{"diffoscope-json-version": 1, "source1": "/srv/reproducible-results/rbuild-debian/r-b-build.eYkYHvZA/b1/opensubdiv_3.5.0-2_amd64.changes", "source2": "/srv/reproducible-results/rbuild-debian/r-b-build.eYkYHvZA/b2/opensubdiv_3.5.0-2_amd64.changes", "unified_diff": null, "details": [{"source1": "Files", "source2": "Files", "unified_diff": "@@ -1,9 +1,9 @@\n \n 30f2dff1d3ed309e7bb0226694b1cad9 143940 libdevel optional libosd-dev_3.5.0-2_amd64.deb\n- 13adbc07e7621a940b17aa19085f9042 31483392 doc optional libosd-doc_3.5.0-2_all.deb\n+ 78cc60dedf9ac5a14f334475ebb198b3 31483352 doc optional libosd-doc_3.5.0-2_all.deb\n 3f2186c3c30098a2ec9c195a114e23d5 3540752 debug optional libosdcpu3.5.0-dbgsym_3.5.0-2_amd64.deb\n 630ded59f081f5c54b5149a1732753d9 330400 libs optional libosdcpu3.5.0_3.5.0-2_amd64.deb\n 2b8f9683537940a21178b0619384ae03 541024 debug optional libosdgpu3.5.0-dbgsym_3.5.0-2_amd64.deb\n f683e27a8366fc93b909b037609c20db 135356 libs optional libosdgpu3.5.0_3.5.0-2_amd64.deb\n c044e60ad4f9165f199cdc4caf0314df 12330168 debug optional opensubdiv-tools-dbgsym_3.5.0-2_amd64.deb\n e2825eb1eeaf55347969a0d14961edf2 638176 graphics optional opensubdiv-tools_3.5.0-2_amd64.deb\n"}, {"source1": "libosd-doc_3.5.0-2_all.deb", "source2": "libosd-doc_3.5.0-2_all.deb", "unified_diff": null, "details": [{"source1": "file list", "source2": "file list", "unified_diff": "@@ -1,3 +1,3 @@\n -rw-r--r-- 0 0 0 4 2022-12-11 09:52:28.000000 debian-binary\n--rw-r--r-- 0 0 0 28052 2022-12-11 09:52:28.000000 control.tar.xz\n--rw-r--r-- 0 0 0 31455148 2022-12-11 09:52:28.000000 data.tar.xz\n+-rw-r--r-- 0 0 0 28004 2022-12-11 09:52:28.000000 control.tar.xz\n+-rw-r--r-- 0 0 0 31455156 2022-12-11 09:52:28.000000 data.tar.xz\n"}, {"source1": "control.tar.xz", "source2": "control.tar.xz", "unified_diff": null, "details": [{"source1": "control.tar", "source2": "control.tar", "unified_diff": null, "details": [{"source1": "./md5sums", "source2": "./md5sums", "unified_diff": null, "details": [{"source1": "./md5sums", "source2": "./md5sums", "comments": ["Files differ"], "unified_diff": null}, {"source1": "line order", "source2": "line order", "unified_diff": "@@ -122,20 +122,20 @@\n usr/share/doc/opensubdiv/doxy_html/a00779.js\n usr/share/doc/opensubdiv/doxy_html/a00779_source.html\n usr/share/doc/opensubdiv/doxy_html/a00782.html\n usr/share/doc/opensubdiv/doxy_html/a00782_source.html\n usr/share/doc/opensubdiv/doxy_html/a00785.html\n usr/share/doc/opensubdiv/doxy_html/a00785_source.html\n usr/share/doc/opensubdiv/doxy_html/a00788.html\n-usr/share/doc/opensubdiv/doxy_html/a00788.js\n usr/share/doc/opensubdiv/doxy_html/a00788_source.html\n usr/share/doc/opensubdiv/doxy_html/a00791.html\n usr/share/doc/opensubdiv/doxy_html/a00791.js\n usr/share/doc/opensubdiv/doxy_html/a00791_source.html\n usr/share/doc/opensubdiv/doxy_html/a00794.html\n+usr/share/doc/opensubdiv/doxy_html/a00794.js\n usr/share/doc/opensubdiv/doxy_html/a00794_source.html\n usr/share/doc/opensubdiv/doxy_html/a00797.html\n usr/share/doc/opensubdiv/doxy_html/a00797.js\n usr/share/doc/opensubdiv/doxy_html/a00797_source.html\n usr/share/doc/opensubdiv/doxy_html/a00800.html\n usr/share/doc/opensubdiv/doxy_html/a00800.js\n usr/share/doc/opensubdiv/doxy_html/a00800_source.html\n@@ -214,17 +214,17 @@\n usr/share/doc/opensubdiv/doxy_html/a00875.html\n usr/share/doc/opensubdiv/doxy_html/a00875.js\n usr/share/doc/opensubdiv/doxy_html/a00875_source.html\n usr/share/doc/opensubdiv/doxy_html/a00878.html\n usr/share/doc/opensubdiv/doxy_html/a00878.js\n usr/share/doc/opensubdiv/doxy_html/a00878_source.html\n usr/share/doc/opensubdiv/doxy_html/a00881.html\n-usr/share/doc/opensubdiv/doxy_html/a00881.js\n usr/share/doc/opensubdiv/doxy_html/a00881_source.html\n usr/share/doc/opensubdiv/doxy_html/a00884.html\n+usr/share/doc/opensubdiv/doxy_html/a00884.js\n usr/share/doc/opensubdiv/doxy_html/a00884_source.html\n usr/share/doc/opensubdiv/doxy_html/a00887.html\n usr/share/doc/opensubdiv/doxy_html/a00887.js\n usr/share/doc/opensubdiv/doxy_html/a00887_source.html\n usr/share/doc/opensubdiv/doxy_html/a00890.html\n usr/share/doc/opensubdiv/doxy_html/a00890.js\n usr/share/doc/opensubdiv/doxy_html/a00890_source.html\n"}]}]}]}, {"source1": "data.tar.xz", "source2": "data.tar.xz", "unified_diff": null, "details": [{"source1": "data.tar", "source2": "data.tar", "unified_diff": null, "details": [{"source1": "file list", "source2": "file list", "unified_diff": "@@ -42,76 +42,76 @@\n -rw-r--r-- 0 root (0) root (0) 26635 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00680_source.html\n -rw-r--r-- 0 root (0) root (0) 6346 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00683.html\n -rw-r--r-- 0 root (0) root (0) 78 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00683.js\n -rw-r--r-- 0 root (0) root (0) 34622 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00683_source.html\n -rw-r--r-- 0 root (0) root (0) 7292 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00686.html\n -rw-r--r-- 0 root (0) root (0) 127 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00686.js\n -rw-r--r-- 0 root (0) root (0) 87199 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00686_source.html\n--rw-r--r-- 0 root (0) root (0) 7539 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00689.html\n--rw-r--r-- 0 root (0) root (0) 150 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00689.js\n--rw-r--r-- 0 root (0) root (0) 39771 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00689_source.html\n--rw-r--r-- 0 root (0) root (0) 6863 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00692.html\n--rw-r--r-- 0 root (0) root (0) 137 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00692.js\n--rw-r--r-- 0 root (0) root (0) 74790 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00692_source.html\n+-rw-r--r-- 0 root (0) root (0) 6863 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00689.html\n+-rw-r--r-- 0 root (0) root (0) 137 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00689.js\n+-rw-r--r-- 0 root (0) root (0) 74790 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00689_source.html\n+-rw-r--r-- 0 root (0) root (0) 7539 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00692.html\n+-rw-r--r-- 0 root (0) root (0) 150 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00692.js\n+-rw-r--r-- 0 root (0) root (0) 39771 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00692_source.html\n -rw-r--r-- 0 root (0) root (0) 6175 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00695.html\n -rw-r--r-- 0 root (0) root (0) 55 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00695.js\n -rw-r--r-- 0 root (0) root (0) 13613 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00695_source.html\n -rw-r--r-- 0 root (0) root (0) 6224 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00698.html\n -rw-r--r-- 0 root (0) root (0) 69 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00698.js\n -rw-r--r-- 0 root (0) root (0) 44798 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00698_source.html\n -rw-r--r-- 0 root (0) root (0) 7625 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00704.html\n -rw-r--r-- 0 root (0) root (0) 222 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00704.js\n -rw-r--r-- 0 root (0) root (0) 143889 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00704_source.html\n -rw-r--r-- 0 root (0) root (0) 7781 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00707.html\n -rw-r--r-- 0 root (0) root (0) 205 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00707.js\n -rw-r--r-- 0 root (0) root (0) 73225 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00707_source.html\n--rw-r--r-- 0 root (0) root (0) 9256 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00710.html\n--rw-r--r-- 0 root (0) root (0) 513 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00710.js\n--rw-r--r-- 0 root (0) root (0) 82688 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00710_source.html\n--rw-r--r-- 0 root (0) root (0) 6924 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00713.html\n--rw-r--r-- 0 root (0) root (0) 139 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00713.js\n--rw-r--r-- 0 root (0) root (0) 37530 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00713_source.html\n--rw-r--r-- 0 root (0) root (0) 6351 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00716.html\n--rw-r--r-- 0 root (0) root (0) 64 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00716.js\n--rw-r--r-- 0 root (0) root (0) 18965 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00716_source.html\n--rw-r--r-- 0 root (0) root (0) 8011 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00719.html\n--rw-r--r-- 0 root (0) root (0) 130 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00719.js\n--rw-r--r-- 0 root (0) root (0) 210952 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00719_source.html\n--rw-r--r-- 0 root (0) root (0) 6486 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00722.html\n--rw-r--r-- 0 root (0) root (0) 66 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00722.js\n--rw-r--r-- 0 root (0) root (0) 65952 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00722_source.html\n--rw-r--r-- 0 root (0) root (0) 9711 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00725.html\n--rw-r--r-- 0 root (0) root (0) 456 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00725.js\n--rw-r--r-- 0 root (0) root (0) 171138 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00725_source.html\n+-rw-r--r-- 0 root (0) root (0) 6924 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00710.html\n+-rw-r--r-- 0 root (0) root (0) 139 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00710.js\n+-rw-r--r-- 0 root (0) root (0) 37530 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00710_source.html\n+-rw-r--r-- 0 root (0) root (0) 9711 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00713.html\n+-rw-r--r-- 0 root (0) root (0) 456 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00713.js\n+-rw-r--r-- 0 root (0) root (0) 171138 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00713_source.html\n+-rw-r--r-- 0 root (0) root (0) 9256 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00716.html\n+-rw-r--r-- 0 root (0) root (0) 513 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00716.js\n+-rw-r--r-- 0 root (0) root (0) 82688 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00716_source.html\n+-rw-r--r-- 0 root (0) root (0) 6486 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00719.html\n+-rw-r--r-- 0 root (0) root (0) 66 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00719.js\n+-rw-r--r-- 0 root (0) root (0) 65952 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00719_source.html\n+-rw-r--r-- 0 root (0) root (0) 6351 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00722.html\n+-rw-r--r-- 0 root (0) root (0) 64 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00722.js\n+-rw-r--r-- 0 root (0) root (0) 18965 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00722_source.html\n+-rw-r--r-- 0 root (0) root (0) 8011 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00725.html\n+-rw-r--r-- 0 root (0) root (0) 130 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00725.js\n+-rw-r--r-- 0 root (0) root (0) 210952 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00725_source.html\n -rw-r--r-- 0 root (0) root (0) 7601 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00728.html\n -rw-r--r-- 0 root (0) root (0) 201 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00728.js\n -rw-r--r-- 0 root (0) root (0) 66897 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00728_source.html\n -rw-r--r-- 0 root (0) root (0) 7124 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00731.html\n -rw-r--r-- 0 root (0) root (0) 123 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00731.js\n -rw-r--r-- 0 root (0) root (0) 156950 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00731_source.html\n -rw-r--r-- 0 root (0) root (0) 6533 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00734.html\n -rw-r--r-- 0 root (0) root (0) 123 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00734.js\n -rw-r--r-- 0 root (0) root (0) 41643 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00734_source.html\n--rw-r--r-- 0 root (0) root (0) 7604 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00737.html\n--rw-r--r-- 0 root (0) root (0) 314 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00737.js\n--rw-r--r-- 0 root (0) root (0) 49522 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00737_source.html\n--rw-r--r-- 0 root (0) root (0) 6977 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00740.html\n--rw-r--r-- 0 root (0) root (0) 160 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00740.js\n--rw-r--r-- 0 root (0) root (0) 52341 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00740_source.html\n+-rw-r--r-- 0 root (0) root (0) 6977 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00737.html\n+-rw-r--r-- 0 root (0) root (0) 160 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00737.js\n+-rw-r--r-- 0 root (0) root (0) 52341 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00737_source.html\n+-rw-r--r-- 0 root (0) root (0) 7604 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00740.html\n+-rw-r--r-- 0 root (0) root (0) 314 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00740.js\n+-rw-r--r-- 0 root (0) root (0) 49522 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00740_source.html\n -rw-r--r-- 0 root (0) root (0) 11046 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00743.html\n -rw-r--r-- 0 root (0) root (0) 1173 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00743.js\n -rw-r--r-- 0 root (0) root (0) 22461 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00743_source.html\n -rw-r--r-- 0 root (0) root (0) 5432 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00746.html\n -rw-r--r-- 0 root (0) root (0) 37914 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00746_source.html\n -rw-r--r-- 0 root (0) root (0) 5687 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00749.html\n -rw-r--r-- 0 root (0) root (0) 21734 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00749_source.html\n--rw-r--r-- 0 root (0) root (0) 5990 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00752.html\n--rw-r--r-- 0 root (0) root (0) 163100 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00752_source.html\n--rw-r--r-- 0 root (0) root (0) 5870 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00755.html\n--rw-r--r-- 0 root (0) root (0) 85578 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00755_source.html\n+-rw-r--r-- 0 root (0) root (0) 5870 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00752.html\n+-rw-r--r-- 0 root (0) root (0) 85578 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00752_source.html\n+-rw-r--r-- 0 root (0) root (0) 5990 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00755.html\n+-rw-r--r-- 0 root (0) root (0) 163100 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00755_source.html\n -rw-r--r-- 0 root (0) root (0) 6077 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00758.html\n -rw-r--r-- 0 root (0) root (0) 23929 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00758_source.html\n -rw-r--r-- 0 root (0) root (0) 6055 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00761.html\n -rw-r--r-- 0 root (0) root (0) 81111 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00761_source.html\n -rw-r--r-- 0 root (0) root (0) 6225 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00764.html\n -rw-r--r-- 0 root (0) root (0) 121 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00764.js\n -rw-r--r-- 0 root (0) root (0) 45393 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00764_source.html\n@@ -124,130 +124,130 @@\n -rw-r--r-- 0 root (0) root (0) 60 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00773.js\n -rw-r--r-- 0 root (0) root (0) 37854 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00773_source.html\n -rw-r--r-- 0 root (0) root (0) 5615 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00776.html\n -rw-r--r-- 0 root (0) root (0) 99219 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00776_source.html\n -rw-r--r-- 0 root (0) root (0) 6262 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00779.html\n -rw-r--r-- 0 root (0) root (0) 59 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00779.js\n -rw-r--r-- 0 root (0) root (0) 52972 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00779_source.html\n--rw-r--r-- 0 root (0) root (0) 5541 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00782.html\n--rw-r--r-- 0 root (0) root (0) 37229 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00782_source.html\n--rw-r--r-- 0 root (0) root (0) 5624 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00785.html\n--rw-r--r-- 0 root (0) root (0) 88720 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00785_source.html\n--rw-r--r-- 0 root (0) root (0) 6581 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00788.html\n--rw-r--r-- 0 root (0) root (0) 67 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00788.js\n--rw-r--r-- 0 root (0) root (0) 33086 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00788_source.html\n--rw-r--r-- 0 root (0) root (0) 6899 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00791.html\n--rw-r--r-- 0 root (0) root (0) 115 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00791.js\n--rw-r--r-- 0 root (0) root (0) 127471 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00791_source.html\n--rw-r--r-- 0 root (0) root (0) 4308 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00794.html\n--rw-r--r-- 0 root (0) root (0) 10202 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00794_source.html\n+-rw-r--r-- 0 root (0) root (0) 5624 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00782.html\n+-rw-r--r-- 0 root (0) root (0) 88720 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00782_source.html\n+-rw-r--r-- 0 root (0) root (0) 5541 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00785.html\n+-rw-r--r-- 0 root (0) root (0) 37229 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00785_source.html\n+-rw-r--r-- 0 root (0) root (0) 4308 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00788.html\n+-rw-r--r-- 0 root (0) root (0) 10202 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00788_source.html\n+-rw-r--r-- 0 root (0) root (0) 6581 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00791.html\n+-rw-r--r-- 0 root (0) root (0) 67 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00791.js\n+-rw-r--r-- 0 root (0) root (0) 33086 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00791_source.html\n+-rw-r--r-- 0 root (0) root (0) 6899 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00794.html\n+-rw-r--r-- 0 root (0) root (0) 115 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00794.js\n+-rw-r--r-- 0 root (0) root (0) 127471 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00794_source.html\n -rw-r--r-- 0 root (0) root (0) 6213 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00797.html\n -rw-r--r-- 0 root (0) root (0) 67 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00797.js\n -rw-r--r-- 0 root (0) root (0) 23232 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00797_source.html\n--rw-r--r-- 0 root (0) root (0) 6948 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00800.html\n--rw-r--r-- 0 root (0) root (0) 115 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00800.js\n--rw-r--r-- 0 root (0) root (0) 249312 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00800_source.html\n--rw-r--r-- 0 root (0) root (0) 6671 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00803.html\n--rw-r--r-- 0 root (0) root (0) 65 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00803.js\n--rw-r--r-- 0 root (0) root (0) 33708 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00803_source.html\n+-rw-r--r-- 0 root (0) root (0) 6671 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00800.html\n+-rw-r--r-- 0 root (0) root (0) 65 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00800.js\n+-rw-r--r-- 0 root (0) root (0) 33708 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00800_source.html\n+-rw-r--r-- 0 root (0) root (0) 6948 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00803.html\n+-rw-r--r-- 0 root (0) root (0) 115 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00803.js\n+-rw-r--r-- 0 root (0) root (0) 249312 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00803_source.html\n -rw-r--r-- 0 root (0) root (0) 5949 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00806.html\n -rw-r--r-- 0 root (0) root (0) 63 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00806.js\n -rw-r--r-- 0 root (0) root (0) 13628 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00806_source.html\n -rw-r--r-- 0 root (0) root (0) 6097 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00809.html\n -rw-r--r-- 0 root (0) root (0) 69 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00809.js\n -rw-r--r-- 0 root (0) root (0) 19929 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00809_source.html\n--rw-r--r-- 0 root (0) root (0) 6253 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00812.html\n--rw-r--r-- 0 root (0) root (0) 99 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00812.js\n--rw-r--r-- 0 root (0) root (0) 12331 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00812_source.html\n--rw-r--r-- 0 root (0) root (0) 6592 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00815.html\n--rw-r--r-- 0 root (0) root (0) 66 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00815.js\n--rw-r--r-- 0 root (0) root (0) 34882 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00815_source.html\n--rw-r--r-- 0 root (0) root (0) 6240 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00818.html\n--rw-r--r-- 0 root (0) root (0) 79 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00818.js\n--rw-r--r-- 0 root (0) root (0) 24609 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00818_source.html\n--rw-r--r-- 0 root (0) root (0) 6034 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00821.html\n--rw-r--r-- 0 root (0) root (0) 71 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00821.js\n--rw-r--r-- 0 root (0) root (0) 21842 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00821_source.html\n--rw-r--r-- 0 root (0) root (0) 6826 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00824.html\n--rw-r--r-- 0 root (0) root (0) 124 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00824.js\n--rw-r--r-- 0 root (0) root (0) 221456 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00824_source.html\n--rw-r--r-- 0 root (0) root (0) 6265 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00827.html\n--rw-r--r-- 0 root (0) root (0) 101 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00827.js\n--rw-r--r-- 0 root (0) root (0) 12448 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00827_source.html\n--rw-r--r-- 0 root (0) root (0) 6146 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00830.html\n--rw-r--r-- 0 root (0) root (0) 70 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00830.js\n--rw-r--r-- 0 root (0) root (0) 22400 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00830_source.html\n--rw-r--r-- 0 root (0) root (0) 6104 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00833.html\n+-rw-r--r-- 0 root (0) root (0) 6592 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00812.html\n+-rw-r--r-- 0 root (0) root (0) 66 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00812.js\n+-rw-r--r-- 0 root (0) root (0) 34882 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00812_source.html\n+-rw-r--r-- 0 root (0) root (0) 6240 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00815.html\n+-rw-r--r-- 0 root (0) root (0) 79 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00815.js\n+-rw-r--r-- 0 root (0) root (0) 24609 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00815_source.html\n+-rw-r--r-- 0 root (0) root (0) 6253 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00818.html\n+-rw-r--r-- 0 root (0) root (0) 99 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00818.js\n+-rw-r--r-- 0 root (0) root (0) 12331 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00818_source.html\n+-rw-r--r-- 0 root (0) root (0) 6826 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00821.html\n+-rw-r--r-- 0 root (0) root (0) 124 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00821.js\n+-rw-r--r-- 0 root (0) root (0) 221456 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00821_source.html\n+-rw-r--r-- 0 root (0) root (0) 6034 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00824.html\n+-rw-r--r-- 0 root (0) root (0) 71 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00824.js\n+-rw-r--r-- 0 root (0) root (0) 21842 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00824_source.html\n+-rw-r--r-- 0 root (0) root (0) 6104 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00827.html\n+-rw-r--r-- 0 root (0) root (0) 70 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00827.js\n+-rw-r--r-- 0 root (0) root (0) 16387 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00827_source.html\n+-rw-r--r-- 0 root (0) root (0) 6265 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00830.html\n+-rw-r--r-- 0 root (0) root (0) 101 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00830.js\n+-rw-r--r-- 0 root (0) root (0) 12448 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00830_source.html\n+-rw-r--r-- 0 root (0) root (0) 6146 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00833.html\n -rw-r--r-- 0 root (0) root (0) 70 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00833.js\n--rw-r--r-- 0 root (0) root (0) 16387 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00833_source.html\n--rw-r--r-- 0 root (0) root (0) 6145 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00836.html\n--rw-r--r-- 0 root (0) root (0) 81 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00836.js\n--rw-r--r-- 0 root (0) root (0) 26150 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00836_source.html\n--rw-r--r-- 0 root (0) root (0) 6791 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00839.html\n--rw-r--r-- 0 root (0) root (0) 128 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00839.js\n--rw-r--r-- 0 root (0) root (0) 49005 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00839_source.html\n--rw-r--r-- 0 root (0) root (0) 6198 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00842.html\n--rw-r--r-- 0 root (0) root (0) 73 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00842.js\n--rw-r--r-- 0 root (0) root (0) 22190 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00842_source.html\n--rw-r--r-- 0 root (0) root (0) 6540 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00845.html\n--rw-r--r-- 0 root (0) root (0) 68 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00845.js\n--rw-r--r-- 0 root (0) root (0) 27167 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00845_source.html\n--rw-r--r-- 0 root (0) root (0) 6999 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00848.html\n--rw-r--r-- 0 root (0) root (0) 126 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00848.js\n--rw-r--r-- 0 root (0) root (0) 221754 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00848_source.html\n--rw-r--r-- 0 root (0) root (0) 6973 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00851.html\n--rw-r--r-- 0 root (0) root (0) 121 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00851.js\n--rw-r--r-- 0 root (0) root (0) 236475 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00851_source.html\n--rw-r--r-- 0 root (0) root (0) 6247 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00854.html\n--rw-r--r-- 0 root (0) root (0) 98 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00854.js\n--rw-r--r-- 0 root (0) root (0) 12412 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00854_source.html\n--rw-r--r-- 0 root (0) root (0) 6274 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00857.html\n--rw-r--r-- 0 root (0) root (0) 67 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00857.js\n--rw-r--r-- 0 root (0) root (0) 19505 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00857_source.html\n--rw-r--r-- 0 root (0) root (0) 6104 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00860.html\n--rw-r--r-- 0 root (0) root (0) 70 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00860.js\n--rw-r--r-- 0 root (0) root (0) 16709 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00860_source.html\n--rw-r--r-- 0 root (0) root (0) 6519 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00863.html\n--rw-r--r-- 0 root (0) root (0) 65 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00863.js\n--rw-r--r-- 0 root (0) root (0) 45624 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00863_source.html\n--rw-r--r-- 0 root (0) root (0) 6233 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00866.html\n--rw-r--r-- 0 root (0) root (0) 78 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00866.js\n--rw-r--r-- 0 root (0) root (0) 22701 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00866_source.html\n--rw-r--r-- 0 root (0) root (0) 9743 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00869.html\n--rw-r--r-- 0 root (0) root (0) 424 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00869.js\n--rw-r--r-- 0 root (0) root (0) 28840 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00869_source.html\n+-rw-r--r-- 0 root (0) root (0) 22400 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00833_source.html\n+-rw-r--r-- 0 root (0) root (0) 6540 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00836.html\n+-rw-r--r-- 0 root (0) root (0) 68 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00836.js\n+-rw-r--r-- 0 root (0) root (0) 27167 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00836_source.html\n+-rw-r--r-- 0 root (0) root (0) 6145 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00839.html\n+-rw-r--r-- 0 root (0) root (0) 81 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00839.js\n+-rw-r--r-- 0 root (0) root (0) 26150 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00839_source.html\n+-rw-r--r-- 0 root (0) root (0) 6791 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00842.html\n+-rw-r--r-- 0 root (0) root (0) 128 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00842.js\n+-rw-r--r-- 0 root (0) root (0) 49005 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00842_source.html\n+-rw-r--r-- 0 root (0) root (0) 6973 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00845.html\n+-rw-r--r-- 0 root (0) root (0) 121 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00845.js\n+-rw-r--r-- 0 root (0) root (0) 236475 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00845_source.html\n+-rw-r--r-- 0 root (0) root (0) 6198 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00848.html\n+-rw-r--r-- 0 root (0) root (0) 73 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00848.js\n+-rw-r--r-- 0 root (0) root (0) 22190 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00848_source.html\n+-rw-r--r-- 0 root (0) root (0) 6999 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00851.html\n+-rw-r--r-- 0 root (0) root (0) 126 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00851.js\n+-rw-r--r-- 0 root (0) root (0) 221754 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00851_source.html\n+-rw-r--r-- 0 root (0) root (0) 6104 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00854.html\n+-rw-r--r-- 0 root (0) root (0) 70 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00854.js\n+-rw-r--r-- 0 root (0) root (0) 16709 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00854_source.html\n+-rw-r--r-- 0 root (0) root (0) 6247 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00857.html\n+-rw-r--r-- 0 root (0) root (0) 98 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00857.js\n+-rw-r--r-- 0 root (0) root (0) 12412 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00857_source.html\n+-rw-r--r-- 0 root (0) root (0) 6519 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00860.html\n+-rw-r--r-- 0 root (0) root (0) 65 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00860.js\n+-rw-r--r-- 0 root (0) root (0) 45624 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00860_source.html\n+-rw-r--r-- 0 root (0) root (0) 6233 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00863.html\n+-rw-r--r-- 0 root (0) root (0) 78 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00863.js\n+-rw-r--r-- 0 root (0) root (0) 22701 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00863_source.html\n+-rw-r--r-- 0 root (0) root (0) 6274 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00866.html\n+-rw-r--r-- 0 root (0) root (0) 67 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00866.js\n+-rw-r--r-- 0 root (0) root (0) 19505 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00866_source.html\n+-rw-r--r-- 0 root (0) root (0) 6169 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00869.html\n+-rw-r--r-- 0 root (0) root (0) 61 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00869.js\n+-rw-r--r-- 0 root (0) root (0) 109731 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00869_source.html\n -rw-r--r-- 0 root (0) root (0) 6285 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00872.html\n -rw-r--r-- 0 root (0) root (0) 70 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00872.js\n -rw-r--r-- 0 root (0) root (0) 20583 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00872_source.html\n--rw-r--r-- 0 root (0) root (0) 6169 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00875.html\n--rw-r--r-- 0 root (0) root (0) 61 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00875.js\n--rw-r--r-- 0 root (0) root (0) 109731 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00875_source.html\n+-rw-r--r-- 0 root (0) root (0) 9743 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00875.html\n+-rw-r--r-- 0 root (0) root (0) 424 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00875.js\n+-rw-r--r-- 0 root (0) root (0) 28840 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00875_source.html\n -rw-r--r-- 0 root (0) root (0) 6169 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00878.html\n -rw-r--r-- 0 root (0) root (0) 61 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00878.js\n -rw-r--r-- 0 root (0) root (0) 109537 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00878_source.html\n--rw-r--r-- 0 root (0) root (0) 7905 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00881.html\n--rw-r--r-- 0 root (0) root (0) 262 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00881.js\n--rw-r--r-- 0 root (0) root (0) 20880 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00881_source.html\n--rw-r--r-- 0 root (0) root (0) 4304 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00884.html\n--rw-r--r-- 0 root (0) root (0) 12502 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00884_source.html\n+-rw-r--r-- 0 root (0) root (0) 4304 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00881.html\n+-rw-r--r-- 0 root (0) root (0) 12502 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00881_source.html\n+-rw-r--r-- 0 root (0) root (0) 7905 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00884.html\n+-rw-r--r-- 0 root (0) root (0) 262 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00884.js\n+-rw-r--r-- 0 root (0) root (0) 20880 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00884_source.html\n -rw-r--r-- 0 root (0) root (0) 13417 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00887.html\n -rw-r--r-- 0 root (0) root (0) 2496 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00887.js\n -rw-r--r-- 0 root (0) root (0) 177843 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00887_source.html\n--rw-r--r-- 0 root (0) root (0) 6685 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00890.html\n--rw-r--r-- 0 root (0) root (0) 66 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00890.js\n--rw-r--r-- 0 root (0) root (0) 42717 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00890_source.html\n--rw-r--r-- 0 root (0) root (0) 6158 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00893.html\n--rw-r--r-- 0 root (0) root (0) 68 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00893.js\n--rw-r--r-- 0 root (0) root (0) 18797 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00893_source.html\n--rw-r--r-- 0 root (0) root (0) 6475 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00896.html\n--rw-r--r-- 0 root (0) root (0) 69 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00896.js\n--rw-r--r-- 0 root (0) root (0) 26583 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00896_source.html\n--rw-r--r-- 0 root (0) root (0) 6169 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00899.html\n--rw-r--r-- 0 root (0) root (0) 61 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00899.js\n--rw-r--r-- 0 root (0) root (0) 109206 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00899_source.html\n+-rw-r--r-- 0 root (0) root (0) 6158 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00890.html\n+-rw-r--r-- 0 root (0) root (0) 68 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00890.js\n+-rw-r--r-- 0 root (0) root (0) 18797 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00890_source.html\n+-rw-r--r-- 0 root (0) root (0) 6685 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00893.html\n+-rw-r--r-- 0 root (0) root (0) 66 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00893.js\n+-rw-r--r-- 0 root (0) root (0) 42717 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00893_source.html\n+-rw-r--r-- 0 root (0) root (0) 6169 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00896.html\n+-rw-r--r-- 0 root (0) root (0) 61 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00896.js\n+-rw-r--r-- 0 root (0) root (0) 109206 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00896_source.html\n+-rw-r--r-- 0 root (0) root (0) 6475 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00899.html\n+-rw-r--r-- 0 root (0) root (0) 69 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00899.js\n+-rw-r--r-- 0 root (0) root (0) 26583 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00899_source.html\n -rw-r--r-- 0 root (0) root (0) 6163 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00902.html\n -rw-r--r-- 0 root (0) root (0) 69 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00902.js\n -rw-r--r-- 0 root (0) root (0) 19559 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00902_source.html\n -rw-r--r-- 0 root (0) root (0) 4475 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00905.html\n -rw-r--r-- 0 root (0) root (0) 71 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00905.js\n -rw-r--r-- 0 root (0) root (0) 5652 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00906.html\n -rw-r--r-- 0 root (0) root (0) 216 2022-12-11 09:52:28.000000 ./usr/share/doc/opensubdiv/doxy_html/a00906.js\n"}, {"source1": "./usr/share/doc/opensubdiv/additional_resources.html", "source2": "./usr/share/doc/opensubdiv/additional_resources.html", "unified_diff": "@@ -234,15 +234,15 @@\n surfaces helps artists be more productive, and how this code is open\n source and engineered for ease of integration.

\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -101,14 +101,14 @@\n meet_the_experts_autodesk2013.png]\n Join special guests, Bill Polson, Dirk Van Gelder, Manuel Kraemer, Takahito\n Tejima, David G. Yu and Dale Ruffolo, from Pixar Animation Studios' GPU team,\n as they show how real time display of subdivision surfaces helps artists be\n more productive, and how this code is open source and engineered for ease of\n integration.\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/api_overview.html", "source2": "./usr/share/doc/opensubdiv/api_overview.html", "unified_diff": "@@ -455,15 +455,15 @@\n
\n

Tutorials and Examples

\n

For more use cases, please see Tutorials and Examples

\n
\n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -247,14 +247,14 @@\n glDrawElements(GL_PATCHES, patch.GetNumPatches() * numVertsPerPatch,\n GL_UNSIGNED_INT, 0);\n 1. As the mesh animates, repeat from step 3 to update positions, refine, and\n draw. See glViewer and other examples for more complete usage.\n **** Tutorials and Examples ****\n For more use cases, please see Tutorials and Examples\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/bfr_overview.html", "source2": "./usr/share/doc/opensubdiv/bfr_overview.html", "unified_diff": "@@ -1260,15 +1260,15 @@\n worth simplifying the template complexity here if only one precision is\n ever required.

\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -808,14 +808,14 @@\n MeshPrimvar const & meshPrimvar,\n Surface * surface);\n which would then determine the nature of the MeshPrimvar data (interpolated as\n vertex, varying or face-varying) and act accordingly. It may also be worth\n simplifying the template complexity here if only one precision is ever\n required.\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/bfr_tutorial_1_1.html", "source2": "./usr/share/doc/opensubdiv/bfr_tutorial_1_1.html", "unified_diff": "@@ -426,15 +426,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -307,14 +307,14 @@\n delete meshTopology;\n return EXIT_SUCCESS;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/bfr_tutorial_1_2.html", "source2": "./usr/share/doc/opensubdiv/bfr_tutorial_1_2.html", "unified_diff": "@@ -436,15 +436,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -317,14 +317,14 @@\n delete meshTopology;\n return EXIT_SUCCESS;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/bfr_tutorial_1_3.html", "source2": "./usr/share/doc/opensubdiv/bfr_tutorial_1_3.html", "unified_diff": "@@ -507,15 +507,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -389,14 +389,14 @@\n delete meshTopology;\n return EXIT_SUCCESS;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/bfr_tutorial_1_4.html", "source2": "./usr/share/doc/opensubdiv/bfr_tutorial_1_4.html", "unified_diff": "@@ -559,15 +559,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -440,14 +440,14 @@\n delete meshTopology;\n return EXIT_SUCCESS;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/bfr_tutorial_1_5.html", "source2": "./usr/share/doc/opensubdiv/bfr_tutorial_1_5.html", "unified_diff": "@@ -468,15 +468,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -349,14 +349,14 @@\n delete meshTopology;\n return EXIT_SUCCESS;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/bfr_tutorial_2_1.html", "source2": "./usr/share/doc/opensubdiv/bfr_tutorial_2_1.html", "unified_diff": "@@ -558,15 +558,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -441,14 +441,14 @@\n delete meshTopology;\n return EXIT_SUCCESS;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/bfr_tutorial_2_2.html", "source2": "./usr/share/doc/opensubdiv/bfr_tutorial_2_2.html", "unified_diff": "@@ -641,15 +641,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -522,14 +522,14 @@\n delete meshTopology;\n return EXIT_SUCCESS;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/bfr_tutorial_3_1.html", "source2": "./usr/share/doc/opensubdiv/bfr_tutorial_3_1.html", "unified_diff": "@@ -509,15 +509,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -391,14 +391,14 @@\n delete meshTopology;\n return EXIT_SUCCESS;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/bfr_tutorial_3_2.html", "source2": "./usr/share/doc/opensubdiv/bfr_tutorial_3_2.html", "unified_diff": "@@ -565,15 +565,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -447,14 +447,14 @@\n delete meshTopology;\n return EXIT_SUCCESS;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/cmake_build.html", "source2": "./usr/share/doc/opensubdiv/cmake_build.html", "unified_diff": "@@ -500,15 +500,15 @@\n warning in both gcc and Clang. It is recommended to turn the warning off with the\n -Wno-invalid-offsetof flag.

\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -282,14 +282,14 @@\n link /nologo /out:myapp.exe /LIBPATH:\"%OPENSUBDIV%\\lib\" libosdGPU.lib\n libosdCPU.lib myapp.obj\n Note:\n HBR uses the offsetof macro on a templated struct, which appears to spurriously\n set off a warning in both gcc and Clang. It is recommended to turn the warning\n off with the -Wno-invalid-offsetof flag.\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/code_examples.html", "source2": "./usr/share/doc/opensubdiv/code_examples.html", "unified_diff": "@@ -290,15 +290,15 @@\n w : switch display mode\n q : quit\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -114,14 +114,14 @@\n level\n +, - : increase / decrease tessellation\n Tab : toggle full-screen\n Esc : turn on / off the HUD\n w : switch display mode\n q : quit\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/compatibility.html", "source2": "./usr/share/doc/opensubdiv/compatibility.html", "unified_diff": "@@ -569,15 +569,15 @@\n not go undetected. Whether or not this can lead to visual artifacts is\n unclear.

\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -363,14 +363,14 @@\n coefficient first in the Hbr implementation. With extremely high valence\n vertices, a difference in magnitude between the most and least significant\n coefficients of several orders of magnitude is likely, and that has a\n significant impact on the single-precision floating point computations.\n The improved accuracy of OpenSubdiv 3.0 can reach a magnitude that will not go\n undetected. Whether or not this can lead to visual artifacts is unclear.\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/contributing.html", "source2": "./usr/share/doc/opensubdiv/contributing.html", "unified_diff": "@@ -304,15 +304,15 @@\n

Once the change is clean, push your changes to "origin" and go to the GitHub\n website to submit your pull request.

\n

Be sure to submit your request against the "dev" branch.

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -141,14 +141,14 @@\n If you've followed the steps above, your pending change should already be\n queued up as required. If you have not, you may need to rebase and squash\n changes at this point.\n Once the change is clean, push your changes to \"origin\" and go to the GitHub\n website to submit your pull request.\n Be sure to submit your request against the \"dev\" branch.\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/downloads.html", "source2": "./usr/share/doc/opensubdiv/downloads.html", "unified_diff": "@@ -299,15 +299,15 @@\n

\n \n

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -103,14 +103,14 @@\n Unless required by applicable law or agreed to in writing, software\n distributed under the Apache License with the above modification is\n distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n KIND, either express or implied. See the Apache License for the specific\n language governing permissions and limitations under the Apache License.\n Accept\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00686.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00686.html", "unified_diff": "@@ -87,15 +87,15 @@\n
\n Classes |\n Namespaces
\n
surfaceFactory.h File Reference
\n \n
\n
#include "../version.h"
\n-#include "../bfr/surface.h"
\n+#include "../bfr/surface.h"
\n #include "../bfr/surfaceFactoryMeshAdapter.h"
\n #include "../sdc/options.h"
\n #include "../sdc/types.h"
\n #include <cstdint>
\n
\n

Go to the source code of this file.

\n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00686_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00686_source.html", "unified_diff": "@@ -432,15 +432,15 @@\n
560
\n
561} // end namespace OpenSubdiv
\n
562
\n
563#endif /* OPENSUBDIV3_BFR_SURFACE_FACTORY_H */
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SchemeType
SchemeType
Enumerated type for all subdivision schemes supported by OpenSubdiv.
Definition: types.h:37
\n
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Parameterization
Simple class defining the 2D parameterization of a face.
Definition: parameterization.h:52
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface
Encapsulates the limit surface for a face of a mesh.
Definition: surface.h:59
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface
Encapsulates the limit surface for a face of a mesh.
Definition: surface.h:59
\n
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory
Base class providing initialization of a Surface for each face of a mesh.
Definition: surfaceFactory.h:103
\n
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::setInternalCache
void setInternalCache(SurfaceFactoryCache *cache)
Subclass to identify an internal cache for use by base class.
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::InitFaceVaryingSurface
bool InitFaceVaryingSurface(Index faceIndex, Surface< REAL > *surface) const
Initialize a Surface for the default face-varying data.
Definition: surfaceFactory.h:500
\n
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::GetSchemeOptions
Sdc::Options GetSchemeOptions() const
Return the set of subdivision options.
Definition: surfaceFactory.h:173
\n
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::GetFaceParameterization
Parameterization GetFaceParameterization(Index faceIndex) const
Return the Parameterization of a face with a limit surface.
\n
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::SurfaceFactory
SurfaceFactory(SurfaceFactory const &)=delete
\n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00689.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00689.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/bfr/refinerSurfaceFactory.h File Reference\n+OpenSubdiv: opensubdiv/bfr/surface.h File Reference\n \n \n \n \n \n \n \n@@ -83,47 +83,46 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
refinerSurfaceFactory.h File Reference
\n+
surface.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../bfr/surfaceFactory.h"
\n-#include "../bfr/surfaceFactoryCache.h"
\n+#include "../bfr/surfaceData.h"
\n+#include "../bfr/parameterization.h"
\n+#include "../vtr/array.h"
\n
\n

Go to the source code of this file.

\n
\n \n-\n-\n+\n+\n \n-\n-\n+\n+\n \n

\n Classes

class  RefinerSurfaceFactoryBase
 Intermediate subclass of SurfaceFactory with Far::TopologyRefiner as the mesh. More...
class  Surface< REAL >
 Encapsulates the limit surface for a face of a mesh. More...
 
class  RefinerSurfaceFactory< CACHE_TYPE >
 Template for concrete subclasses of RefinerSurfaceFactoryBase. More...
struct  Surface< REAL >::PointDescriptor
 Simple struct defining the size and stride of points in arrays. More...
 
\n \n \n \n \n \n-\n-\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Bfr
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,33 +5,32 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-refinerSurfaceFactory.h File Reference\n+surface.h File Reference\n #include \"../version.h\"\n-#include \"../bfr/surfaceFactory.h\"\n-#include \"../bfr/surfaceFactoryCache.h\"\n+#include \"../bfr/surfaceData.h\"\n+#include \"../bfr/parameterization.h\"\n+#include \"../vtr/array.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0RefinerSurfaceFactoryBase\n-\u00a0 Intermediate subclass of SurfaceFactory with Far::TopologyRefiner as\n- the mesh. More...\n+ class \u00a0Surface<_REAL_>\n+\u00a0 Encapsulates the limit surface for a face of a mesh. More...\n \u00a0\n-class \u00a0RefinerSurfaceFactory<_CACHE_TYPE_>\n-\u00a0 Template for concrete subclasses of RefinerSurfaceFactoryBase. More...\n+struct \u00a0Surface<_REAL_>::PointDescriptor\n+\u00a0 Simple struct defining the size and stride of points in arrays.\n+ More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n-\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Bfr\n \u00a0\n * opensubdiv\n * bfr\n- * refinerSurfaceFactory.h\n+ * surface.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00689.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00689.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,4 @@\n var a00689 = [\n- [\"RefinerSurfaceFactoryBase\", \"a00925.html\", \"a00925\"],\n- [\"RefinerSurfaceFactory< CACHE_TYPE >\", \"a00929.html\", \"a00929\"]\n+ [\"Surface< REAL >\", \"a00933.html\", \"a00933\"],\n+ [\"Surface< REAL >::PointDescriptor\", \"a00937.html\", \"a00937\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00689_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00689_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/bfr/refinerSurfaceFactory.h Source File\n+OpenSubdiv: opensubdiv/bfr/surface.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
refinerSurfaceFactory.h
\n+
surface.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2021 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,155 +107,338 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H
\n-
26#define OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H
\n+
25#ifndef OPENSUBDIV3_BFR_SURFACE_H
\n+
26#define OPENSUBDIV3_BFR_SURFACE_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30#include "../bfr/surfaceFactory.h"
\n-
31#include "../bfr/surfaceFactoryCache.h"
\n-
32
\n-
33namespace OpenSubdiv {
\n-
34namespace OPENSUBDIV_VERSION {
\n-
35
\n-
36namespace Far {
\n-
37 class TopologyRefiner;
\n-
38}
\n-
39
\n-
40namespace Bfr {
\n-
41
\n-
58class RefinerSurfaceFactoryBase : public SurfaceFactory {
\n-
59public:
\n-
61
\n-
65
\n-
66 RefinerSurfaceFactoryBase(Far::TopologyRefiner const & mesh,
\n-
67 Options const & options);
\n-
68
\n-
69 ~RefinerSurfaceFactoryBase() override = default;
\n-
71
\n+
30#include "../bfr/surfaceData.h"
\n+
31#include "../bfr/parameterization.h"
\n+
32#include "../vtr/array.h"
\n+
33
\n+
34namespace OpenSubdiv {
\n+
35namespace OPENSUBDIV_VERSION {
\n+
36
\n+
37namespace Bfr {
\n+
38
\n+
58template <typename REAL>
\n+
59class Surface {
\n+
60public:
\n+
63 struct PointDescriptor {
\n+
64 PointDescriptor() : size(0), stride(0) { }
\n+
65 PointDescriptor(int n) : size(n), stride(n) { }
\n+
66 PointDescriptor(int n, int m) : size(n), stride(m) { }
\n+
67
\n+
68 int size, stride;
\n+
69 };
\n+
70
\n+
72 typedef int Index;
\n
73
\n-
77
\n-
79 Far::TopologyRefiner const & GetMesh() const { return _mesh; }
\n-
80
\n-
82 int GetNumFaces() const { return _numFaces; }
\n-
83
\n-
85 int GetNumFVarChannels() const { return _numFVarChannels; }
\n+
74public:
\n+
76
\n+
84
\n+
86 bool IsValid() const { return _data.isValid(); }
\n
87
\n-
88protected:
\n-
90 //
\n-
91 // Virtual overrides to satisfy the SurfaceFactoryMeshAdapter interface:
\n-
92 //
\n-
93 bool isFaceHole( Index faceIndex) const override;
\n-
94 int getFaceSize(Index faceIndex) const override;
\n-
95
\n-
96 int getFaceVertexIndices(Index faceIndex,
\n-
97 Index vertexIndices[]) const override;
\n-
98 int getFaceFVarValueIndices(Index faceIndex,
\n-
99 FVarID fvarID, Index fvarValueIndices[]) const override;
\n+
89 void Clear() { _data.reinitialize(); }
\n+
90
\n+
92 Surface();
\n+
93
\n+
94 Surface(Surface const & src) = default;
\n+
95 Surface& operator=(Surface const & src) = default;
\n+
96 ~Surface() = default;
\n+
98
\n
100
\n-
101 int populateFaceVertexDescriptor(Index faceIndex, int faceVertex,
\n-
102 VertexDescriptor * vertexDescriptor) const override;
\n-
103
\n-
104 int getFaceVertexIncidentFaceVertexIndices(
\n-
105 Index faceIndex, int faceVertex,
\n-
106 Index vertexIndices[]) const override;
\n-
107 int getFaceVertexIncidentFaceFVarValueIndices(
\n-
108 Index faceIndex, int faceVertex,
\n-
109 FVarID fvarID, Index fvarValueIndices[]) const override;
\n+
104
\n+
106 Parameterization GetParameterization() const { return _data.getParam(); }
\n+
107
\n+
109 int GetFaceSize() const { return GetParameterization().GetFaceSize(); }
\n
110
\n-
111 // Optional SurfaceFactoryMeshAdapter overrides for regular patches:
\n-
112 bool getFaceNeighborhoodVertexIndicesIfRegular(
\n-
113 Index faceIndex,
\n-
114 Index vertexIndices[]) const override;
\n-
115
\n-
116 bool getFaceNeighborhoodFVarValueIndicesIfRegular(
\n-
117 Index faceIndex,
\n-
118 FVarID fvarID, Index fvarValueIndices[]) const override;
\n-
120
\n-
121private:
\n-
122 //
\n-
123 // Internal supporting methods:
\n-
124 //
\n-
125 int getFaceVaryingChannel(FVarID fvarID) const;
\n-
126
\n-
127 int getFaceVertexPointIndices(Index faceIndex, int faceVertex,
\n-
128 Index indices[], int vtxOrFVarChannel) const;
\n-
129
\n-
130 int getFacePatchPointIndices(Index faceIndex,
\n-
131 Index indices[], int vtxOrFVarChannel) const;
\n-
132
\n-
133private:
\n-
134 // Additional members for the subclass:
\n-
135 Far::TopologyRefiner const & _mesh;
\n-
136
\n-
137 int _numFaces;
\n-
138 int _numFVarChannels;
\n-
139};
\n-
140
\n+
112 bool IsRegular() const { return _data.isRegular(); }
\n+
113
\n+
115 bool IsLinear() const { return _data.isLinear(); }
\n+
117
\n+
119
\n+
135
\n+
137 int GetNumControlPoints() const { return _data.getNumCVs(); }
\n+
138
\n+
140 int GetControlPointIndices(Index meshPointIndices[]) const;
\n
141
\n-
142//
\n-
156template <class CACHE_TYPE = SurfaceFactoryCache>
\n-
157class RefinerSurfaceFactory : public RefinerSurfaceFactoryBase {
\n-
158public:
\n-
159 RefinerSurfaceFactory(Far::TopologyRefiner const & mesh,
\n-
160 Options const & options = Options()) :
\n-
161 RefinerSurfaceFactoryBase(mesh, options),
\n-
162 _localCache() {
\n-
163
\n-
164 SurfaceFactory::setInternalCache(&_localCache);
\n-
165 }
\n-
166 ~RefinerSurfaceFactory() override = default;
\n-
167
\n-
168private:
\n-
169 CACHE_TYPE _localCache;
\n-
170};
\n+
151 template <typename REAL_MESH>
\n+
152 void GatherControlPoints(REAL_MESH const meshPoints[],
\n+
153 PointDescriptor const & meshPointDesc,
\n+
154 REAL controlPoints[],
\n+
155 PointDescriptor const & controlPointDesc) const;
\n+
156
\n+
158 void BoundControlPoints(REAL const controlPoints[],
\n+
159 PointDescriptor const & controlPointDesc,
\n+
160 REAL minExtent[],
\n+
161 REAL maxExtent[]) const;
\n+
162
\n+
164 void BoundControlPointsFromMesh(REAL const meshPoints[],
\n+
165 PointDescriptor const & meshPointDesc,
\n+
166 REAL minExtent[],
\n+
167 REAL maxExtent[]) const;
\n+
169
\n
171
\n-
172} // end namespace Bfr
\n-
173
\n-
174} // end namespace OPENSUBDIV_VERSION
\n-
175using namespace OPENSUBDIV_VERSION;
\n-
176
\n-
177} // end namespace OpenSubdiv
\n-
178
\n-
179#endif /* OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H */
\n+
177
\n+
179 int GetNumPatchPoints() const;
\n+
180
\n+
199 void PreparePatchPoints(REAL const meshPoints[],
\n+
200 PointDescriptor const & meshPointDesc,
\n+
201 REAL patchPoints[],
\n+
202 PointDescriptor const & patchPointDesc) const;
\n+
203
\n+
213 void ComputePatchPoints(REAL patchPoints[],
\n+
214 PointDescriptor const & patchPointDesc) const;
\n+
216
\n+
218
\n+
225
\n+
227 void Evaluate(REAL const uv[2],
\n+
228 REAL const patchPoints[], PointDescriptor const & pointDesc,
\n+
229 REAL P[]) const;
\n+
230
\n+
232 void Evaluate(REAL const uv[2],
\n+
233 REAL const patchPoints[], PointDescriptor const & pointDesc,
\n+
234 REAL P[], REAL Du[], REAL Dv[]) const;
\n+
235
\n+
237 void Evaluate(REAL const uv[2],
\n+
238 REAL const patchPoints[], PointDescriptor const & pointDesc,
\n+
239 REAL P[], REAL Du[], REAL Dv[],
\n+
240 REAL Duu[], REAL Duv[], REAL Dvv[]) const;
\n+
242
\n+
244
\n+
256
\n+
258 int EvaluateStencil(REAL const uv[2], REAL sP[]) const;
\n+
259
\n+
261 int EvaluateStencil(REAL const uv[2], REAL sP[],
\n+
262 REAL sDu[], REAL sDv[]) const;
\n+
263
\n+
265 int EvaluateStencil(REAL const uv[2], REAL sP[],
\n+
266 REAL sDu[], REAL sDv[],
\n+
267 REAL sDuu[], REAL sDuv[], REAL sDvv[]) const;
\n+
268
\n+
270 void ApplyStencil(REAL const stencil[],
\n+
271 REAL const controlPoints[], PointDescriptor const &,
\n+
272 REAL result[]) const;
\n+
273
\n+
275 void ApplyStencilFromMesh(REAL const stencil[],
\n+
276 REAL const meshPoints[], PointDescriptor const &,
\n+
277 REAL result[]) const;
\n+
279
\n+
280private:
\n+
281 // Internal methods for evaluating derivatives, basis weights and
\n+
282 // stencils for regular, irregular and irregular linear patches:
\n+
283 typedef Vtr::ConstArray<int> IndexArray;
\n+
284
\n+
285 void evaluateDerivs(REAL const uv[2], REAL const patchPoints[],
\n+
286 PointDescriptor const &, REAL * derivs[]) const;
\n+
287 void evalRegularDerivs(REAL const uv[2], REAL const patchPoints[],
\n+
288 PointDescriptor const &, REAL * derivs[]) const;
\n+
289 void evalIrregularDerivs(REAL const uv[2], REAL const patchPoints[],
\n+
290 PointDescriptor const &, REAL * derivs[]) const;
\n+
291 void evalMultiLinearDerivs(REAL const uv[2], REAL const patchPoints[],
\n+
292 PointDescriptor const &, REAL * derivs[]) const;
\n+
293
\n+
294 void evalRegularBasis(REAL const uv[2], REAL * wDeriv[]) const;
\n+
295 IndexArray evalIrregularBasis(REAL const uv[2], REAL * wDeriv[]) const;
\n+
296 int evalMultiLinearBasis(REAL const uv[2], REAL * wDeriv[]) const;
\n+
297
\n+
298 int evaluateStencils(REAL const uv[2], REAL * sDeriv[]) const;
\n+
299 int evalRegularStencils(REAL const uv[2], REAL * sDeriv[]) const;
\n+
300 int evalIrregularStencils(REAL const uv[2], REAL * sDeriv[]) const;
\n+
301 int evalMultiLinearStencils(REAL const uv[2], REAL * sDeriv[]) const;
\n+
302
\n+
303 // Internal methods to compute patch points:
\n+
304 void computeLinearPatchPoints(REAL p[], PointDescriptor const &) const;
\n+
305 void computeIrregularPatchPoints(REAL p[], PointDescriptor const &) const;
\n+
306
\n+
307 // Internal methods specific to regular or irregular patches:
\n+
308 unsigned char getRegPatchType() const { return _data.getRegPatchType(); }
\n+
309 unsigned char getRegPatchMask() const { return _data.getRegPatchMask(); }
\n+
310
\n+
311 internal::IrregularPatchType const & getIrregPatch() const;
\n+
312
\n+
313private:
\n+
314 // Access to the set of member variables - provided to the Factory:
\n+
315 friend class SurfaceFactory;
\n+
316
\n+
317 internal::SurfaceData & getSurfaceData() { return _data; }
\n+
318 internal::SurfaceData const & getSurfaceData() const { return _data; }
\n+
319
\n+
320private:
\n+
321 // All member variables encapsulated in a single class:
\n+
322 internal::SurfaceData _data;
\n+
323};
\n+
324
\n+
325
\n+
326//
\n+
327// Simple inline methods composed of other methods:
\n+
328//
\n+
329template <typename REAL>
\n+
330inline void
\n+
331Surface<REAL>::ComputePatchPoints(REAL points[],
\n+
332 PointDescriptor const & pointDesc) const {
\n+
333
\n+
334 if (!IsRegular()) {
\n+
335 if (IsLinear()) {
\n+
336 computeLinearPatchPoints(points, pointDesc);
\n+
337 } else {
\n+
338 computeIrregularPatchPoints(points, pointDesc);
\n+
339 }
\n+
340 }
\n+
341}
\n+
342
\n+
343template <typename REAL>
\n+
344inline void
\n+
345Surface<REAL>::PreparePatchPoints(
\n+
346 REAL const meshPoints[], PointDescriptor const & meshPointDesc,
\n+
347 REAL patchPoints[], PointDescriptor const & patchPointDesc) const {
\n+
348
\n+
349 GatherControlPoints(meshPoints, meshPointDesc, patchPoints, patchPointDesc);
\n+
350 ComputePatchPoints(patchPoints, patchPointDesc);
\n+
351}
\n+
352
\n+
353//
\n+
354// Inline invocations of more general methods for derivative overloads:
\n+
355//
\n+
356template <typename REAL>
\n+
357inline void
\n+
358Surface<REAL>::evaluateDerivs(REAL const uv[2],
\n+
359 REAL const patchPoints[],
\n+
360 PointDescriptor const & pointDesc,
\n+
361 REAL * derivatives[]) const {
\n+
362 if (IsRegular()) {
\n+
363 evalRegularDerivs(uv, patchPoints, pointDesc, derivatives);
\n+
364 } else if (IsLinear()) {
\n+
365 evalMultiLinearDerivs(uv, patchPoints, pointDesc, derivatives);
\n+
366 } else {
\n+
367 evalIrregularDerivs(uv, patchPoints, pointDesc, derivatives);
\n+
368 }
\n+
369}
\n+
370template <typename REAL>
\n+
371inline void
\n+
372Surface<REAL>::Evaluate(REAL const uv[2],
\n+
373 REAL const patchPoints[],
\n+
374 PointDescriptor const & pointDesc,
\n+
375 REAL P[]) const {
\n+
376
\n+
377 REAL * derivatives[6] = { P, 0, 0, 0, 0, 0 };
\n+
378 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);
\n+
379}
\n+
380template <typename REAL>
\n+
381inline void
\n+
382Surface<REAL>::Evaluate(REAL const uv[2],
\n+
383 REAL const patchPoints[],
\n+
384 PointDescriptor const & pointDesc,
\n+
385 REAL P[], REAL Du[], REAL Dv[]) const {
\n+
386
\n+
387 REAL * derivatives[6] = { P, Du, Dv, 0, 0, 0 };
\n+
388 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);
\n+
389}
\n+
390template <typename REAL>
\n+
391inline void
\n+
392Surface<REAL>::Evaluate(REAL const uv[2],
\n+
393 REAL const patchPoints[],
\n+
394 PointDescriptor const & pointDesc,
\n+
395 REAL P[], REAL Du[], REAL Dv[],
\n+
396 REAL Duu[], REAL Duv[], REAL Dvv[]) const {
\n+
397
\n+
398 REAL * derivatives[6] = { P, Du, Dv, Duu, Duv, Dvv };
\n+
399 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);
\n+
400}
\n+
401
\n+
402template <typename REAL>
\n+
403inline int
\n+
404Surface<REAL>::evaluateStencils(REAL const uv[2], REAL * sDeriv[]) const {
\n+
405
\n+
406 if (IsRegular()) {
\n+
407 return evalRegularStencils(uv, sDeriv);
\n+
408 } else if (IsLinear()) {
\n+
409 return evalMultiLinearStencils(uv, sDeriv);
\n+
410 } else {
\n+
411 return evalIrregularStencils(uv, sDeriv);
\n+
412 }
\n+
413}
\n+
414template <typename REAL>
\n+
415inline int
\n+
416Surface<REAL>::EvaluateStencil(REAL const uv[2], REAL sP[]) const {
\n+
417
\n+
418 REAL * derivativeStencils[6] = { sP, 0, 0, 0, 0, 0 };
\n+
419 return evaluateStencils(uv, derivativeStencils);
\n+
420}
\n+
421template <typename REAL>
\n+
422inline int
\n+
423Surface<REAL>::EvaluateStencil(REAL const uv[2],
\n+
424 REAL sP[], REAL sDu[], REAL sDv[]) const {
\n+
425
\n+
426 REAL * derivativeStencils[6] = { sP, sDu, sDv, 0, 0, 0 };
\n+
427 return evaluateStencils(uv, derivativeStencils);
\n+
428}
\n+
429template <typename REAL>
\n+
430inline int
\n+
431Surface<REAL>::EvaluateStencil(REAL const uv[2],
\n+
432 REAL sP[], REAL sDu[], REAL sDv[],
\n+
433 REAL sDuu[], REAL sDuv[], REAL sDvv[]) const {
\n+
434
\n+
435 REAL * derivativeStencils[6] = { sP, sDu, sDv, sDuu, sDuv, sDvv };
\n+
436 return evaluateStencils(uv, derivativeStencils);
\n+
437}
\n+
438
\n+
439} // end namespace Bfr
\n+
440
\n+
441} // end namespace OPENSUBDIV_VERSION
\n+
442using namespace OPENSUBDIV_VERSION;
\n+
443
\n+
444} // end namespace OpenSubdiv
\n+
445
\n+
446#endif /* OPENSUBDIV3_BFR_SURFACE */
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase
Intermediate subclass of SurfaceFactory with Far::TopologyRefiner as the mesh.
Definition: refinerSurfaceFactory.h:58
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::GetNumFVarChannels
int GetNumFVarChannels() const
Return the number of face-varying channels.
Definition: refinerSurfaceFactory.h:85
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::GetMesh
Far::TopologyRefiner const & GetMesh() const
Return the instance of the mesh.
Definition: refinerSurfaceFactory.h:79
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::RefinerSurfaceFactoryBase
RefinerSurfaceFactoryBase(Far::TopologyRefiner const &mesh, Options const &options)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactory
Template for concrete subclasses of RefinerSurfaceFactoryBase.
Definition: refinerSurfaceFactory.h:157
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactory::RefinerSurfaceFactory
RefinerSurfaceFactory(Far::TopologyRefiner const &mesh, Options const &options=Options())
Definition: refinerSurfaceFactory.h:159
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Parameterization
Simple class defining the 2D parameterization of a face.
Definition: parameterization.h:52
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Parameterization::GetFaceSize
int GetFaceSize() const
Returns the size (number of vertices) of the corresponding face.
Definition: parameterization.h:101
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface
Encapsulates the limit surface for a face of a mesh.
Definition: surface.h:59
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetNumControlPoints
int GetNumControlPoints() const
Return the number of control points affecting the Surface.
Definition: surface.h:137
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Evaluate
void Evaluate(REAL const uv[2], REAL const patchPoints[], PointDescriptor const &pointDesc, REAL P[]) const
Evaluation of position.
Definition: surface.h:372
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetFaceSize
int GetFaceSize() const
Return the size of the face.
Definition: surface.h:109
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsRegular
bool IsRegular() const
Return if the Surface is a single regular patch.
Definition: surface.h:112
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::BoundControlPoints
void BoundControlPoints(REAL const controlPoints[], PointDescriptor const &controlPointDesc, REAL minExtent[], REAL maxExtent[]) const
Compute bounds of control points from a local array.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetControlPointIndices
int GetControlPointIndices(Index meshPointIndices[]) const
Identify indices of control points in the mesh.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Index
int Index
Integer type representing a mesh index.
Definition: surface.h:72
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ApplyStencil
void ApplyStencil(REAL const stencil[], REAL const controlPoints[], PointDescriptor const &, REAL result[]) const
Apply a single stencil to control points from a local array.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetParameterization
Parameterization GetParameterization() const
Return the Parameterization.
Definition: surface.h:106
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Surface
Surface()
Default construction produces an invalid instance.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ApplyStencilFromMesh
void ApplyStencilFromMesh(REAL const stencil[], REAL const meshPoints[], PointDescriptor const &, REAL result[]) const
Apply a single stencil to control points from the mesh data.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::operator=
Surface & operator=(Surface const &src)=default
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::BoundControlPointsFromMesh
void BoundControlPointsFromMesh(REAL const meshPoints[], PointDescriptor const &meshPointDesc, REAL minExtent[], REAL maxExtent[]) const
Compute bounds of control points from the mesh data.
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Clear
void Clear()
Clear a previously initialized Surface.
Definition: surface.h:89
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PreparePatchPoints
void PreparePatchPoints(REAL const meshPoints[], PointDescriptor const &meshPointDesc, REAL patchPoints[], PointDescriptor const &patchPointDesc) const
Prepare patch points in a local array for evaluation.
Definition: surface.h:345
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsValid
bool IsValid() const
Return true if successfully initialized.
Definition: surface.h:86
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GatherControlPoints
void GatherControlPoints(REAL_MESH const meshPoints[], PointDescriptor const &meshPointDesc, REAL controlPoints[], PointDescriptor const &controlPointDesc) const
Gather control points in a local array.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ComputePatchPoints
void ComputePatchPoints(REAL patchPoints[], PointDescriptor const &patchPointDesc) const
Compute all patch points following the control points.
Definition: surface.h:331
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsLinear
bool IsLinear() const
Return if the Surface is linear.
Definition: surface.h:115
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetNumPatchPoints
int GetNumPatchPoints() const
Return the number of patch points representing the Surface.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::EvaluateStencil
int EvaluateStencil(REAL const uv[2], REAL sP[]) const
Evaluation of the limit stencil for position.
Definition: surface.h:416
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor
Simple struct defining the size and stride of points in arrays.
Definition: surface.h:63
\n+\n+\n+\n+\n+\n
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory
Base class providing initialization of a Surface for each face of a mesh.
Definition: surfaceFactory.h:103
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::setInternalCache
void setInternalCache(SurfaceFactoryCache *cache)
Subclass to identify an internal cache for use by base class.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::Options
Simple set of options assigned to instances of SurfaceFactory.
Definition: surfaceFactory.h:116
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceVertexIncidentFaceFVarValueIndices
virtual int getFaceVertexIncidentFaceFVarValueIndices(Index faceIndex, int faceVertex, FVarID fvarID, Index fvarValueIndices[]) const =0
Gather face-varying indices of incident faces around a face-vertex.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceVertexIncidentFaceVertexIndices
virtual int getFaceVertexIncidentFaceVertexIndices(Index faceIndex, int faceVertex, Index vertexIndices[]) const =0
Gather vertex indices of incident faces around a face-vertex.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceSize
virtual int getFaceSize(Index faceIndex) const =0
Returns the size of a face (number of vertices)
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::populateFaceVertexDescriptor
virtual int populateFaceVertexDescriptor(Index faceIndex, int faceVertex, VertexDescriptor *vertexDescriptor) const =0
Describe the topology of incident faces around a face-vertex.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceNeighborhoodVertexIndicesIfRegular
virtual bool getFaceNeighborhoodVertexIndicesIfRegular(Index faceIndex, Index vertexIndices[]) const
Definition: surfaceFactoryMeshAdapter.h:218
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceNeighborhoodFVarValueIndicesIfRegular
virtual bool getFaceNeighborhoodFVarValueIndicesIfRegular(Index faceIndex, FVarID fvarID, Index fvarValueIndices[]) const
Definition: surfaceFactoryMeshAdapter.h:224
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceFVarValueIndices
virtual int getFaceFVarValueIndices(Index faceIndex, FVarID fvarID, Index fvarValueIndices[]) const =0
Gather the face-varying indices of the face's vertices.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceVertexIndices
virtual int getFaceVertexIndices(Index faceIndex, Index vertexIndices[]) const =0
Gather the indices of the face's vertices.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::isFaceHole
virtual bool isFaceHole(Index faceIndex) const =0
Returns if a face is a hole.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::FVarID
std::intptr_t FVarID
Type used to identify and specify face-varying primvars.
Definition: surfaceFactoryMeshAdapter.h:80
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::VertexDescriptor
Simple class used by subclasses of SurfaceFactory to describe a vertex.
Definition: vertexDescriptor.h:131
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner
Stores topology data for a specified set of refinement options.
Definition: topologyRefiner.h:51
\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-refinerSurfaceFactory.h\n+surface.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2021 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,222 +30,420 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H\n- 26#define OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H\n+ 25#ifndef OPENSUBDIV3_BFR_SURFACE_H\n+ 26#define OPENSUBDIV3_BFR_SURFACE_H\n 27\n 28#include \"../version.h\"\n 29\n- 30#include \"../bfr/surfaceFactory.h\"\n- 31#include \"../bfr/surfaceFactoryCache.h\"\n- 32\n- 33namespace OpenSubdiv {\n- 34namespace OPENSUBDIV_VERSION {\n- 35\n-36namespace Far {\n- 37 class TopologyRefiner;\n- 38}\n- 39\n- 40namespace Bfr {\n- 41\n-58class RefinerSurfaceFactoryBase : public SurfaceFactory {\n- 59public:\n- 61\n- 65\n-66 RefinerSurfaceFactoryBase(Far::TopologyRefiner const & mesh,\n- 67 Options const & options);\n- 68\n-69 ~RefinerSurfaceFactoryBase() override = default;\n- 71\n+ 30#include \"../bfr/surfaceData.h\"\n+ 31#include \"../bfr/parameterization.h\"\n+ 32#include \"../vtr/array.h\"\n+ 33\n+ 34namespace OpenSubdiv {\n+ 35namespace OPENSUBDIV_VERSION {\n+ 36\n+ 37namespace Bfr {\n+ 38\n+ 58template \n+59class Surface {\n+ 60public:\n+63 struct PointDescriptor {\n+64 PointDescriptor() : size(0), stride(0) { }\n+65 PointDescriptor(int n) : size(n), stride(n) { }\n+66 PointDescriptor(int n, int m) : size(n), stride(m) { }\n+ 67\n+68 int size, stride;\n+ 69 };\n+ 70\n+72 typedef int Index;\n 73\n- 77\n-79 Far::TopologyRefiner const & GetMesh() const { return _mesh; }\n- 80\n-82 int GetNumFaces() const { return _numFaces; }\n- 83\n-85 int GetNumFVarChannels() const { return _numFVarChannels; }\n+ 74public:\n+ 76\n+ 84\n+86 bool IsValid() const { return _data.isValid(); }\n 87\n- 88protected:\n- 90 //\n- 91 // Virtual overrides to satisfy the SurfaceFactoryMeshAdapter interface:\n- 92 //\n- 93 bool isFaceHole( Index faceIndex) const override;\n- 94 int getFaceSize(Index faceIndex) const override;\n- 95\n- 96 int getFaceVertexIndices(Index faceIndex,\n- 97 Index vertexIndices[]) const override;\n- 98 int getFaceFVarValueIndices(Index faceIndex,\n- 99 FVarID fvarID, Index fvarValueIndices[]) const override;\n+89 void Clear() { _data.reinitialize(); }\n+ 90\n+92 Surface();\n+ 93\n+94 Surface(Surface const & src) = default;\n+95 Surface& operator=(Surface const & src) = default;\n+96 ~Surface() = default;\n+ 98\n 100\n- 101 int populateFaceVertexDescriptor(Index faceIndex, int faceVertex,\n- 102 VertexDescriptor * vertexDescriptor) const override;\n- 103\n- 104 int getFaceVertexIncidentFaceVertexIndices(\n- 105 Index faceIndex, int faceVertex,\n- 106 Index vertexIndices[]) const override;\n- 107 int getFaceVertexIncidentFaceFVarValueIndices(\n- 108 Index faceIndex, int faceVertex,\n- 109 FVarID fvarID, Index fvarValueIndices[]) const override;\n+ 104\n+106 Parameterization GetParameterization() const { return _data.getParam(); }\n+ 107\n+109 int GetFaceSize() const { return GetParameterization().GetFaceSize(); }\n 110\n- 111 // Optional SurfaceFactoryMeshAdapter overrides for regular patches:\n- 112 bool getFaceNeighborhoodVertexIndicesIfRegular(\n- 113 Index faceIndex,\n- 114 Index vertexIndices[]) const override;\n- 115\n- 116 bool getFaceNeighborhoodFVarValueIndicesIfRegular(\n- 117 Index faceIndex,\n- 118 FVarID fvarID, Index fvarValueIndices[]) const override;\n- 120\n- 121private:\n- 122 //\n- 123 // Internal supporting methods:\n- 124 //\n- 125 int getFaceVaryingChannel(FVarID fvarID) const;\n- 126\n- 127 int getFaceVertexPointIndices(Index faceIndex, int faceVertex,\n- 128 Index indices[], int vtxOrFVarChannel) const;\n- 129\n- 130 int getFacePatchPointIndices(Index faceIndex,\n- 131 Index indices[], int vtxOrFVarChannel) const;\n- 132\n- 133private:\n- 134 // Additional members for the subclass:\n- 135 Far::TopologyRefiner const & _mesh;\n- 136\n- 137 int _numFaces;\n- 138 int _numFVarChannels;\n- 139};\n- 140\n+112 bool IsRegular() const { return _data.isRegular(); }\n+ 113\n+115 bool IsLinear() const { return _data.isLinear(); }\n+ 117\n+ 119\n+ 135\n+137 int GetNumControlPoints() const { return _data.getNumCVs(); }\n+ 138\n+140 int GetControlPointIndices(Index meshPointIndices[]) const;\n 141\n- 142//\n- 156template \n-157class RefinerSurfaceFactory : public RefinerSurfaceFactoryBase {\n- 158public:\n-159 RefinerSurfaceFactory(Far::TopologyRefiner const & mesh,\n- 160 Options const & options = Options()) :\n- 161 RefinerSurfaceFactoryBase(mesh, options),\n- 162 _localCache() {\n- 163\n- 164 SurfaceFactory::setInternalCache(&_localCache);\n- 165 }\n-166 ~RefinerSurfaceFactory() override = default;\n- 167\n- 168private:\n- 169 CACHE_TYPE _localCache;\n- 170};\n+ 151 template \n+152 void GatherControlPoints(REAL_MESH const meshPoints[],\n+ 153 PointDescriptor const & meshPointDesc,\n+ 154 REAL controlPoints[],\n+ 155 PointDescriptor const & controlPointDesc) const;\n+ 156\n+158 void BoundControlPoints(REAL const controlPoints[],\n+ 159 PointDescriptor const & controlPointDesc,\n+ 160 REAL minExtent[],\n+ 161 REAL maxExtent[]) const;\n+ 162\n+164 void BoundControlPointsFromMesh(REAL const meshPoints[],\n+ 165 PointDescriptor const & meshPointDesc,\n+ 166 REAL minExtent[],\n+ 167 REAL maxExtent[]) const;\n+ 169\n 171\n- 172} // end namespace Bfr\n- 173\n- 174} // end namespace OPENSUBDIV_VERSION\n- 175using namespace OPENSUBDIV_VERSION;\n- 176\n- 177} // end namespace OpenSubdiv\n- 178\n- 179#endif /* OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H */\n+ 177\n+179 int GetNumPatchPoints() const;\n+ 180\n+ 199 void PreparePatchPoints(REAL const meshPoints[],\n+ 200 PointDescriptor const & meshPointDesc,\n+ 201 REAL patchPoints[],\n+ 202 PointDescriptor const & patchPointDesc) const;\n+ 203\n+ 213 void ComputePatchPoints(REAL patchPoints[],\n+ 214 PointDescriptor const & patchPointDesc) const;\n+ 216\n+ 218\n+ 225\n+ 227 void Evaluate(REAL const uv[2],\n+ 228 REAL const patchPoints[], PointDescriptor const & pointDesc,\n+ 229 REAL P[]) const;\n+ 230\n+ 232 void Evaluate(REAL const uv[2],\n+ 233 REAL const patchPoints[], PointDescriptor const & pointDesc,\n+ 234 REAL P[], REAL Du[], REAL Dv[]) const;\n+ 235\n+ 237 void Evaluate(REAL const uv[2],\n+ 238 REAL const patchPoints[], PointDescriptor const & pointDesc,\n+ 239 REAL P[], REAL Du[], REAL Dv[],\n+ 240 REAL Duu[], REAL Duv[], REAL Dvv[]) const;\n+ 242\n+ 244\n+ 256\n+ 258 int EvaluateStencil(REAL const uv[2], REAL sP[]) const;\n+ 259\n+ 261 int EvaluateStencil(REAL const uv[2], REAL sP[],\n+ 262 REAL sDu[], REAL sDv[]) const;\n+ 263\n+ 265 int EvaluateStencil(REAL const uv[2], REAL sP[],\n+ 266 REAL sDu[], REAL sDv[],\n+ 267 REAL sDuu[], REAL sDuv[], REAL sDvv[]) const;\n+ 268\n+270 void ApplyStencil(REAL const stencil[],\n+ 271 REAL const controlPoints[], PointDescriptor const &,\n+ 272 REAL result[]) const;\n+ 273\n+275 void ApplyStencilFromMesh(REAL const stencil[],\n+ 276 REAL const meshPoints[], PointDescriptor const &,\n+ 277 REAL result[]) const;\n+ 279\n+ 280private:\n+ 281 // Internal methods for evaluating derivatives, basis weights and\n+ 282 // stencils for regular, irregular and irregular linear patches:\n+ 283 typedef Vtr::ConstArray IndexArray;\n+ 284\n+ 285 void evaluateDerivs(REAL const uv[2], REAL const patchPoints[],\n+ 286 PointDescriptor const &, REAL * derivs[]) const;\n+ 287 void evalRegularDerivs(REAL const uv[2], REAL const patchPoints[],\n+ 288 PointDescriptor const &, REAL * derivs[]) const;\n+ 289 void evalIrregularDerivs(REAL const uv[2], REAL const patchPoints[],\n+ 290 PointDescriptor const &, REAL * derivs[]) const;\n+ 291 void evalMultiLinearDerivs(REAL const uv[2], REAL const patchPoints[],\n+ 292 PointDescriptor const &, REAL * derivs[]) const;\n+ 293\n+ 294 void evalRegularBasis(REAL const uv[2], REAL * wDeriv[]) const;\n+ 295 IndexArray evalIrregularBasis(REAL const uv[2], REAL * wDeriv[]) const;\n+ 296 int evalMultiLinearBasis(REAL const uv[2], REAL * wDeriv[]) const;\n+ 297\n+ 298 int evaluateStencils(REAL const uv[2], REAL * sDeriv[]) const;\n+ 299 int evalRegularStencils(REAL const uv[2], REAL * sDeriv[]) const;\n+ 300 int evalIrregularStencils(REAL const uv[2], REAL * sDeriv[]) const;\n+ 301 int evalMultiLinearStencils(REAL const uv[2], REAL * sDeriv[]) const;\n+ 302\n+ 303 // Internal methods to compute patch points:\n+ 304 void computeLinearPatchPoints(REAL p[], PointDescriptor const &) const;\n+ 305 void computeIrregularPatchPoints(REAL p[], PointDescriptor const &) const;\n+ 306\n+ 307 // Internal methods specific to regular or irregular patches:\n+ 308 unsigned char getRegPatchType() const { return _data.getRegPatchType(); }\n+ 309 unsigned char getRegPatchMask() const { return _data.getRegPatchMask(); }\n+ 310\n+ 311 internal::IrregularPatchType const & getIrregPatch() const;\n+ 312\n+ 313private:\n+ 314 // Access to the set of member variables - provided to the Factory:\n+315 friend class SurfaceFactory;\n+ 316\n+ 317 internal::SurfaceData & getSurfaceData() { return _data; }\n+ 318 internal::SurfaceData const & getSurfaceData() const { return _data; }\n+ 319\n+ 320private:\n+ 321 // All member variables encapsulated in a single class:\n+ 322 internal::SurfaceData _data;\n+ 323};\n+ 324\n+ 325\n+ 326//\n+ 327// Simple inline methods composed of other methods:\n+ 328//\n+ 329template \n+ 330inline void\n+331Surface::ComputePatchPoints(REAL points[],\n+ 332 PointDescriptor const & pointDesc) const {\n+ 333\n+ 334 if (!IsRegular()) {\n+ 335 if (IsLinear()) {\n+ 336 computeLinearPatchPoints(points, pointDesc);\n+ 337 } else {\n+ 338 computeIrregularPatchPoints(points, pointDesc);\n+ 339 }\n+ 340 }\n+ 341}\n+ 342\n+ 343template \n+ 344inline void\n+345Surface::PreparePatchPoints(\n+ 346 REAL const meshPoints[], PointDescriptor const & meshPointDesc,\n+ 347 REAL patchPoints[], PointDescriptor const & patchPointDesc) const {\n+ 348\n+ 349 GatherControlPoints(meshPoints, meshPointDesc, patchPoints,\n+patchPointDesc);\n+ 350 ComputePatchPoints(patchPoints, patchPointDesc);\n+ 351}\n+ 352\n+ 353//\n+ 354// Inline invocations of more general methods for derivative overloads:\n+ 355//\n+ 356template \n+ 357inline void\n+ 358Surface::evaluateDerivs(REAL const uv[2],\n+ 359 REAL const patchPoints[],\n+ 360 PointDescriptor const & pointDesc,\n+ 361 REAL * derivatives[]) const {\n+ 362 if (IsRegular()) {\n+ 363 evalRegularDerivs(uv, patchPoints, pointDesc, derivatives);\n+ 364 } else if (IsLinear()) {\n+ 365 evalMultiLinearDerivs(uv, patchPoints, pointDesc, derivatives);\n+ 366 } else {\n+ 367 evalIrregularDerivs(uv, patchPoints, pointDesc, derivatives);\n+ 368 }\n+ 369}\n+ 370template \n+ 371inline void\n+372Surface::Evaluate(REAL const uv[2],\n+ 373 REAL const patchPoints[],\n+ 374 PointDescriptor const & pointDesc,\n+ 375 REAL P[]) const {\n+ 376\n+ 377 REAL * derivatives[6] = { P, 0, 0, 0, 0, 0 };\n+ 378 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);\n+ 379}\n+ 380template \n+ 381inline void\n+382Surface::Evaluate(REAL const uv[2],\n+ 383 REAL const patchPoints[],\n+ 384 PointDescriptor const & pointDesc,\n+ 385 REAL P[], REAL Du[], REAL Dv[]) const {\n+ 386\n+ 387 REAL * derivatives[6] = { P, Du, Dv, 0, 0, 0 };\n+ 388 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);\n+ 389}\n+ 390template \n+ 391inline void\n+392Surface::Evaluate(REAL const uv[2],\n+ 393 REAL const patchPoints[],\n+ 394 PointDescriptor const & pointDesc,\n+ 395 REAL P[], REAL Du[], REAL Dv[],\n+ 396 REAL Duu[], REAL Duv[], REAL Dvv[]) const {\n+ 397\n+ 398 REAL * derivatives[6] = { P, Du, Dv, Duu, Duv, Dvv };\n+ 399 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);\n+ 400}\n+ 401\n+ 402template \n+ 403inline int\n+ 404Surface::evaluateStencils(REAL const uv[2], REAL * sDeriv[]) const {\n+ 405\n+ 406 if (IsRegular()) {\n+ 407 return evalRegularStencils(uv, sDeriv);\n+ 408 } else if (IsLinear()) {\n+ 409 return evalMultiLinearStencils(uv, sDeriv);\n+ 410 } else {\n+ 411 return evalIrregularStencils(uv, sDeriv);\n+ 412 }\n+ 413}\n+ 414template \n+ 415inline int\n+416Surface::EvaluateStencil(REAL const uv[2], REAL sP[]) const {\n+ 417\n+ 418 REAL * derivativeStencils[6] = { sP, 0, 0, 0, 0, 0 };\n+ 419 return evaluateStencils(uv, derivativeStencils);\n+ 420}\n+ 421template \n+ 422inline int\n+423Surface::EvaluateStencil(REAL const uv[2],\n+ 424 REAL sP[], REAL sDu[], REAL sDv[]) const {\n+ 425\n+ 426 REAL * derivativeStencils[6] = { sP, sDu, sDv, 0, 0, 0 };\n+ 427 return evaluateStencils(uv, derivativeStencils);\n+ 428}\n+ 429template \n+ 430inline int\n+431Surface::EvaluateStencil(REAL const uv[2],\n+ 432 REAL sP[], REAL sDu[], REAL sDv[],\n+ 433 REAL sDuu[], REAL sDuv[], REAL sDvv[]) const {\n+ 434\n+ 435 REAL * derivativeStencils[6] = { sP, sDu, sDv, sDuu, sDuv, sDvv };\n+ 436 return evaluateStencils(uv, derivativeStencils);\n+ 437}\n+ 438\n+ 439} // end namespace Bfr\n+ 440\n+ 441} // end namespace OPENSUBDIV_VERSION\n+ 442using namespace OPENSUBDIV_VERSION;\n+ 443\n+ 444} // end namespace OpenSubdiv\n+ 445\n+ 446#endif /* OPENSUBDIV3_BFR_SURFACE */\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase\n-Intermediate subclass of SurfaceFactory with Far::TopologyRefiner as the mesh.\n-Definition: refinerSurfaceFactory.h:58\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::GetNumFaces\n-int GetNumFaces() const\n-Return the number of faces.\n-Definition: refinerSurfaceFactory.h:82\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::\n-GetNumFVarChannels\n-int GetNumFVarChannels() const\n-Return the number of face-varying channels.\n-Definition: refinerSurfaceFactory.h:85\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::GetMesh\n-Far::TopologyRefiner const & GetMesh() const\n-Return the instance of the mesh.\n-Definition: refinerSurfaceFactory.h:79\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::\n-~RefinerSurfaceFactoryBase\n-~RefinerSurfaceFactoryBase() override=default\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::\n-RefinerSurfaceFactoryBase\n-RefinerSurfaceFactoryBase(Far::TopologyRefiner const &mesh, Options const\n-&options)\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactory\n-Template for concrete subclasses of RefinerSurfaceFactoryBase.\n-Definition: refinerSurfaceFactory.h:157\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactory::\n-~RefinerSurfaceFactory\n-~RefinerSurfaceFactory() override=default\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactory::\n-RefinerSurfaceFactory\n-RefinerSurfaceFactory(Far::TopologyRefiner const &mesh, Options const\n-&options=Options())\n-Definition: refinerSurfaceFactory.h:159\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Parameterization\n+Simple class defining the 2D parameterization of a face.\n+Definition: parameterization.h:52\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Parameterization::GetFaceSize\n+int GetFaceSize() const\n+Returns the size (number of vertices) of the corresponding face.\n+Definition: parameterization.h:101\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface\n+Encapsulates the limit surface for a face of a mesh.\n+Definition: surface.h:59\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetNumControlPoints\n+int GetNumControlPoints() const\n+Return the number of control points affecting the Surface.\n+Definition: surface.h:137\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Evaluate\n+void Evaluate(REAL const uv[2], REAL const patchPoints[], PointDescriptor const\n+&pointDesc, REAL P[]) const\n+Evaluation of position.\n+Definition: surface.h:372\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetFaceSize\n+int GetFaceSize() const\n+Return the size of the face.\n+Definition: surface.h:109\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsRegular\n+bool IsRegular() const\n+Return if the Surface is a single regular patch.\n+Definition: surface.h:112\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::BoundControlPoints\n+void BoundControlPoints(REAL const controlPoints[], PointDescriptor const\n+&controlPointDesc, REAL minExtent[], REAL maxExtent[]) const\n+Compute bounds of control points from a local array.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetControlPointIndices\n+int GetControlPointIndices(Index meshPointIndices[]) const\n+Identify indices of control points in the mesh.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Index\n+int Index\n+Integer type representing a mesh index.\n+Definition: surface.h:72\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ApplyStencil\n+void ApplyStencil(REAL const stencil[], REAL const controlPoints[],\n+PointDescriptor const &, REAL result[]) const\n+Apply a single stencil to control points from a local array.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetParameterization\n+Parameterization GetParameterization() const\n+Return the Parameterization.\n+Definition: surface.h:106\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Surface\n+Surface()\n+Default construction produces an invalid instance.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ApplyStencilFromMesh\n+void ApplyStencilFromMesh(REAL const stencil[], REAL const meshPoints[],\n+PointDescriptor const &, REAL result[]) const\n+Apply a single stencil to control points from the mesh data.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::operator=\n+Surface & operator=(Surface const &src)=default\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::BoundControlPointsFromMesh\n+void BoundControlPointsFromMesh(REAL const meshPoints[], PointDescriptor const\n+&meshPointDesc, REAL minExtent[], REAL maxExtent[]) const\n+Compute bounds of control points from the mesh data.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::~Surface\n+~Surface()=default\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Clear\n+void Clear()\n+Clear a previously initialized Surface.\n+Definition: surface.h:89\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Surface\n+Surface(Surface const &src)=default\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PreparePatchPoints\n+void PreparePatchPoints(REAL const meshPoints[], PointDescriptor const\n+&meshPointDesc, REAL patchPoints[], PointDescriptor const &patchPointDesc)\n+const\n+Prepare patch points in a local array for evaluation.\n+Definition: surface.h:345\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsValid\n+bool IsValid() const\n+Return true if successfully initialized.\n+Definition: surface.h:86\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GatherControlPoints\n+void GatherControlPoints(REAL_MESH const meshPoints[], PointDescriptor const\n+&meshPointDesc, REAL controlPoints[], PointDescriptor const &controlPointDesc)\n+const\n+Gather control points in a local array.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ComputePatchPoints\n+void ComputePatchPoints(REAL patchPoints[], PointDescriptor const\n+&patchPointDesc) const\n+Compute all patch points following the control points.\n+Definition: surface.h:331\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsLinear\n+bool IsLinear() const\n+Return if the Surface is linear.\n+Definition: surface.h:115\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetNumPatchPoints\n+int GetNumPatchPoints() const\n+Return the number of patch points representing the Surface.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::EvaluateStencil\n+int EvaluateStencil(REAL const uv[2], REAL sP[]) const\n+Evaluation of the limit stencil for position.\n+Definition: surface.h:416\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor\n+Simple struct defining the size and stride of points in arrays.\n+Definition: surface.h:63\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor::stride\n+int stride\n+Definition: surface.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor::size\n+int size\n+Definition: surface.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor::PointDescriptor\n+PointDescriptor(int n)\n+Definition: surface.h:65\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor::PointDescriptor\n+PointDescriptor(int n, int m)\n+Definition: surface.h:66\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor::PointDescriptor\n+PointDescriptor()\n+Definition: surface.h:64\n OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory\n Base class providing initialization of a Surface for each face of a mesh.\n Definition: surfaceFactory.h:103\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::setInternalCache\n-void setInternalCache(SurfaceFactoryCache *cache)\n-Subclass to identify an internal cache for use by base class.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::Options\n-Simple set of options assigned to instances of SurfaceFactory.\n-Definition: surfaceFactory.h:116\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n-getFaceVertexIncidentFaceFVarValueIndices\n-virtual int getFaceVertexIncidentFaceFVarValueIndices(Index faceIndex, int\n-faceVertex, FVarID fvarID, Index fvarValueIndices[]) const =0\n-Gather face-varying indices of incident faces around a face-vertex.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n-getFaceVertexIncidentFaceVertexIndices\n-virtual int getFaceVertexIncidentFaceVertexIndices(Index faceIndex, int\n-faceVertex, Index vertexIndices[]) const =0\n-Gather vertex indices of incident faces around a face-vertex.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceSize\n-virtual int getFaceSize(Index faceIndex) const =0\n-Returns the size of a face (number of vertices)\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::Index\n-int Index\n-Integer type representing a mesh index.\n-Definition: surfaceFactoryMeshAdapter.h:68\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n-populateFaceVertexDescriptor\n-virtual int populateFaceVertexDescriptor(Index faceIndex, int faceVertex,\n-VertexDescriptor *vertexDescriptor) const =0\n-Describe the topology of incident faces around a face-vertex.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n-getFaceNeighborhoodVertexIndicesIfRegular\n-virtual bool getFaceNeighborhoodVertexIndicesIfRegular(Index faceIndex, Index\n-vertexIndices[]) const\n-Definition: surfaceFactoryMeshAdapter.h:218\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n-getFaceNeighborhoodFVarValueIndicesIfRegular\n-virtual bool getFaceNeighborhoodFVarValueIndicesIfRegular(Index faceIndex,\n-FVarID fvarID, Index fvarValueIndices[]) const\n-Definition: surfaceFactoryMeshAdapter.h:224\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n-getFaceFVarValueIndices\n-virtual int getFaceFVarValueIndices(Index faceIndex, FVarID fvarID, Index\n-fvarValueIndices[]) const =0\n-Gather the face-varying indices of the face's vertices.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n-getFaceVertexIndices\n-virtual int getFaceVertexIndices(Index faceIndex, Index vertexIndices[]) const\n-=0\n-Gather the indices of the face's vertices.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::isFaceHole\n-virtual bool isFaceHole(Index faceIndex) const =0\n-Returns if a face is a hole.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::FVarID\n-std::intptr_t FVarID\n-Type used to identify and specify face-varying primvars.\n-Definition: surfaceFactoryMeshAdapter.h:80\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::VertexDescriptor\n-Simple class used by subclasses of SurfaceFactory to describe a vertex.\n-Definition: vertexDescriptor.h:131\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner\n-Stores topology data for a specified set of refinement options.\n-Definition: topologyRefiner.h:51\n+OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstArray\n+Definition: array.h:53\n * opensubdiv\n * bfr\n- * refinerSurfaceFactory.h\n+ * surface.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00692.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00692.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/bfr/surface.h File Reference\n+OpenSubdiv: opensubdiv/bfr/refinerSurfaceFactory.h File Reference\n \n \n \n \n \n \n \n@@ -83,46 +83,47 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
surface.h File Reference
\n+
refinerSurfaceFactory.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../bfr/surfaceData.h"
\n-#include "../bfr/parameterization.h"
\n-#include "../vtr/array.h"
\n+#include "../bfr/surfaceFactory.h"
\n+#include "../bfr/surfaceFactoryCache.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n+\n+\n \n-\n-\n+\n+\n \n

\n Classes

class  Surface< REAL >
 Encapsulates the limit surface for a face of a mesh. More...
class  RefinerSurfaceFactoryBase
 Intermediate subclass of SurfaceFactory with Far::TopologyRefiner as the mesh. More...
 
struct  Surface< REAL >::PointDescriptor
 Simple struct defining the size and stride of points in arrays. More...
class  RefinerSurfaceFactory< CACHE_TYPE >
 Template for concrete subclasses of RefinerSurfaceFactoryBase. More...
 
\n \n \n \n \n \n+\n+\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Bfr
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,32 +5,33 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-surface.h File Reference\n+refinerSurfaceFactory.h File Reference\n #include \"../version.h\"\n-#include \"../bfr/surfaceData.h\"\n-#include \"../bfr/parameterization.h\"\n-#include \"../vtr/array.h\"\n+#include \"../bfr/surfaceFactory.h\"\n+#include \"../bfr/surfaceFactoryCache.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n- class \u00a0Surface<_REAL_>\n-\u00a0 Encapsulates the limit surface for a face of a mesh. More...\n+class \u00a0RefinerSurfaceFactoryBase\n+\u00a0 Intermediate subclass of SurfaceFactory with Far::TopologyRefiner as\n+ the mesh. More...\n \u00a0\n-struct \u00a0Surface<_REAL_>::PointDescriptor\n-\u00a0 Simple struct defining the size and stride of points in arrays.\n- More...\n+class \u00a0RefinerSurfaceFactory<_CACHE_TYPE_>\n+\u00a0 Template for concrete subclasses of RefinerSurfaceFactoryBase. More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n+\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Bfr\n \u00a0\n * opensubdiv\n * bfr\n- * surface.h\n+ * refinerSurfaceFactory.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00692.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00692.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,4 @@\n var a00692 = [\n- [\"Surface< REAL >\", \"a00933.html\", \"a00933\"],\n- [\"Surface< REAL >::PointDescriptor\", \"a00937.html\", \"a00937\"]\n+ [\"RefinerSurfaceFactoryBase\", \"a00925.html\", \"a00925\"],\n+ [\"RefinerSurfaceFactory< CACHE_TYPE >\", \"a00929.html\", \"a00929\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00692_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00692_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/bfr/surface.h Source File\n+OpenSubdiv: opensubdiv/bfr/refinerSurfaceFactory.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
surface.h
\n+
refinerSurfaceFactory.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2021 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,338 +107,155 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_BFR_SURFACE_H
\n-
26#define OPENSUBDIV3_BFR_SURFACE_H
\n+
25#ifndef OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H
\n+
26#define OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30#include "../bfr/surfaceData.h"
\n-
31#include "../bfr/parameterization.h"
\n-
32#include "../vtr/array.h"
\n-
33
\n-
34namespace OpenSubdiv {
\n-
35namespace OPENSUBDIV_VERSION {
\n-
36
\n-
37namespace Bfr {
\n-
38
\n-
58template <typename REAL>
\n-
59class Surface {
\n-
60public:
\n-
63 struct PointDescriptor {
\n-
64 PointDescriptor() : size(0), stride(0) { }
\n-
65 PointDescriptor(int n) : size(n), stride(n) { }
\n-
66 PointDescriptor(int n, int m) : size(n), stride(m) { }
\n-
67
\n-
68 int size, stride;
\n-
69 };
\n-
70
\n-
72 typedef int Index;
\n+
30#include "../bfr/surfaceFactory.h"
\n+
31#include "../bfr/surfaceFactoryCache.h"
\n+
32
\n+
33namespace OpenSubdiv {
\n+
34namespace OPENSUBDIV_VERSION {
\n+
35
\n+
36namespace Far {
\n+
37 class TopologyRefiner;
\n+
38}
\n+
39
\n+
40namespace Bfr {
\n+
41
\n+
58class RefinerSurfaceFactoryBase : public SurfaceFactory {
\n+
59public:
\n+
61
\n+
65
\n+
66 RefinerSurfaceFactoryBase(Far::TopologyRefiner const & mesh,
\n+
67 Options const & options);
\n+
68
\n+
69 ~RefinerSurfaceFactoryBase() override = default;
\n+
71
\n
73
\n-
74public:
\n-
76
\n-
84
\n-
86 bool IsValid() const { return _data.isValid(); }
\n+
77
\n+
79 Far::TopologyRefiner const & GetMesh() const { return _mesh; }
\n+
80
\n+
82 int GetNumFaces() const { return _numFaces; }
\n+
83
\n+
85 int GetNumFVarChannels() const { return _numFVarChannels; }
\n
87
\n-
89 void Clear() { _data.reinitialize(); }
\n-
90
\n-
92 Surface();
\n-
93
\n-
94 Surface(Surface const & src) = default;
\n-
95 Surface& operator=(Surface const & src) = default;
\n-
96 ~Surface() = default;
\n-
98
\n+
88protected:
\n+
90 //
\n+
91 // Virtual overrides to satisfy the SurfaceFactoryMeshAdapter interface:
\n+
92 //
\n+
93 bool isFaceHole( Index faceIndex) const override;
\n+
94 int getFaceSize(Index faceIndex) const override;
\n+
95
\n+
96 int getFaceVertexIndices(Index faceIndex,
\n+
97 Index vertexIndices[]) const override;
\n+
98 int getFaceFVarValueIndices(Index faceIndex,
\n+
99 FVarID fvarID, Index fvarValueIndices[]) const override;
\n
100
\n-
104
\n-
106 Parameterization GetParameterization() const { return _data.getParam(); }
\n-
107
\n-
109 int GetFaceSize() const { return GetParameterization().GetFaceSize(); }
\n+
101 int populateFaceVertexDescriptor(Index faceIndex, int faceVertex,
\n+
102 VertexDescriptor * vertexDescriptor) const override;
\n+
103
\n+
104 int getFaceVertexIncidentFaceVertexIndices(
\n+
105 Index faceIndex, int faceVertex,
\n+
106 Index vertexIndices[]) const override;
\n+
107 int getFaceVertexIncidentFaceFVarValueIndices(
\n+
108 Index faceIndex, int faceVertex,
\n+
109 FVarID fvarID, Index fvarValueIndices[]) const override;
\n
110
\n-
112 bool IsRegular() const { return _data.isRegular(); }
\n-
113
\n-
115 bool IsLinear() const { return _data.isLinear(); }
\n-
117
\n-
119
\n-
135
\n-
137 int GetNumControlPoints() const { return _data.getNumCVs(); }
\n-
138
\n-
140 int GetControlPointIndices(Index meshPointIndices[]) const;
\n+
111 // Optional SurfaceFactoryMeshAdapter overrides for regular patches:
\n+
112 bool getFaceNeighborhoodVertexIndicesIfRegular(
\n+
113 Index faceIndex,
\n+
114 Index vertexIndices[]) const override;
\n+
115
\n+
116 bool getFaceNeighborhoodFVarValueIndicesIfRegular(
\n+
117 Index faceIndex,
\n+
118 FVarID fvarID, Index fvarValueIndices[]) const override;
\n+
120
\n+
121private:
\n+
122 //
\n+
123 // Internal supporting methods:
\n+
124 //
\n+
125 int getFaceVaryingChannel(FVarID fvarID) const;
\n+
126
\n+
127 int getFaceVertexPointIndices(Index faceIndex, int faceVertex,
\n+
128 Index indices[], int vtxOrFVarChannel) const;
\n+
129
\n+
130 int getFacePatchPointIndices(Index faceIndex,
\n+
131 Index indices[], int vtxOrFVarChannel) const;
\n+
132
\n+
133private:
\n+
134 // Additional members for the subclass:
\n+
135 Far::TopologyRefiner const & _mesh;
\n+
136
\n+
137 int _numFaces;
\n+
138 int _numFVarChannels;
\n+
139};
\n+
140
\n
141
\n-
151 template <typename REAL_MESH>
\n-
152 void GatherControlPoints(REAL_MESH const meshPoints[],
\n-
153 PointDescriptor const & meshPointDesc,
\n-
154 REAL controlPoints[],
\n-
155 PointDescriptor const & controlPointDesc) const;
\n-
156
\n-
158 void BoundControlPoints(REAL const controlPoints[],
\n-
159 PointDescriptor const & controlPointDesc,
\n-
160 REAL minExtent[],
\n-
161 REAL maxExtent[]) const;
\n-
162
\n-
164 void BoundControlPointsFromMesh(REAL const meshPoints[],
\n-
165 PointDescriptor const & meshPointDesc,
\n-
166 REAL minExtent[],
\n-
167 REAL maxExtent[]) const;
\n-
169
\n+
142//
\n+
156template <class CACHE_TYPE = SurfaceFactoryCache>
\n+
157class RefinerSurfaceFactory : public RefinerSurfaceFactoryBase {
\n+
158public:
\n+
159 RefinerSurfaceFactory(Far::TopologyRefiner const & mesh,
\n+
160 Options const & options = Options()) :
\n+
161 RefinerSurfaceFactoryBase(mesh, options),
\n+
162 _localCache() {
\n+
163
\n+
164 SurfaceFactory::setInternalCache(&_localCache);
\n+
165 }
\n+
166 ~RefinerSurfaceFactory() override = default;
\n+
167
\n+
168private:
\n+
169 CACHE_TYPE _localCache;
\n+
170};
\n
171
\n-
177
\n-
179 int GetNumPatchPoints() const;
\n-
180
\n-
199 void PreparePatchPoints(REAL const meshPoints[],
\n-
200 PointDescriptor const & meshPointDesc,
\n-
201 REAL patchPoints[],
\n-
202 PointDescriptor const & patchPointDesc) const;
\n-
203
\n-
213 void ComputePatchPoints(REAL patchPoints[],
\n-
214 PointDescriptor const & patchPointDesc) const;
\n-
216
\n-
218
\n-
225
\n-
227 void Evaluate(REAL const uv[2],
\n-
228 REAL const patchPoints[], PointDescriptor const & pointDesc,
\n-
229 REAL P[]) const;
\n-
230
\n-
232 void Evaluate(REAL const uv[2],
\n-
233 REAL const patchPoints[], PointDescriptor const & pointDesc,
\n-
234 REAL P[], REAL Du[], REAL Dv[]) const;
\n-
235
\n-
237 void Evaluate(REAL const uv[2],
\n-
238 REAL const patchPoints[], PointDescriptor const & pointDesc,
\n-
239 REAL P[], REAL Du[], REAL Dv[],
\n-
240 REAL Duu[], REAL Duv[], REAL Dvv[]) const;
\n-
242
\n-
244
\n-
256
\n-
258 int EvaluateStencil(REAL const uv[2], REAL sP[]) const;
\n-
259
\n-
261 int EvaluateStencil(REAL const uv[2], REAL sP[],
\n-
262 REAL sDu[], REAL sDv[]) const;
\n-
263
\n-
265 int EvaluateStencil(REAL const uv[2], REAL sP[],
\n-
266 REAL sDu[], REAL sDv[],
\n-
267 REAL sDuu[], REAL sDuv[], REAL sDvv[]) const;
\n-
268
\n-
270 void ApplyStencil(REAL const stencil[],
\n-
271 REAL const controlPoints[], PointDescriptor const &,
\n-
272 REAL result[]) const;
\n-
273
\n-
275 void ApplyStencilFromMesh(REAL const stencil[],
\n-
276 REAL const meshPoints[], PointDescriptor const &,
\n-
277 REAL result[]) const;
\n-
279
\n-
280private:
\n-
281 // Internal methods for evaluating derivatives, basis weights and
\n-
282 // stencils for regular, irregular and irregular linear patches:
\n-
283 typedef Vtr::ConstArray<int> IndexArray;
\n-
284
\n-
285 void evaluateDerivs(REAL const uv[2], REAL const patchPoints[],
\n-
286 PointDescriptor const &, REAL * derivs[]) const;
\n-
287 void evalRegularDerivs(REAL const uv[2], REAL const patchPoints[],
\n-
288 PointDescriptor const &, REAL * derivs[]) const;
\n-
289 void evalIrregularDerivs(REAL const uv[2], REAL const patchPoints[],
\n-
290 PointDescriptor const &, REAL * derivs[]) const;
\n-
291 void evalMultiLinearDerivs(REAL const uv[2], REAL const patchPoints[],
\n-
292 PointDescriptor const &, REAL * derivs[]) const;
\n-
293
\n-
294 void evalRegularBasis(REAL const uv[2], REAL * wDeriv[]) const;
\n-
295 IndexArray evalIrregularBasis(REAL const uv[2], REAL * wDeriv[]) const;
\n-
296 int evalMultiLinearBasis(REAL const uv[2], REAL * wDeriv[]) const;
\n-
297
\n-
298 int evaluateStencils(REAL const uv[2], REAL * sDeriv[]) const;
\n-
299 int evalRegularStencils(REAL const uv[2], REAL * sDeriv[]) const;
\n-
300 int evalIrregularStencils(REAL const uv[2], REAL * sDeriv[]) const;
\n-
301 int evalMultiLinearStencils(REAL const uv[2], REAL * sDeriv[]) const;
\n-
302
\n-
303 // Internal methods to compute patch points:
\n-
304 void computeLinearPatchPoints(REAL p[], PointDescriptor const &) const;
\n-
305 void computeIrregularPatchPoints(REAL p[], PointDescriptor const &) const;
\n-
306
\n-
307 // Internal methods specific to regular or irregular patches:
\n-
308 unsigned char getRegPatchType() const { return _data.getRegPatchType(); }
\n-
309 unsigned char getRegPatchMask() const { return _data.getRegPatchMask(); }
\n-
310
\n-
311 internal::IrregularPatchType const & getIrregPatch() const;
\n-
312
\n-
313private:
\n-
314 // Access to the set of member variables - provided to the Factory:
\n-
315 friend class SurfaceFactory;
\n-
316
\n-
317 internal::SurfaceData & getSurfaceData() { return _data; }
\n-
318 internal::SurfaceData const & getSurfaceData() const { return _data; }
\n-
319
\n-
320private:
\n-
321 // All member variables encapsulated in a single class:
\n-
322 internal::SurfaceData _data;
\n-
323};
\n-
324
\n-
325
\n-
326//
\n-
327// Simple inline methods composed of other methods:
\n-
328//
\n-
329template <typename REAL>
\n-
330inline void
\n-
331Surface<REAL>::ComputePatchPoints(REAL points[],
\n-
332 PointDescriptor const & pointDesc) const {
\n-
333
\n-
334 if (!IsRegular()) {
\n-
335 if (IsLinear()) {
\n-
336 computeLinearPatchPoints(points, pointDesc);
\n-
337 } else {
\n-
338 computeIrregularPatchPoints(points, pointDesc);
\n-
339 }
\n-
340 }
\n-
341}
\n-
342
\n-
343template <typename REAL>
\n-
344inline void
\n-
345Surface<REAL>::PreparePatchPoints(
\n-
346 REAL const meshPoints[], PointDescriptor const & meshPointDesc,
\n-
347 REAL patchPoints[], PointDescriptor const & patchPointDesc) const {
\n-
348
\n-
349 GatherControlPoints(meshPoints, meshPointDesc, patchPoints, patchPointDesc);
\n-
350 ComputePatchPoints(patchPoints, patchPointDesc);
\n-
351}
\n-
352
\n-
353//
\n-
354// Inline invocations of more general methods for derivative overloads:
\n-
355//
\n-
356template <typename REAL>
\n-
357inline void
\n-
358Surface<REAL>::evaluateDerivs(REAL const uv[2],
\n-
359 REAL const patchPoints[],
\n-
360 PointDescriptor const & pointDesc,
\n-
361 REAL * derivatives[]) const {
\n-
362 if (IsRegular()) {
\n-
363 evalRegularDerivs(uv, patchPoints, pointDesc, derivatives);
\n-
364 } else if (IsLinear()) {
\n-
365 evalMultiLinearDerivs(uv, patchPoints, pointDesc, derivatives);
\n-
366 } else {
\n-
367 evalIrregularDerivs(uv, patchPoints, pointDesc, derivatives);
\n-
368 }
\n-
369}
\n-
370template <typename REAL>
\n-
371inline void
\n-
372Surface<REAL>::Evaluate(REAL const uv[2],
\n-
373 REAL const patchPoints[],
\n-
374 PointDescriptor const & pointDesc,
\n-
375 REAL P[]) const {
\n-
376
\n-
377 REAL * derivatives[6] = { P, 0, 0, 0, 0, 0 };
\n-
378 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);
\n-
379}
\n-
380template <typename REAL>
\n-
381inline void
\n-
382Surface<REAL>::Evaluate(REAL const uv[2],
\n-
383 REAL const patchPoints[],
\n-
384 PointDescriptor const & pointDesc,
\n-
385 REAL P[], REAL Du[], REAL Dv[]) const {
\n-
386
\n-
387 REAL * derivatives[6] = { P, Du, Dv, 0, 0, 0 };
\n-
388 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);
\n-
389}
\n-
390template <typename REAL>
\n-
391inline void
\n-
392Surface<REAL>::Evaluate(REAL const uv[2],
\n-
393 REAL const patchPoints[],
\n-
394 PointDescriptor const & pointDesc,
\n-
395 REAL P[], REAL Du[], REAL Dv[],
\n-
396 REAL Duu[], REAL Duv[], REAL Dvv[]) const {
\n-
397
\n-
398 REAL * derivatives[6] = { P, Du, Dv, Duu, Duv, Dvv };
\n-
399 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);
\n-
400}
\n-
401
\n-
402template <typename REAL>
\n-
403inline int
\n-
404Surface<REAL>::evaluateStencils(REAL const uv[2], REAL * sDeriv[]) const {
\n-
405
\n-
406 if (IsRegular()) {
\n-
407 return evalRegularStencils(uv, sDeriv);
\n-
408 } else if (IsLinear()) {
\n-
409 return evalMultiLinearStencils(uv, sDeriv);
\n-
410 } else {
\n-
411 return evalIrregularStencils(uv, sDeriv);
\n-
412 }
\n-
413}
\n-
414template <typename REAL>
\n-
415inline int
\n-
416Surface<REAL>::EvaluateStencil(REAL const uv[2], REAL sP[]) const {
\n-
417
\n-
418 REAL * derivativeStencils[6] = { sP, 0, 0, 0, 0, 0 };
\n-
419 return evaluateStencils(uv, derivativeStencils);
\n-
420}
\n-
421template <typename REAL>
\n-
422inline int
\n-
423Surface<REAL>::EvaluateStencil(REAL const uv[2],
\n-
424 REAL sP[], REAL sDu[], REAL sDv[]) const {
\n-
425
\n-
426 REAL * derivativeStencils[6] = { sP, sDu, sDv, 0, 0, 0 };
\n-
427 return evaluateStencils(uv, derivativeStencils);
\n-
428}
\n-
429template <typename REAL>
\n-
430inline int
\n-
431Surface<REAL>::EvaluateStencil(REAL const uv[2],
\n-
432 REAL sP[], REAL sDu[], REAL sDv[],
\n-
433 REAL sDuu[], REAL sDuv[], REAL sDvv[]) const {
\n-
434
\n-
435 REAL * derivativeStencils[6] = { sP, sDu, sDv, sDuu, sDuv, sDvv };
\n-
436 return evaluateStencils(uv, derivativeStencils);
\n-
437}
\n-
438
\n-
439} // end namespace Bfr
\n-
440
\n-
441} // end namespace OPENSUBDIV_VERSION
\n-
442using namespace OPENSUBDIV_VERSION;
\n-
443
\n-
444} // end namespace OpenSubdiv
\n-
445
\n-
446#endif /* OPENSUBDIV3_BFR_SURFACE */
\n+
172} // end namespace Bfr
\n+
173
\n+
174} // end namespace OPENSUBDIV_VERSION
\n+
175using namespace OPENSUBDIV_VERSION;
\n+
176
\n+
177} // end namespace OpenSubdiv
\n+
178
\n+
179#endif /* OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H */
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Parameterization
Simple class defining the 2D parameterization of a face.
Definition: parameterization.h:52
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Parameterization::GetFaceSize
int GetFaceSize() const
Returns the size (number of vertices) of the corresponding face.
Definition: parameterization.h:101
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface
Encapsulates the limit surface for a face of a mesh.
Definition: surface.h:59
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetNumControlPoints
int GetNumControlPoints() const
Return the number of control points affecting the Surface.
Definition: surface.h:137
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Evaluate
void Evaluate(REAL const uv[2], REAL const patchPoints[], PointDescriptor const &pointDesc, REAL P[]) const
Evaluation of position.
Definition: surface.h:372
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetFaceSize
int GetFaceSize() const
Return the size of the face.
Definition: surface.h:109
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsRegular
bool IsRegular() const
Return if the Surface is a single regular patch.
Definition: surface.h:112
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::BoundControlPoints
void BoundControlPoints(REAL const controlPoints[], PointDescriptor const &controlPointDesc, REAL minExtent[], REAL maxExtent[]) const
Compute bounds of control points from a local array.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetControlPointIndices
int GetControlPointIndices(Index meshPointIndices[]) const
Identify indices of control points in the mesh.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Index
int Index
Integer type representing a mesh index.
Definition: surface.h:72
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ApplyStencil
void ApplyStencil(REAL const stencil[], REAL const controlPoints[], PointDescriptor const &, REAL result[]) const
Apply a single stencil to control points from a local array.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetParameterization
Parameterization GetParameterization() const
Return the Parameterization.
Definition: surface.h:106
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Surface
Surface()
Default construction produces an invalid instance.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ApplyStencilFromMesh
void ApplyStencilFromMesh(REAL const stencil[], REAL const meshPoints[], PointDescriptor const &, REAL result[]) const
Apply a single stencil to control points from the mesh data.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::operator=
Surface & operator=(Surface const &src)=default
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::BoundControlPointsFromMesh
void BoundControlPointsFromMesh(REAL const meshPoints[], PointDescriptor const &meshPointDesc, REAL minExtent[], REAL maxExtent[]) const
Compute bounds of control points from the mesh data.
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Clear
void Clear()
Clear a previously initialized Surface.
Definition: surface.h:89
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PreparePatchPoints
void PreparePatchPoints(REAL const meshPoints[], PointDescriptor const &meshPointDesc, REAL patchPoints[], PointDescriptor const &patchPointDesc) const
Prepare patch points in a local array for evaluation.
Definition: surface.h:345
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsValid
bool IsValid() const
Return true if successfully initialized.
Definition: surface.h:86
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GatherControlPoints
void GatherControlPoints(REAL_MESH const meshPoints[], PointDescriptor const &meshPointDesc, REAL controlPoints[], PointDescriptor const &controlPointDesc) const
Gather control points in a local array.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ComputePatchPoints
void ComputePatchPoints(REAL patchPoints[], PointDescriptor const &patchPointDesc) const
Compute all patch points following the control points.
Definition: surface.h:331
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsLinear
bool IsLinear() const
Return if the Surface is linear.
Definition: surface.h:115
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetNumPatchPoints
int GetNumPatchPoints() const
Return the number of patch points representing the Surface.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::EvaluateStencil
int EvaluateStencil(REAL const uv[2], REAL sP[]) const
Evaluation of the limit stencil for position.
Definition: surface.h:416
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor
Simple struct defining the size and stride of points in arrays.
Definition: surface.h:63
\n-\n-\n-\n-\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase
Intermediate subclass of SurfaceFactory with Far::TopologyRefiner as the mesh.
Definition: refinerSurfaceFactory.h:58
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::GetNumFVarChannels
int GetNumFVarChannels() const
Return the number of face-varying channels.
Definition: refinerSurfaceFactory.h:85
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::GetMesh
Far::TopologyRefiner const & GetMesh() const
Return the instance of the mesh.
Definition: refinerSurfaceFactory.h:79
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::RefinerSurfaceFactoryBase
RefinerSurfaceFactoryBase(Far::TopologyRefiner const &mesh, Options const &options)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactory
Template for concrete subclasses of RefinerSurfaceFactoryBase.
Definition: refinerSurfaceFactory.h:157
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactory::RefinerSurfaceFactory
RefinerSurfaceFactory(Far::TopologyRefiner const &mesh, Options const &options=Options())
Definition: refinerSurfaceFactory.h:159
\n
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory
Base class providing initialization of a Surface for each face of a mesh.
Definition: surfaceFactory.h:103
\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::setInternalCache
void setInternalCache(SurfaceFactoryCache *cache)
Subclass to identify an internal cache for use by base class.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::Options
Simple set of options assigned to instances of SurfaceFactory.
Definition: surfaceFactory.h:116
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceVertexIncidentFaceFVarValueIndices
virtual int getFaceVertexIncidentFaceFVarValueIndices(Index faceIndex, int faceVertex, FVarID fvarID, Index fvarValueIndices[]) const =0
Gather face-varying indices of incident faces around a face-vertex.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceVertexIncidentFaceVertexIndices
virtual int getFaceVertexIncidentFaceVertexIndices(Index faceIndex, int faceVertex, Index vertexIndices[]) const =0
Gather vertex indices of incident faces around a face-vertex.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceSize
virtual int getFaceSize(Index faceIndex) const =0
Returns the size of a face (number of vertices)
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::populateFaceVertexDescriptor
virtual int populateFaceVertexDescriptor(Index faceIndex, int faceVertex, VertexDescriptor *vertexDescriptor) const =0
Describe the topology of incident faces around a face-vertex.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceNeighborhoodVertexIndicesIfRegular
virtual bool getFaceNeighborhoodVertexIndicesIfRegular(Index faceIndex, Index vertexIndices[]) const
Definition: surfaceFactoryMeshAdapter.h:218
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceNeighborhoodFVarValueIndicesIfRegular
virtual bool getFaceNeighborhoodFVarValueIndicesIfRegular(Index faceIndex, FVarID fvarID, Index fvarValueIndices[]) const
Definition: surfaceFactoryMeshAdapter.h:224
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceFVarValueIndices
virtual int getFaceFVarValueIndices(Index faceIndex, FVarID fvarID, Index fvarValueIndices[]) const =0
Gather the face-varying indices of the face's vertices.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceVertexIndices
virtual int getFaceVertexIndices(Index faceIndex, Index vertexIndices[]) const =0
Gather the indices of the face's vertices.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::isFaceHole
virtual bool isFaceHole(Index faceIndex) const =0
Returns if a face is a hole.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::FVarID
std::intptr_t FVarID
Type used to identify and specify face-varying primvars.
Definition: surfaceFactoryMeshAdapter.h:80
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Bfr::VertexDescriptor
Simple class used by subclasses of SurfaceFactory to describe a vertex.
Definition: vertexDescriptor.h:131
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner
Stores topology data for a specified set of refinement options.
Definition: topologyRefiner.h:51
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-surface.h\n+refinerSurfaceFactory.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2021 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,420 +30,222 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_BFR_SURFACE_H\n- 26#define OPENSUBDIV3_BFR_SURFACE_H\n+ 25#ifndef OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H\n+ 26#define OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H\n 27\n 28#include \"../version.h\"\n 29\n- 30#include \"../bfr/surfaceData.h\"\n- 31#include \"../bfr/parameterization.h\"\n- 32#include \"../vtr/array.h\"\n- 33\n- 34namespace OpenSubdiv {\n- 35namespace OPENSUBDIV_VERSION {\n- 36\n- 37namespace Bfr {\n- 38\n- 58template \n-59class Surface {\n- 60public:\n-63 struct PointDescriptor {\n-64 PointDescriptor() : size(0), stride(0) { }\n-65 PointDescriptor(int n) : size(n), stride(n) { }\n-66 PointDescriptor(int n, int m) : size(n), stride(m) { }\n- 67\n-68 int size, stride;\n- 69 };\n- 70\n-72 typedef int Index;\n+ 30#include \"../bfr/surfaceFactory.h\"\n+ 31#include \"../bfr/surfaceFactoryCache.h\"\n+ 32\n+ 33namespace OpenSubdiv {\n+ 34namespace OPENSUBDIV_VERSION {\n+ 35\n+36namespace Far {\n+ 37 class TopologyRefiner;\n+ 38}\n+ 39\n+ 40namespace Bfr {\n+ 41\n+58class RefinerSurfaceFactoryBase : public SurfaceFactory {\n+ 59public:\n+ 61\n+ 65\n+66 RefinerSurfaceFactoryBase(Far::TopologyRefiner const & mesh,\n+ 67 Options const & options);\n+ 68\n+69 ~RefinerSurfaceFactoryBase() override = default;\n+ 71\n 73\n- 74public:\n- 76\n- 84\n-86 bool IsValid() const { return _data.isValid(); }\n+ 77\n+79 Far::TopologyRefiner const & GetMesh() const { return _mesh; }\n+ 80\n+82 int GetNumFaces() const { return _numFaces; }\n+ 83\n+85 int GetNumFVarChannels() const { return _numFVarChannels; }\n 87\n-89 void Clear() { _data.reinitialize(); }\n- 90\n-92 Surface();\n- 93\n-94 Surface(Surface const & src) = default;\n-95 Surface& operator=(Surface const & src) = default;\n-96 ~Surface() = default;\n- 98\n+ 88protected:\n+ 90 //\n+ 91 // Virtual overrides to satisfy the SurfaceFactoryMeshAdapter interface:\n+ 92 //\n+ 93 bool isFaceHole( Index faceIndex) const override;\n+ 94 int getFaceSize(Index faceIndex) const override;\n+ 95\n+ 96 int getFaceVertexIndices(Index faceIndex,\n+ 97 Index vertexIndices[]) const override;\n+ 98 int getFaceFVarValueIndices(Index faceIndex,\n+ 99 FVarID fvarID, Index fvarValueIndices[]) const override;\n 100\n- 104\n-106 Parameterization GetParameterization() const { return _data.getParam(); }\n- 107\n-109 int GetFaceSize() const { return GetParameterization().GetFaceSize(); }\n+ 101 int populateFaceVertexDescriptor(Index faceIndex, int faceVertex,\n+ 102 VertexDescriptor * vertexDescriptor) const override;\n+ 103\n+ 104 int getFaceVertexIncidentFaceVertexIndices(\n+ 105 Index faceIndex, int faceVertex,\n+ 106 Index vertexIndices[]) const override;\n+ 107 int getFaceVertexIncidentFaceFVarValueIndices(\n+ 108 Index faceIndex, int faceVertex,\n+ 109 FVarID fvarID, Index fvarValueIndices[]) const override;\n 110\n-112 bool IsRegular() const { return _data.isRegular(); }\n- 113\n-115 bool IsLinear() const { return _data.isLinear(); }\n- 117\n- 119\n- 135\n-137 int GetNumControlPoints() const { return _data.getNumCVs(); }\n- 138\n-140 int GetControlPointIndices(Index meshPointIndices[]) const;\n+ 111 // Optional SurfaceFactoryMeshAdapter overrides for regular patches:\n+ 112 bool getFaceNeighborhoodVertexIndicesIfRegular(\n+ 113 Index faceIndex,\n+ 114 Index vertexIndices[]) const override;\n+ 115\n+ 116 bool getFaceNeighborhoodFVarValueIndicesIfRegular(\n+ 117 Index faceIndex,\n+ 118 FVarID fvarID, Index fvarValueIndices[]) const override;\n+ 120\n+ 121private:\n+ 122 //\n+ 123 // Internal supporting methods:\n+ 124 //\n+ 125 int getFaceVaryingChannel(FVarID fvarID) const;\n+ 126\n+ 127 int getFaceVertexPointIndices(Index faceIndex, int faceVertex,\n+ 128 Index indices[], int vtxOrFVarChannel) const;\n+ 129\n+ 130 int getFacePatchPointIndices(Index faceIndex,\n+ 131 Index indices[], int vtxOrFVarChannel) const;\n+ 132\n+ 133private:\n+ 134 // Additional members for the subclass:\n+ 135 Far::TopologyRefiner const & _mesh;\n+ 136\n+ 137 int _numFaces;\n+ 138 int _numFVarChannels;\n+ 139};\n+ 140\n 141\n- 151 template \n-152 void GatherControlPoints(REAL_MESH const meshPoints[],\n- 153 PointDescriptor const & meshPointDesc,\n- 154 REAL controlPoints[],\n- 155 PointDescriptor const & controlPointDesc) const;\n- 156\n-158 void BoundControlPoints(REAL const controlPoints[],\n- 159 PointDescriptor const & controlPointDesc,\n- 160 REAL minExtent[],\n- 161 REAL maxExtent[]) const;\n- 162\n-164 void BoundControlPointsFromMesh(REAL const meshPoints[],\n- 165 PointDescriptor const & meshPointDesc,\n- 166 REAL minExtent[],\n- 167 REAL maxExtent[]) const;\n- 169\n+ 142//\n+ 156template \n+157class RefinerSurfaceFactory : public RefinerSurfaceFactoryBase {\n+ 158public:\n+159 RefinerSurfaceFactory(Far::TopologyRefiner const & mesh,\n+ 160 Options const & options = Options()) :\n+ 161 RefinerSurfaceFactoryBase(mesh, options),\n+ 162 _localCache() {\n+ 163\n+ 164 SurfaceFactory::setInternalCache(&_localCache);\n+ 165 }\n+166 ~RefinerSurfaceFactory() override = default;\n+ 167\n+ 168private:\n+ 169 CACHE_TYPE _localCache;\n+ 170};\n 171\n- 177\n-179 int GetNumPatchPoints() const;\n- 180\n- 199 void PreparePatchPoints(REAL const meshPoints[],\n- 200 PointDescriptor const & meshPointDesc,\n- 201 REAL patchPoints[],\n- 202 PointDescriptor const & patchPointDesc) const;\n- 203\n- 213 void ComputePatchPoints(REAL patchPoints[],\n- 214 PointDescriptor const & patchPointDesc) const;\n- 216\n- 218\n- 225\n- 227 void Evaluate(REAL const uv[2],\n- 228 REAL const patchPoints[], PointDescriptor const & pointDesc,\n- 229 REAL P[]) const;\n- 230\n- 232 void Evaluate(REAL const uv[2],\n- 233 REAL const patchPoints[], PointDescriptor const & pointDesc,\n- 234 REAL P[], REAL Du[], REAL Dv[]) const;\n- 235\n- 237 void Evaluate(REAL const uv[2],\n- 238 REAL const patchPoints[], PointDescriptor const & pointDesc,\n- 239 REAL P[], REAL Du[], REAL Dv[],\n- 240 REAL Duu[], REAL Duv[], REAL Dvv[]) const;\n- 242\n- 244\n- 256\n- 258 int EvaluateStencil(REAL const uv[2], REAL sP[]) const;\n- 259\n- 261 int EvaluateStencil(REAL const uv[2], REAL sP[],\n- 262 REAL sDu[], REAL sDv[]) const;\n- 263\n- 265 int EvaluateStencil(REAL const uv[2], REAL sP[],\n- 266 REAL sDu[], REAL sDv[],\n- 267 REAL sDuu[], REAL sDuv[], REAL sDvv[]) const;\n- 268\n-270 void ApplyStencil(REAL const stencil[],\n- 271 REAL const controlPoints[], PointDescriptor const &,\n- 272 REAL result[]) const;\n- 273\n-275 void ApplyStencilFromMesh(REAL const stencil[],\n- 276 REAL const meshPoints[], PointDescriptor const &,\n- 277 REAL result[]) const;\n- 279\n- 280private:\n- 281 // Internal methods for evaluating derivatives, basis weights and\n- 282 // stencils for regular, irregular and irregular linear patches:\n- 283 typedef Vtr::ConstArray IndexArray;\n- 284\n- 285 void evaluateDerivs(REAL const uv[2], REAL const patchPoints[],\n- 286 PointDescriptor const &, REAL * derivs[]) const;\n- 287 void evalRegularDerivs(REAL const uv[2], REAL const patchPoints[],\n- 288 PointDescriptor const &, REAL * derivs[]) const;\n- 289 void evalIrregularDerivs(REAL const uv[2], REAL const patchPoints[],\n- 290 PointDescriptor const &, REAL * derivs[]) const;\n- 291 void evalMultiLinearDerivs(REAL const uv[2], REAL const patchPoints[],\n- 292 PointDescriptor const &, REAL * derivs[]) const;\n- 293\n- 294 void evalRegularBasis(REAL const uv[2], REAL * wDeriv[]) const;\n- 295 IndexArray evalIrregularBasis(REAL const uv[2], REAL * wDeriv[]) const;\n- 296 int evalMultiLinearBasis(REAL const uv[2], REAL * wDeriv[]) const;\n- 297\n- 298 int evaluateStencils(REAL const uv[2], REAL * sDeriv[]) const;\n- 299 int evalRegularStencils(REAL const uv[2], REAL * sDeriv[]) const;\n- 300 int evalIrregularStencils(REAL const uv[2], REAL * sDeriv[]) const;\n- 301 int evalMultiLinearStencils(REAL const uv[2], REAL * sDeriv[]) const;\n- 302\n- 303 // Internal methods to compute patch points:\n- 304 void computeLinearPatchPoints(REAL p[], PointDescriptor const &) const;\n- 305 void computeIrregularPatchPoints(REAL p[], PointDescriptor const &) const;\n- 306\n- 307 // Internal methods specific to regular or irregular patches:\n- 308 unsigned char getRegPatchType() const { return _data.getRegPatchType(); }\n- 309 unsigned char getRegPatchMask() const { return _data.getRegPatchMask(); }\n- 310\n- 311 internal::IrregularPatchType const & getIrregPatch() const;\n- 312\n- 313private:\n- 314 // Access to the set of member variables - provided to the Factory:\n-315 friend class SurfaceFactory;\n- 316\n- 317 internal::SurfaceData & getSurfaceData() { return _data; }\n- 318 internal::SurfaceData const & getSurfaceData() const { return _data; }\n- 319\n- 320private:\n- 321 // All member variables encapsulated in a single class:\n- 322 internal::SurfaceData _data;\n- 323};\n- 324\n- 325\n- 326//\n- 327// Simple inline methods composed of other methods:\n- 328//\n- 329template \n- 330inline void\n-331Surface::ComputePatchPoints(REAL points[],\n- 332 PointDescriptor const & pointDesc) const {\n- 333\n- 334 if (!IsRegular()) {\n- 335 if (IsLinear()) {\n- 336 computeLinearPatchPoints(points, pointDesc);\n- 337 } else {\n- 338 computeIrregularPatchPoints(points, pointDesc);\n- 339 }\n- 340 }\n- 341}\n- 342\n- 343template \n- 344inline void\n-345Surface::PreparePatchPoints(\n- 346 REAL const meshPoints[], PointDescriptor const & meshPointDesc,\n- 347 REAL patchPoints[], PointDescriptor const & patchPointDesc) const {\n- 348\n- 349 GatherControlPoints(meshPoints, meshPointDesc, patchPoints,\n-patchPointDesc);\n- 350 ComputePatchPoints(patchPoints, patchPointDesc);\n- 351}\n- 352\n- 353//\n- 354// Inline invocations of more general methods for derivative overloads:\n- 355//\n- 356template \n- 357inline void\n- 358Surface::evaluateDerivs(REAL const uv[2],\n- 359 REAL const patchPoints[],\n- 360 PointDescriptor const & pointDesc,\n- 361 REAL * derivatives[]) const {\n- 362 if (IsRegular()) {\n- 363 evalRegularDerivs(uv, patchPoints, pointDesc, derivatives);\n- 364 } else if (IsLinear()) {\n- 365 evalMultiLinearDerivs(uv, patchPoints, pointDesc, derivatives);\n- 366 } else {\n- 367 evalIrregularDerivs(uv, patchPoints, pointDesc, derivatives);\n- 368 }\n- 369}\n- 370template \n- 371inline void\n-372Surface::Evaluate(REAL const uv[2],\n- 373 REAL const patchPoints[],\n- 374 PointDescriptor const & pointDesc,\n- 375 REAL P[]) const {\n- 376\n- 377 REAL * derivatives[6] = { P, 0, 0, 0, 0, 0 };\n- 378 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);\n- 379}\n- 380template \n- 381inline void\n-382Surface::Evaluate(REAL const uv[2],\n- 383 REAL const patchPoints[],\n- 384 PointDescriptor const & pointDesc,\n- 385 REAL P[], REAL Du[], REAL Dv[]) const {\n- 386\n- 387 REAL * derivatives[6] = { P, Du, Dv, 0, 0, 0 };\n- 388 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);\n- 389}\n- 390template \n- 391inline void\n-392Surface::Evaluate(REAL const uv[2],\n- 393 REAL const patchPoints[],\n- 394 PointDescriptor const & pointDesc,\n- 395 REAL P[], REAL Du[], REAL Dv[],\n- 396 REAL Duu[], REAL Duv[], REAL Dvv[]) const {\n- 397\n- 398 REAL * derivatives[6] = { P, Du, Dv, Duu, Duv, Dvv };\n- 399 evaluateDerivs(uv, patchPoints, pointDesc, derivatives);\n- 400}\n- 401\n- 402template \n- 403inline int\n- 404Surface::evaluateStencils(REAL const uv[2], REAL * sDeriv[]) const {\n- 405\n- 406 if (IsRegular()) {\n- 407 return evalRegularStencils(uv, sDeriv);\n- 408 } else if (IsLinear()) {\n- 409 return evalMultiLinearStencils(uv, sDeriv);\n- 410 } else {\n- 411 return evalIrregularStencils(uv, sDeriv);\n- 412 }\n- 413}\n- 414template \n- 415inline int\n-416Surface::EvaluateStencil(REAL const uv[2], REAL sP[]) const {\n- 417\n- 418 REAL * derivativeStencils[6] = { sP, 0, 0, 0, 0, 0 };\n- 419 return evaluateStencils(uv, derivativeStencils);\n- 420}\n- 421template \n- 422inline int\n-423Surface::EvaluateStencil(REAL const uv[2],\n- 424 REAL sP[], REAL sDu[], REAL sDv[]) const {\n- 425\n- 426 REAL * derivativeStencils[6] = { sP, sDu, sDv, 0, 0, 0 };\n- 427 return evaluateStencils(uv, derivativeStencils);\n- 428}\n- 429template \n- 430inline int\n-431Surface::EvaluateStencil(REAL const uv[2],\n- 432 REAL sP[], REAL sDu[], REAL sDv[],\n- 433 REAL sDuu[], REAL sDuv[], REAL sDvv[]) const {\n- 434\n- 435 REAL * derivativeStencils[6] = { sP, sDu, sDv, sDuu, sDuv, sDvv };\n- 436 return evaluateStencils(uv, derivativeStencils);\n- 437}\n- 438\n- 439} // end namespace Bfr\n- 440\n- 441} // end namespace OPENSUBDIV_VERSION\n- 442using namespace OPENSUBDIV_VERSION;\n- 443\n- 444} // end namespace OpenSubdiv\n- 445\n- 446#endif /* OPENSUBDIV3_BFR_SURFACE */\n+ 172} // end namespace Bfr\n+ 173\n+ 174} // end namespace OPENSUBDIV_VERSION\n+ 175using namespace OPENSUBDIV_VERSION;\n+ 176\n+ 177} // end namespace OpenSubdiv\n+ 178\n+ 179#endif /* OPENSUBDIV3_BFR_REFINER_SURFACE_FACTORY_H */\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Parameterization\n-Simple class defining the 2D parameterization of a face.\n-Definition: parameterization.h:52\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Parameterization::GetFaceSize\n-int GetFaceSize() const\n-Returns the size (number of vertices) of the corresponding face.\n-Definition: parameterization.h:101\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface\n-Encapsulates the limit surface for a face of a mesh.\n-Definition: surface.h:59\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetNumControlPoints\n-int GetNumControlPoints() const\n-Return the number of control points affecting the Surface.\n-Definition: surface.h:137\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Evaluate\n-void Evaluate(REAL const uv[2], REAL const patchPoints[], PointDescriptor const\n-&pointDesc, REAL P[]) const\n-Evaluation of position.\n-Definition: surface.h:372\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetFaceSize\n-int GetFaceSize() const\n-Return the size of the face.\n-Definition: surface.h:109\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsRegular\n-bool IsRegular() const\n-Return if the Surface is a single regular patch.\n-Definition: surface.h:112\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::BoundControlPoints\n-void BoundControlPoints(REAL const controlPoints[], PointDescriptor const\n-&controlPointDesc, REAL minExtent[], REAL maxExtent[]) const\n-Compute bounds of control points from a local array.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetControlPointIndices\n-int GetControlPointIndices(Index meshPointIndices[]) const\n-Identify indices of control points in the mesh.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Index\n-int Index\n-Integer type representing a mesh index.\n-Definition: surface.h:72\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ApplyStencil\n-void ApplyStencil(REAL const stencil[], REAL const controlPoints[],\n-PointDescriptor const &, REAL result[]) const\n-Apply a single stencil to control points from a local array.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetParameterization\n-Parameterization GetParameterization() const\n-Return the Parameterization.\n-Definition: surface.h:106\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Surface\n-Surface()\n-Default construction produces an invalid instance.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ApplyStencilFromMesh\n-void ApplyStencilFromMesh(REAL const stencil[], REAL const meshPoints[],\n-PointDescriptor const &, REAL result[]) const\n-Apply a single stencil to control points from the mesh data.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::operator=\n-Surface & operator=(Surface const &src)=default\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::BoundControlPointsFromMesh\n-void BoundControlPointsFromMesh(REAL const meshPoints[], PointDescriptor const\n-&meshPointDesc, REAL minExtent[], REAL maxExtent[]) const\n-Compute bounds of control points from the mesh data.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::~Surface\n-~Surface()=default\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Clear\n-void Clear()\n-Clear a previously initialized Surface.\n-Definition: surface.h:89\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Surface\n-Surface(Surface const &src)=default\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PreparePatchPoints\n-void PreparePatchPoints(REAL const meshPoints[], PointDescriptor const\n-&meshPointDesc, REAL patchPoints[], PointDescriptor const &patchPointDesc)\n-const\n-Prepare patch points in a local array for evaluation.\n-Definition: surface.h:345\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsValid\n-bool IsValid() const\n-Return true if successfully initialized.\n-Definition: surface.h:86\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GatherControlPoints\n-void GatherControlPoints(REAL_MESH const meshPoints[], PointDescriptor const\n-&meshPointDesc, REAL controlPoints[], PointDescriptor const &controlPointDesc)\n-const\n-Gather control points in a local array.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::ComputePatchPoints\n-void ComputePatchPoints(REAL patchPoints[], PointDescriptor const\n-&patchPointDesc) const\n-Compute all patch points following the control points.\n-Definition: surface.h:331\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::IsLinear\n-bool IsLinear() const\n-Return if the Surface is linear.\n-Definition: surface.h:115\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::GetNumPatchPoints\n-int GetNumPatchPoints() const\n-Return the number of patch points representing the Surface.\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::EvaluateStencil\n-int EvaluateStencil(REAL const uv[2], REAL sP[]) const\n-Evaluation of the limit stencil for position.\n-Definition: surface.h:416\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor\n-Simple struct defining the size and stride of points in arrays.\n-Definition: surface.h:63\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor::stride\n-int stride\n-Definition: surface.h:68\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor::size\n-int size\n-Definition: surface.h:68\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor::PointDescriptor\n-PointDescriptor(int n)\n-Definition: surface.h:65\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor::PointDescriptor\n-PointDescriptor(int n, int m)\n-Definition: surface.h:66\n-OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::PointDescriptor::PointDescriptor\n-PointDescriptor()\n-Definition: surface.h:64\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase\n+Intermediate subclass of SurfaceFactory with Far::TopologyRefiner as the mesh.\n+Definition: refinerSurfaceFactory.h:58\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::GetNumFaces\n+int GetNumFaces() const\n+Return the number of faces.\n+Definition: refinerSurfaceFactory.h:82\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::\n+GetNumFVarChannels\n+int GetNumFVarChannels() const\n+Return the number of face-varying channels.\n+Definition: refinerSurfaceFactory.h:85\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::GetMesh\n+Far::TopologyRefiner const & GetMesh() const\n+Return the instance of the mesh.\n+Definition: refinerSurfaceFactory.h:79\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::\n+~RefinerSurfaceFactoryBase\n+~RefinerSurfaceFactoryBase() override=default\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::\n+RefinerSurfaceFactoryBase\n+RefinerSurfaceFactoryBase(Far::TopologyRefiner const &mesh, Options const\n+&options)\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactory\n+Template for concrete subclasses of RefinerSurfaceFactoryBase.\n+Definition: refinerSurfaceFactory.h:157\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactory::\n+~RefinerSurfaceFactory\n+~RefinerSurfaceFactory() override=default\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactory::\n+RefinerSurfaceFactory\n+RefinerSurfaceFactory(Far::TopologyRefiner const &mesh, Options const\n+&options=Options())\n+Definition: refinerSurfaceFactory.h:159\n OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory\n Base class providing initialization of a Surface for each face of a mesh.\n Definition: surfaceFactory.h:103\n-OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstArray\n-Definition: array.h:53\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::setInternalCache\n+void setInternalCache(SurfaceFactoryCache *cache)\n+Subclass to identify an internal cache for use by base class.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::Options\n+Simple set of options assigned to instances of SurfaceFactory.\n+Definition: surfaceFactory.h:116\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n+getFaceVertexIncidentFaceFVarValueIndices\n+virtual int getFaceVertexIncidentFaceFVarValueIndices(Index faceIndex, int\n+faceVertex, FVarID fvarID, Index fvarValueIndices[]) const =0\n+Gather face-varying indices of incident faces around a face-vertex.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n+getFaceVertexIncidentFaceVertexIndices\n+virtual int getFaceVertexIncidentFaceVertexIndices(Index faceIndex, int\n+faceVertex, Index vertexIndices[]) const =0\n+Gather vertex indices of incident faces around a face-vertex.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::getFaceSize\n+virtual int getFaceSize(Index faceIndex) const =0\n+Returns the size of a face (number of vertices)\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::Index\n+int Index\n+Integer type representing a mesh index.\n+Definition: surfaceFactoryMeshAdapter.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n+populateFaceVertexDescriptor\n+virtual int populateFaceVertexDescriptor(Index faceIndex, int faceVertex,\n+VertexDescriptor *vertexDescriptor) const =0\n+Describe the topology of incident faces around a face-vertex.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n+getFaceNeighborhoodVertexIndicesIfRegular\n+virtual bool getFaceNeighborhoodVertexIndicesIfRegular(Index faceIndex, Index\n+vertexIndices[]) const\n+Definition: surfaceFactoryMeshAdapter.h:218\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n+getFaceNeighborhoodFVarValueIndicesIfRegular\n+virtual bool getFaceNeighborhoodFVarValueIndicesIfRegular(Index faceIndex,\n+FVarID fvarID, Index fvarValueIndices[]) const\n+Definition: surfaceFactoryMeshAdapter.h:224\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n+getFaceFVarValueIndices\n+virtual int getFaceFVarValueIndices(Index faceIndex, FVarID fvarID, Index\n+fvarValueIndices[]) const =0\n+Gather the face-varying indices of the face's vertices.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::\n+getFaceVertexIndices\n+virtual int getFaceVertexIndices(Index faceIndex, Index vertexIndices[]) const\n+=0\n+Gather the indices of the face's vertices.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::isFaceHole\n+virtual bool isFaceHole(Index faceIndex) const =0\n+Returns if a face is a hole.\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactoryMeshAdapter::FVarID\n+std::intptr_t FVarID\n+Type used to identify and specify face-varying primvars.\n+Definition: surfaceFactoryMeshAdapter.h:80\n+OpenSubdiv::OPENSUBDIV_VERSION::Bfr::VertexDescriptor\n+Simple class used by subclasses of SurfaceFactory to describe a vertex.\n+Definition: vertexDescriptor.h:131\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner\n+Stores topology data for a specified set of refinement options.\n+Definition: topologyRefiner.h:51\n * opensubdiv\n * bfr\n- * surface.h\n+ * refinerSurfaceFactory.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00707.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00707.html", "unified_diff": "@@ -90,15 +90,15 @@\n
topologyRefiner.h File Reference
\n \n
\n
#include "../version.h"
\n #include "../sdc/types.h"
\n #include "../sdc/options.h"
\n #include "../far/types.h"
\n-#include "../far/topologyLevel.h"
\n+#include "../far/topologyLevel.h"
\n #include <vector>
\n
\n

Go to the source code of this file.

\n \n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00707_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00707_source.html", "unified_diff": "@@ -325,17 +325,17 @@\n
306using namespace OPENSUBDIV_VERSION;
\n
307} // end namespace OpenSubdiv
\n
308
\n
309#endif /* OPENSUBDIV3_FAR_TOPOLOGY_REFINER_H */
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstIndexArray
Vtr::ConstIndexArray ConstIndexArray
Definition: types.h:47
\n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SchemeType
SchemeType
Enumerated type for all subdivision schemes supported by OpenSubdiv.
Definition: types.h:37
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal
Applies refinement operations to generic primvar data.
Definition: primvarRefiner.h:56
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices
Object used to compute and query ptex face indices.
Definition: ptexIndices.h:46
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel
An interface for accessing data in a specific level of a refined topology hierarchy.
Definition: topologyLevel.h:49
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal
Applies refinement operations to generic primvar data.
Definition: primvarRefiner.h:56
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices
Object used to compute and query ptex face indices.
Definition: ptexIndices.h:46
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel
An interface for accessing data in a specific level of a refined topology hierarchy.
Definition: topologyLevel.h:49
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory
Factory for constructing TopologyRefiners from specific mesh classes.
Definition: topologyRefinerFactory.h:78
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner
Stores topology data for a specified set of refinement options.
Definition: topologyRefiner.h:51
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::TopologyRefiner
TopologyRefiner(TopologyRefiner const &source)
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::GetMaxValence
int GetMaxValence() const
Returns the maximum vertex valence in all levels.
Definition: topologyRefiner.h:77
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::getLevel
Vtr::internal::Level const & getLevel(int l) const
Definition: topologyRefiner.h:237
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::HasHoles
bool HasHoles() const
Returns true if faces have been tagged as holes.
Definition: topologyRefiner.h:80
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::GetNumLevels
int GetNumLevels() const
Returns the number of refinement levels.
Definition: topologyRefiner.h:71
\n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00710.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00710.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/stencilTableFactory.h File Reference\n+OpenSubdiv: opensubdiv/far/topologyDescriptor.h File Reference\n \n \n \n \n \n \n \n@@ -83,43 +83,31 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
stencilTableFactory.h File Reference
\n+
topologyDescriptor.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../far/patchTable.h"
\n-#include <vector>
\n+#include "../far/topologyRefiner.h"
\n+#include "../far/topologyRefinerFactory.h"
\n+#include "../far/error.h"
\n+#include <cassert>
\n
\n

Go to the source code of this file.

\n

\n Classes

class  TopologyRefiner
\n \n-\n-\n+\n+\n \n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n+\n \n

\n Classes

class  StencilTableFactoryReal< REAL >
 A specialized factory for StencilTable. More...
struct  TopologyDescriptor
 A simple reference to raw topology data for use with TopologyRefinerFactory. More...
 
struct  StencilTableFactoryReal< REAL >::Options
 
class  LimitStencilTableFactoryReal< REAL >
 A specialized factory for LimitStencilTable. More...
 
struct  LimitStencilTableFactoryReal< REAL >::Options
 
struct  LimitStencilTableFactoryReal< REAL >::LocationArray
 Descriptor for limit surface locations. More...
 
class  StencilTableFactory
 Stencil table factory class wrapping the template for compatibility. More...
 
class  LimitStencilTableFactory
 Stencil table factory class wrapping the template for compatibility. More...
struct  TopologyDescriptor::FVarChannel
 
\n \n \n \n \n@@ -128,13 +116,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,45 +5,32 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-stencilTableFactory.h File Reference\n+topologyDescriptor.h File Reference\n #include \"../version.h\"\n-#include \"../far/patchTable.h\"\n-#include \n+#include \"../far/topologyRefiner.h\"\n+#include \"../far/topologyRefinerFactory.h\"\n+#include \"../far/error.h\"\n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n- class \u00a0StencilTableFactoryReal<_REAL_>\n-\u00a0 A specialized factory for StencilTable. More...\n+struct \u00a0TopologyDescriptor\n+\u00a0 A simple reference to raw topology data for use with\n+ TopologyRefinerFactory. More...\n \u00a0\n-struct \u00a0StencilTableFactoryReal<_REAL_>::Options\n-\u00a0\n- class \u00a0LimitStencilTableFactoryReal<_REAL_>\n-\u00a0 A specialized factory for LimitStencilTable. More...\n-\u00a0\n-struct \u00a0LimitStencilTableFactoryReal<_REAL_>::Options\n-\u00a0\n-struct \u00a0LimitStencilTableFactoryReal<_REAL_>::LocationArray\n-\u00a0 Descriptor for limit surface locations. More...\n-\u00a0\n- class \u00a0StencilTableFactory\n-\u00a0 Stencil table factory class wrapping the template for compatibility.\n- More...\n-\u00a0\n- class \u00a0LimitStencilTableFactory\n-\u00a0 Stencil table factory class wrapping the template for compatibility.\n- More...\n+struct \u00a0TopologyDescriptor::FVarChannel\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n * opensubdiv\n * far\n- * stencilTableFactory.h\n+ * topologyDescriptor.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00710.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00710.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,9 +1,4 @@\n var a00710 = [\n- [\"StencilTableFactoryReal< REAL >\", \"a01037.html\", \"a01037\"],\n- [\"StencilTableFactoryReal< REAL >::Options\", \"a01077.html\", \"a01077\"],\n- [\"LimitStencilTableFactoryReal< REAL >\", \"a01041.html\", \"a01041\"],\n- [\"LimitStencilTableFactoryReal< REAL >::Options\", \"a01081.html\", \"a01081\"],\n- [\"LimitStencilTableFactoryReal< REAL >::LocationArray\", \"a01085.html\", \"a01085\"],\n- [\"StencilTableFactory\", \"a01089.html\", null],\n- [\"LimitStencilTableFactory\", \"a01093.html\", null]\n+ [\"TopologyDescriptor\", \"a01097.html\", \"a01097\"],\n+ [\"TopologyDescriptor::FVarChannel\", \"a01101.html\", \"a01101\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00710_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00710_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/stencilTableFactory.h Source File\n+OpenSubdiv: opensubdiv/far/topologyDescriptor.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
stencilTableFactory.h
\n+
topologyDescriptor.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2013 Pixar
\n+
2// Copyright 2014 DreamWorks Animation LLC.
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -106,322 +106,142 @@\n
17//
\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n-
24
\n-
25#ifndef OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H
\n-
26#define OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H
\n-
27
\n-
28#include "../version.h"
\n-
29
\n-
30#include "../far/patchTable.h"
\n-
31
\n-
32#include <vector>
\n-
33
\n-
34namespace OpenSubdiv {
\n-
35namespace OPENSUBDIV_VERSION {
\n-
36
\n-
37namespace Far {
\n-
38
\n-
39class TopologyRefiner;
\n-
40
\n-
41template <typename REAL> class StencilReal;
\n-
42template <typename REAL> class StencilTableReal;
\n-
43
\n-
44template <typename REAL> class LimitStencilReal;
\n-
45template <typename REAL> class LimitStencilTableReal;
\n-
46
\n-
47
\n-
50template <typename REAL>
\n-
51class StencilTableFactoryReal {
\n-
52
\n-
53public:
\n-
54
\n-
55 enum Mode {
\n-
56 INTERPOLATE_VERTEX=0,
\n-
57 INTERPOLATE_VARYING,
\n-
58 INTERPOLATE_FACE_VARYING
\n-
59 };
\n+
24#ifndef OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H
\n+
25#define OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H
\n+
26
\n+
27#include "../version.h"
\n+
28
\n+
29#include "../far/topologyRefiner.h"
\n+
30#include "../far/topologyRefinerFactory.h"
\n+
31#include "../far/error.h"
\n+
32
\n+
33#include <cassert>
\n+
34
\n+
35namespace OpenSubdiv {
\n+
36namespace OPENSUBDIV_VERSION {
\n+
37
\n+
38namespace Far {
\n+
39
\n+
49struct TopologyDescriptor {
\n+
50
\n+
51 int numVertices,
\n+
52 numFaces;
\n+
53
\n+
54 int const * numVertsPerFace;
\n+
55 Index const * vertIndicesPerFace;
\n+
56
\n+
57 int numCreases;
\n+
58 Index const * creaseVertexIndexPairs;
\n+
59 float const * creaseWeights;
\n
60
\n-
61 struct Options {
\n-
62
\n-
63 Options() : interpolationMode(INTERPOLATE_VERTEX),
\n-
64 generateOffsets(false),
\n-
65 generateControlVerts(false),
\n-
66 generateIntermediateLevels(true),
\n-
67 factorizeIntermediateLevels(true),
\n-
68 maxLevel(10),
\n-
69 fvarChannel(0) { }
\n-
70
\n-
71 unsigned int interpolationMode : 2,
\n-
72 generateOffsets : 1,
\n-
73 generateControlVerts : 1,
\n-
74 generateIntermediateLevels : 1,
\n-
75 factorizeIntermediateLevels : 1,
\n-
78 maxLevel : 4;
\n-
79 unsigned int fvarChannel;
\n-
81 };
\n-
82
\n-
94 static StencilTableReal<REAL> const * Create(
\n-
95 TopologyRefiner const & refiner, Options options = Options());
\n-
96
\n-
97
\n-
110 static StencilTableReal<REAL> const * Create(
\n-
111 int numTables, StencilTableReal<REAL> const ** tables);
\n-
112
\n+
61 int numCorners;
\n+
62 Index const * cornerVertexIndices;
\n+
63 float const * cornerWeights;
\n+
64
\n+
65 int numHoles;
\n+
66 Index const * holeIndices;
\n+
67
\n+
68 bool isLeftHanded;
\n+
69
\n+
70 // Face-varying data channel -- value indices correspond to vertex indices,
\n+
71 // i.e. one for every vertex of every face:
\n+
72 //
\n+
73 struct FVarChannel {
\n+
74
\n+
75 int numValues;
\n+
76 Index const * valueIndices;
\n+
77
\n+
78 FVarChannel() : numValues(0), valueIndices(0) { }
\n+
79 };
\n+
80
\n+
81 int numFVarChannels;
\n+
82 FVarChannel const * fvarChannels;
\n+
83
\n+
84 TopologyDescriptor();
\n+
85};
\n+
86
\n+
87
\n+
88//
\n+
89// Forward declarations of required TopologyRefinerFactory<TopologyDescriptor>
\n+
90// specializations (defined internally):
\n+
91//
\n+
92// @cond EXCLUDE_DOXYGEN
\n+
93
\n+
94template <>
\n+
95bool
\n+
96TopologyRefinerFactory<TopologyDescriptor>::resizeComponentTopology(
\n+
97 TopologyRefiner & refiner, TopologyDescriptor const & desc);
\n+
98
\n+
99template <>
\n+
100bool
\n+
101TopologyRefinerFactory<TopologyDescriptor>::assignComponentTopology(
\n+
102 TopologyRefiner & refiner, TopologyDescriptor const & desc);
\n+
103
\n+
104template <>
\n+
105bool
\n+
106TopologyRefinerFactory<TopologyDescriptor>::assignComponentTags(
\n+
107 TopologyRefiner & refiner, TopologyDescriptor const & desc);
\n+
108
\n+
109template <>
\n+
110bool
\n+
111TopologyRefinerFactory<TopologyDescriptor>::assignFaceVaryingTopology(
\n+
112 TopologyRefiner & refiner, TopologyDescriptor const & desc);
\n
113
\n-
128 static StencilTableReal<REAL> const * AppendLocalPointStencilTable(
\n-
129 TopologyRefiner const &refiner,
\n-
130 StencilTableReal<REAL> const *baseStencilTable,
\n-
131 StencilTableReal<REAL> const *localPointStencilTable,
\n-
132 bool factorize = true);
\n-
133
\n-
148 static StencilTableReal<REAL> const * AppendLocalPointStencilTableVarying(
\n-
149 TopologyRefiner const &refiner,
\n-
150 StencilTableReal<REAL> const *baseStencilTable,
\n-
151 StencilTableReal<REAL> const *localPointStencilTable,
\n-
152 bool factorize = true) {
\n-
153 return AppendLocalPointStencilTable(
\n-
154 refiner, baseStencilTable, localPointStencilTable, factorize);
\n-
155 }
\n-
156
\n-
174 static StencilTableReal<REAL> const * AppendLocalPointStencilTableFaceVarying(
\n-
175 TopologyRefiner const &refiner,
\n-
176 StencilTableReal<REAL> const *baseStencilTable,
\n-
177 StencilTableReal<REAL> const *localPointStencilTable,
\n-
178 int channel = 0,
\n-
179 bool factorize = true);
\n-
180
\n-
181private:
\n-
182
\n-
183 // Generate stencils for the coarse control-vertices (single weight = 1.0f)
\n-
184 static void generateControlVertStencils(
\n-
185 int numControlVerts,
\n-
186 StencilReal<REAL> & dst);
\n-
187
\n-
188 // Internal method to splice local point stencils
\n-
189 static StencilTableReal<REAL> const * appendLocalPointStencilTable(
\n-
190 TopologyRefiner const &refiner,
\n-
191 StencilTableReal<REAL> const * baseStencilTable,
\n-
192 StencilTableReal<REAL> const * localPointStencilTable,
\n-
193 int channel,
\n-
194 bool factorize);
\n-
195};
\n-
196
\n-
208template <typename REAL>
\n-
209class LimitStencilTableFactoryReal {
\n-
210
\n-
211public:
\n-
212
\n-
213 enum Mode {
\n-
214 INTERPOLATE_VERTEX=0,
\n-
215 INTERPOLATE_VARYING,
\n-
216 INTERPOLATE_FACE_VARYING
\n-
217 };
\n-
218
\n-
219 struct Options {
\n-
220
\n-
221 Options() : interpolationMode(INTERPOLATE_VERTEX),
\n-
222 generate1stDerivatives(true),
\n-
223 generate2ndDerivatives(false),
\n-
224 fvarChannel(0) { }
\n-
225
\n-
226 unsigned int interpolationMode : 2,
\n-
227 generate1stDerivatives : 1,
\n-
228 generate2ndDerivatives : 1;
\n-
229 unsigned int fvarChannel;
\n-
230 };
\n-
231
\n-
233 struct LocationArray {
\n-
234
\n-
235 LocationArray() : ptexIdx(-1), numLocations(0), s(0), t(0) { }
\n-
236
\n-
237 int ptexIdx,
\n-
238 numLocations;
\n-
239
\n-
240 REAL const * s,
\n-
241 * t;
\n-
242 };
\n-
243
\n-
244 typedef std::vector<LocationArray> LocationArrayVec;
\n-
245
\n-
269 static LimitStencilTableReal<REAL> const * Create(
\n-
270 TopologyRefiner const & refiner,
\n-
271 LocationArrayVec const & locationArrays,
\n-
272 StencilTableReal<REAL> const * cvStencils = 0,
\n-
273 PatchTable const * patchTable = 0,
\n-
274 Options options = Options());
\n-
275
\n-
276};
\n-
277
\n-
278
\n-
279//
\n-
280// Public wrapper classes for the templates
\n-
281//
\n-
282class Stencil;
\n-
283class StencilTable;
\n-
284
\n-
287class StencilTableFactory : public StencilTableFactoryReal<float> {
\n-
288private:
\n-
289 typedef StencilTableFactoryReal<float> BaseFactory;
\n-
290 typedef StencilTableReal<float> BaseTable;
\n-
291
\n-
292public:
\n-
293 static StencilTable const * Create(
\n-
294 TopologyRefiner const & refiner, Options options = Options()) {
\n-
295
\n-
296 return static_cast<StencilTable const *>(
\n-
297 BaseFactory::Create(refiner, options));
\n-
298 }
\n-
299
\n-
300 static StencilTable const * Create(
\n-
301 int numTables, StencilTable const ** tables) {
\n-
302
\n-
303 return static_cast<StencilTable const *>(
\n-
304 BaseFactory::Create(numTables,
\n-
305 reinterpret_cast<BaseTable const **>(tables)));
\n-
306 }
\n-
307
\n-
308 static StencilTable const * AppendLocalPointStencilTable(
\n-
309 TopologyRefiner const &refiner,
\n-
310 StencilTable const *baseStencilTable,
\n-
311 StencilTable const *localPointStencilTable,
\n-
312 bool factorize = true) {
\n-
313
\n-
314 return static_cast<StencilTable const *>(
\n-
315 BaseFactory::AppendLocalPointStencilTable(refiner,
\n-
316 static_cast<BaseTable const *>(baseStencilTable),
\n-
317 static_cast<BaseTable const *>(localPointStencilTable),
\n-
318 factorize));
\n-
319 }
\n-
320
\n-
321 static StencilTable const * AppendLocalPointStencilTableVarying(
\n-
322 TopologyRefiner const &refiner,
\n-
323 StencilTable const *baseStencilTable,
\n-
324 StencilTable const *localPointStencilTable,
\n-
325 bool factorize = true) {
\n-
326
\n-
327 return static_cast<StencilTable const *>(
\n-
328 BaseFactory::AppendLocalPointStencilTableVarying(refiner,
\n-
329 static_cast<BaseTable const *>(baseStencilTable),
\n-
330 static_cast<BaseTable const *>(localPointStencilTable),
\n-
331 factorize));
\n-
332 }
\n-
333
\n-
334 static StencilTable const * AppendLocalPointStencilTableFaceVarying(
\n-
335 TopologyRefiner const &refiner,
\n-
336 StencilTable const *baseStencilTable,
\n-
337 StencilTable const *localPointStencilTable,
\n-
338 int channel = 0,
\n-
339 bool factorize = true) {
\n-
340
\n-
341 return static_cast<StencilTable const *>(
\n-
342 BaseFactory::AppendLocalPointStencilTableFaceVarying(refiner,
\n-
343 static_cast<BaseTable const *>(baseStencilTable),
\n-
344 static_cast<BaseTable const *>(localPointStencilTable),
\n-
345 channel, factorize));
\n-
346 }
\n-
347};
\n-
348
\n-
349class LimitStencil;
\n-
350class LimitStencilTable;
\n-
351
\n-
354class LimitStencilTableFactory : public LimitStencilTableFactoryReal<float> {
\n-
355private:
\n-
356 typedef LimitStencilTableFactoryReal<float> BaseFactory;
\n-
357 typedef StencilTableReal<float> BaseTable;
\n-
358
\n-
359public:
\n-
360 static LimitStencilTable const * Create(
\n-
361 TopologyRefiner const & refiner,
\n-
362 LocationArrayVec const & locationArrays,
\n-
363 StencilTable const * cvStencils = 0,
\n-
364 PatchTable const * patchTable = 0,
\n-
365 Options options = Options()) {
\n-
366
\n-
367 return static_cast<LimitStencilTable const *>(
\n-
368 BaseFactory::Create(
\n-
369 refiner,
\n-
370 locationArrays,
\n-
371 static_cast<BaseTable const *>(cvStencils),
\n-
372 patchTable,
\n-
373 options));
\n-
374 }
\n-
375};
\n-
376
\n-
377} // end namespace Far
\n-
378
\n-
379} // end namespace OPENSUBDIV_VERSION
\n-
380using namespace OPENSUBDIV_VERSION;
\n-
381
\n-
382} // end namespace OpenSubdiv
\n-
383
\n-
384#endif // OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H
\n+
114template <>
\n+
115void
\n+
116TopologyRefinerFactory<TopologyDescriptor>::reportInvalidTopology(
\n+
117 TopologyError errCode, char const * msg, TopologyDescriptor const & desc);
\n+
118
\n+
119// @endcond
\n+
120
\n+
121} // end namespace Far
\n+
122
\n+
123} // end namespace OPENSUBDIV_VERSION
\n+
124using namespace OPENSUBDIV_VERSION;
\n+
125} // end namespace OpenSubdiv
\n+
126
\n+
127#endif /* OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H */
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Create
static StencilTableReal< REAL > const * Create(TopologyRefiner const &refiner, Options options=Options())
Instantiates StencilTable from TopologyRefiner that have been refined uniformly or adaptively.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::AppendLocalPointStencilTableFaceVarying
static StencilTableReal< REAL > const * AppendLocalPointStencilTableFaceVarying(TopologyRefiner const &refiner, StencilTableReal< REAL > const *baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, int channel=0, bool factorize=true)
Utility function for stencil splicing for local point face-varying stencils.
\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::AppendLocalPointStencilTableVarying
static StencilTableReal< REAL > const * AppendLocalPointStencilTableVarying(TopologyRefiner const &refiner, StencilTableReal< REAL > const *baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, bool factorize=true)
Utility function for stencil splicing for local point varying stencils.
Definition: stencilTableFactory.h:148
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::AppendLocalPointStencilTable
static StencilTableReal< REAL > const * AppendLocalPointStencilTable(TopologyRefiner const &refiner, StencilTableReal< REAL > const *baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, bool factorize=true)
Utility function for stencil splicing for local point stencils.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Create
static StencilTableReal< REAL > const * Create(int numTables, StencilTableReal< REAL > const **tables)
Instantiates StencilTable by concatenating an array of existing stencil tables.
\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Create
static LimitStencilTableReal< REAL > const * Create(TopologyRefiner const &refiner, LocationArrayVec const &locationArrays, StencilTableReal< REAL > const *cvStencils=0, PatchTable const *patchTable=0, Options options=Options())
Instantiates LimitStencilTable from a TopologyRefiner that has been refined either uniformly or adapt...
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil
Vertex stencil class wrapping the template for compatibility.
Definition: stencilTable.h:116
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::generateIntermediateLevels
unsigned int generateIntermediateLevels
vertices at all levels or highest only
Definition: stencilTableFactory.h:74
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::generateControlVerts
unsigned int generateControlVerts
generate stencils for control-vertices
Definition: stencilTableFactory.h:73
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::generateOffsets
unsigned int generateOffsets
populate optional "_offsets" field
Definition: stencilTableFactory.h:72
\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::generate2ndDerivatives
unsigned int generate2ndDerivatives
Generate weights for 2nd derivatives.
Definition: stencilTableFactory.h:228
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::generate1stDerivatives
unsigned int generate1stDerivatives
Generate weights for 1st derivatives.
Definition: stencilTableFactory.h:227
\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory
Stencil table factory class wrapping the template for compatibility.
Definition: stencilTableFactory.h:287
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::AppendLocalPointStencilTableVarying
static StencilTable const * AppendLocalPointStencilTableVarying(TopologyRefiner const &refiner, StencilTable const *baseStencilTable, StencilTable const *localPointStencilTable, bool factorize=true)
Definition: stencilTableFactory.h:321
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::Create
static StencilTable const * Create(TopologyRefiner const &refiner, Options options=Options())
Definition: stencilTableFactory.h:293
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::AppendLocalPointStencilTableFaceVarying
static StencilTable const * AppendLocalPointStencilTableFaceVarying(TopologyRefiner const &refiner, StencilTable const *baseStencilTable, StencilTable const *localPointStencilTable, int channel=0, bool factorize=true)
Definition: stencilTableFactory.h:334
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::AppendLocalPointStencilTable
static StencilTable const * AppendLocalPointStencilTable(TopologyRefiner const &refiner, StencilTable const *baseStencilTable, StencilTable const *localPointStencilTable, bool factorize=true)
Definition: stencilTableFactory.h:308
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::Create
static StencilTable const * Create(int numTables, StencilTable const **tables)
Definition: stencilTableFactory.h:300
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactory
Stencil table factory class wrapping the template for compatibility.
Definition: stencilTableFactory.h:354
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactory::Create
static LimitStencilTable const * Create(TopologyRefiner const &refiner, LocationArrayVec const &locationArrays, StencilTable const *cvStencils=0, PatchTable const *patchTable=0, Options options=Options())
Definition: stencilTableFactory.h:360
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor
A simple reference to raw topology data for use with TopologyRefinerFactory.
Definition: topologyDescriptor.h:49
\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::assignComponentTopology
static bool assignComponentTopology(TopologyRefiner &newRefiner, MESH const &mesh)
Specify the relationships between vertices, faces, etc. ie the face-vertices, vertex-faces,...
Definition: topologyRefinerFactory.h:621
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::reportInvalidTopology
static void reportInvalidTopology(TopologyError errCode, char const *msg, MESH const &mesh)
(Optional) Control run-time topology validation and error reporting
Definition: topologyRefinerFactory.h:707
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::assignComponentTags
static bool assignComponentTags(TopologyRefiner &newRefiner, MESH const &mesh)
(Optional) Specify edge or vertex sharpness or face holes
Definition: topologyRefinerFactory.h:690
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::resizeComponentTopology
static bool resizeComponentTopology(TopologyRefiner &newRefiner, MESH const &mesh)
Specify the number of vertices, faces, face-vertices, etc.
Definition: topologyRefinerFactory.h:580
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::assignFaceVaryingTopology
static bool assignFaceVaryingTopology(TopologyRefiner &newRefiner, MESH const &mesh)
(Optional) Specify face-varying data per face
Definition: topologyRefinerFactory.h:674
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner
Stores topology data for a specified set of refinement options.
Definition: topologyRefiner.h:51
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-stencilTableFactory.h\n+topologyDescriptor.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2013 Pixar\n+ 2// Copyright 2014 DreamWorks Animation LLC.\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -29,499 +29,208 @@\n 17//\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n- 24\n- 25#ifndef OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H\n- 26#define OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H\n- 27\n- 28#include \"../version.h\"\n- 29\n- 30#include \"../far/patchTable.h\"\n- 31\n- 32#include \n- 33\n- 34namespace OpenSubdiv {\n- 35namespace OPENSUBDIV_VERSION {\n- 36\n- 37namespace Far {\n- 38\n- 39class TopologyRefiner;\n- 40\n- 41template class StencilReal;\n- 42template class StencilTableReal;\n- 43\n- 44template class LimitStencilReal;\n- 45template class LimitStencilTableReal;\n- 46\n- 47\n- 50template \n-51class StencilTableFactoryReal {\n- 52\n- 53public:\n- 54\n-55 enum Mode {\n-56 INTERPOLATE_VERTEX=0,\n-57 INTERPOLATE_VARYING,\n- 58 INTERPOLATE_FACE_VARYING\n-59 };\n+ 24#ifndef OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H\n+ 25#define OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H\n+ 26\n+ 27#include \"../version.h\"\n+ 28\n+ 29#include \"../far/topologyRefiner.h\"\n+ 30#include \"../far/topologyRefinerFactory.h\"\n+ 31#include \"../far/error.h\"\n+ 32\n+ 33#include \n+ 34\n+ 35namespace OpenSubdiv {\n+ 36namespace OPENSUBDIV_VERSION {\n+ 37\n+ 38namespace Far {\n+ 39\n+49struct TopologyDescriptor {\n+ 50\n+51 int numVertices,\n+52 numFaces;\n+ 53\n+54 int const * numVertsPerFace;\n+55 Index const * vertIndicesPerFace;\n+ 56\n+57 int numCreases;\n+58 Index const * creaseVertexIndexPairs;\n+59 float const * creaseWeights;\n 60\n-61 struct Options {\n- 62\n-63 Options() : interpolationMode(INTERPOLATE_VERTEX),\n- 64 generateOffsets(false),\n- 65 generateControlVerts(false),\n- 66 generateIntermediateLevels(true),\n- 67 factorizeIntermediateLevels(true),\n- 68 maxLevel(10),\n- 69 fvarChannel(0) { }\n- 70\n-71 unsigned int interpolationMode : 2,\n-72 generateOffsets : 1,\n-73 generateControlVerts : 1,\n-74 generateIntermediateLevels : 1,\n-75 factorizeIntermediateLevels : 1,\n-78 maxLevel : 4;\n-79 unsigned int fvarChannel;\n- 81 };\n- 82\n-94 static StencilTableReal const * Create(\n- 95 TopologyRefiner const & refiner, Options options = Options());\n- 96\n- 97\n-110 static StencilTableReal const * Create(\n- 111 int numTables, StencilTableReal const ** tables);\n- 112\n+61 int numCorners;\n+62 Index const * cornerVertexIndices;\n+63 float const * cornerWeights;\n+ 64\n+65 int numHoles;\n+66 Index const * holeIndices;\n+ 67\n+68 bool isLeftHanded;\n+ 69\n+ 70 // Face-varying data channel -- value indices correspond to vertex indices,\n+ 71 // i.e. one for every vertex of every face:\n+ 72 //\n+73 struct FVarChannel {\n+ 74\n+75 int numValues;\n+76 Index const * valueIndices;\n+ 77\n+78 FVarChannel() : numValues(0), valueIndices(0) { }\n+ 79 };\n+ 80\n+81 int numFVarChannels;\n+82 FVarChannel const * fvarChannels;\n+ 83\n+84 TopologyDescriptor();\n+ 85};\n+ 86\n+ 87\n+ 88//\n+ 89// Forward declarations of required\n+TopologyRefinerFactory\n+ 90// specializations (defined internally):\n+ 91//\n+ 92// @cond EXCLUDE_DOXYGEN\n+ 93\n+ 94template <>\n+ 95bool\n+ 96TopologyRefinerFactory::resizeComponentTopology(\n+ 97 TopologyRefiner & refiner, TopologyDescriptor const & desc);\n+ 98\n+ 99template <>\n+ 100bool\n+ 101TopologyRefinerFactory::assignComponentTopology(\n+ 102 TopologyRefiner & refiner, TopologyDescriptor const & desc);\n+ 103\n+ 104template <>\n+ 105bool\n+ 106TopologyRefinerFactory::assignComponentTags(\n+ 107 TopologyRefiner & refiner, TopologyDescriptor const & desc);\n+ 108\n+ 109template <>\n+ 110bool\n+ 111TopologyRefinerFactory::assignFaceVaryingTopology(\n+ 112 TopologyRefiner & refiner, TopologyDescriptor const & desc);\n 113\n-128 static StencilTableReal const * AppendLocalPointStencilTable(\n- 129 TopologyRefiner const &refiner,\n- 130 StencilTableReal const *baseStencilTable,\n- 131 StencilTableReal const *localPointStencilTable,\n- 132 bool factorize = true);\n- 133\n-148 static StencilTableReal const * AppendLocalPointStencilTableVarying(\n- 149 TopologyRefiner const &refiner,\n- 150 StencilTableReal const *baseStencilTable,\n- 151 StencilTableReal const *localPointStencilTable,\n- 152 bool factorize = true) {\n- 153 return AppendLocalPointStencilTable(\n- 154 refiner, baseStencilTable, localPointStencilTable, factorize);\n- 155 }\n- 156\n-174 static StencilTableReal const *\n-AppendLocalPointStencilTableFaceVarying(\n- 175 TopologyRefiner const &refiner,\n- 176 StencilTableReal const *baseStencilTable,\n- 177 StencilTableReal const *localPointStencilTable,\n- 178 int channel = 0,\n- 179 bool factorize = true);\n- 180\n- 181private:\n- 182\n- 183 // Generate stencils for the coarse control-vertices (single weight =\n-1.0f)\n- 184 static void generateControlVertStencils(\n- 185 int numControlVerts,\n- 186 StencilReal & dst);\n- 187\n- 188 // Internal method to splice local point stencils\n- 189 static StencilTableReal const * appendLocalPointStencilTable(\n- 190 TopologyRefiner const &refiner,\n- 191 StencilTableReal const * baseStencilTable,\n- 192 StencilTableReal const * localPointStencilTable,\n- 193 int channel,\n- 194 bool factorize);\n- 195};\n- 196\n- 208template \n-209class LimitStencilTableFactoryReal {\n- 210\n- 211public:\n- 212\n-213 enum Mode {\n-214 INTERPOLATE_VERTEX=0,\n-215 INTERPOLATE_VARYING,\n- 216 INTERPOLATE_FACE_VARYING\n-217 };\n- 218\n-219 struct Options {\n- 220\n-221 Options() : interpolationMode(INTERPOLATE_VERTEX),\n- 222 generate1stDerivatives(true),\n- 223 generate2ndDerivatives(false),\n- 224 fvarChannel(0) { }\n- 225\n-226 unsigned int interpolationMode : 2,\n-227 generate1stDerivatives : 1,\n-228 generate2ndDerivatives : 1;\n-229 unsigned int fvarChannel;\n- 230 };\n- 231\n-233 struct LocationArray {\n- 234\n-235 LocationArray() : ptexIdx(-1), numLocations(0), s(0), t(0) { }\n- 236\n-237 int ptexIdx,\n-238 numLocations;\n- 239\n-240 REAL const * s,\n-241 * t;\n- 242 };\n- 243\n-244 typedef std::vector LocationArrayVec;\n- 245\n-269 static LimitStencilTableReal const * Create(\n- 270 TopologyRefiner const & refiner,\n- 271 LocationArrayVec const & locationArrays,\n- 272 StencilTableReal const * cvStencils = 0,\n- 273 PatchTable const * patchTable = 0,\n- 274 Options options = Options());\n- 275\n- 276};\n- 277\n- 278\n- 279//\n- 280// Public wrapper classes for the templates\n- 281//\n- 282class Stencil;\n- 283class StencilTable;\n- 284\n-287class StencilTableFactory : public StencilTableFactoryReal {\n- 288private:\n- 289 typedef StencilTableFactoryReal BaseFactory;\n- 290 typedef StencilTableReal BaseTable;\n- 291\n- 292public:\n-293 static StencilTable const * Create(\n- 294 TopologyRefiner const & refiner, Options options = Options()) {\n- 295\n- 296 return static_cast(\n- 297 BaseFactory::Create(refiner, options));\n- 298 }\n- 299\n-300 static StencilTable const * Create(\n- 301 int numTables, StencilTable const ** tables) {\n- 302\n- 303 return static_cast(\n- 304 BaseFactory::Create(numTables,\n- 305 reinterpret_cast(tables)));\n- 306 }\n- 307\n-308 static StencilTable const * AppendLocalPointStencilTable(\n- 309 TopologyRefiner const &refiner,\n- 310 StencilTable const *baseStencilTable,\n- 311 StencilTable const *localPointStencilTable,\n- 312 bool factorize = true) {\n- 313\n- 314 return static_cast(\n- 315 BaseFactory::AppendLocalPointStencilTable(refiner,\n- 316 static_cast(baseStencilTable),\n- 317 static_cast(localPointStencilTable),\n- 318 factorize));\n- 319 }\n- 320\n-321 static StencilTable const * AppendLocalPointStencilTableVarying(\n- 322 TopologyRefiner const &refiner,\n- 323 StencilTable const *baseStencilTable,\n- 324 StencilTable const *localPointStencilTable,\n- 325 bool factorize = true) {\n- 326\n- 327 return static_cast(\n- 328 BaseFactory::AppendLocalPointStencilTableVarying(refiner,\n- 329 static_cast(baseStencilTable),\n- 330 static_cast(localPointStencilTable),\n- 331 factorize));\n- 332 }\n- 333\n-334 static StencilTable const * AppendLocalPointStencilTableFaceVarying(\n- 335 TopologyRefiner const &refiner,\n- 336 StencilTable const *baseStencilTable,\n- 337 StencilTable const *localPointStencilTable,\n- 338 int channel = 0,\n- 339 bool factorize = true) {\n- 340\n- 341 return static_cast(\n- 342 BaseFactory::AppendLocalPointStencilTableFaceVarying(refiner,\n- 343 static_cast(baseStencilTable),\n- 344 static_cast(localPointStencilTable),\n- 345 channel, factorize));\n- 346 }\n- 347};\n- 348\n- 349class LimitStencil;\n- 350class LimitStencilTable;\n- 351\n-354class LimitStencilTableFactory : public LimitStencilTableFactoryReal\n-{\n- 355private:\n- 356 typedef LimitStencilTableFactoryReal BaseFactory;\n- 357 typedef StencilTableReal BaseTable;\n- 358\n- 359public:\n-360 static LimitStencilTable const * Create(\n- 361 TopologyRefiner const & refiner,\n- 362 LocationArrayVec const & locationArrays,\n- 363 StencilTable const * cvStencils = 0,\n- 364 PatchTable const * patchTable = 0,\n- 365 Options options = Options()) {\n- 366\n- 367 return static_cast(\n- 368 BaseFactory::Create(\n- 369 refiner,\n- 370 locationArrays,\n- 371 static_cast(cvStencils),\n- 372 patchTable,\n- 373 options));\n- 374 }\n- 375};\n- 376\n- 377} // end namespace Far\n- 378\n- 379} // end namespace OPENSUBDIV_VERSION\n- 380using namespace OPENSUBDIV_VERSION;\n- 381\n- 382} // end namespace OpenSubdiv\n- 383\n- 384#endif // OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H\n+ 114template <>\n+ 115void\n+ 116TopologyRefinerFactory::reportInvalidTopology(\n+ 117 TopologyError errCode, char const * msg, TopologyDescriptor const & desc);\n+ 118\n+ 119// @endcond\n+ 120\n+ 121} // end namespace Far\n+ 122\n+ 123} // end namespace OPENSUBDIV_VERSION\n+ 124using namespace OPENSUBDIV_VERSION;\n+ 125} // end namespace OpenSubdiv\n+ 126\n+ 127#endif /* OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H */\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n-Container for arrays of parametric patches.\n-Definition: patchTable.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal\n-A specialized factory for StencilTable.\n-Definition: stencilTableFactory.h:51\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Create\n-static StencilTableReal< REAL > const * Create(TopologyRefiner const &refiner,\n-Options options=Options())\n-Instantiates StencilTable from TopologyRefiner that have been refined uniformly\n-or adaptively.\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n-AppendLocalPointStencilTableFaceVarying\n-static StencilTableReal< REAL > const * AppendLocalPointStencilTableFaceVarying\n-(TopologyRefiner const &refiner, StencilTableReal< REAL > const\n-*baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, int\n-channel=0, bool factorize=true)\n-Utility function for stencil splicing for local point face-varying stencils.\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Mode\n-Mode\n-Definition: stencilTableFactory.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n-INTERPOLATE_FACE_VARYING\n-@ INTERPOLATE_FACE_VARYING\n-face-varying primvar stencils\n-Definition: stencilTableFactory.h:58\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n-INTERPOLATE_VARYING\n-@ INTERPOLATE_VARYING\n-varying primvar stencils\n-Definition: stencilTableFactory.h:57\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n-INTERPOLATE_VERTEX\n-@ INTERPOLATE_VERTEX\n-vertex primvar stencils\n-Definition: stencilTableFactory.h:56\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n-AppendLocalPointStencilTableVarying\n-static StencilTableReal< REAL > const * AppendLocalPointStencilTableVarying\n-(TopologyRefiner const &refiner, StencilTableReal< REAL > const\n-*baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, bool\n-factorize=true)\n-Utility function for stencil splicing for local point varying stencils.\n-Definition: stencilTableFactory.h:148\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n-AppendLocalPointStencilTable\n-static StencilTableReal< REAL > const * AppendLocalPointStencilTable\n-(TopologyRefiner const &refiner, StencilTableReal< REAL > const\n-*baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, bool\n-factorize=true)\n-Utility function for stencil splicing for local point stencils.\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Create\n-static StencilTableReal< REAL > const * Create(int numTables, StencilTableReal<\n-REAL > const **tables)\n-Instantiates StencilTable by concatenating an array of existing stencil tables.\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal\n-A specialized factory for LimitStencilTable.\n-Definition: stencilTableFactory.h:209\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Mode\n-Mode\n-Definition: stencilTableFactory.h:213\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n-INTERPOLATE_FACE_VARYING\n-@ INTERPOLATE_FACE_VARYING\n-face-varying primvar stencils\n-Definition: stencilTableFactory.h:216\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n-INTERPOLATE_VARYING\n-@ INTERPOLATE_VARYING\n-varying primvar stencils\n-Definition: stencilTableFactory.h:215\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n-INTERPOLATE_VERTEX\n-@ INTERPOLATE_VERTEX\n-vertex primvar stencils\n-Definition: stencilTableFactory.h:214\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n-LocationArrayVec\n-std::vector< LocationArray > LocationArrayVec\n-Definition: stencilTableFactory.h:244\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Create\n-static LimitStencilTableReal< REAL > const * Create(TopologyRefiner const\n-&refiner, LocationArrayVec const &locationArrays, StencilTableReal< REAL >\n-const *cvStencils=0, PatchTable const *patchTable=0, Options options=Options())\n-Instantiates LimitStencilTable from a TopologyRefiner that has been refined\n-either uniformly or adapt...\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal\n-Vertex stencil descriptor.\n-Definition: stencilTable.h:53\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil\n-Vertex stencil class wrapping the template for compatibility.\n-Definition: stencilTable.h:116\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal\n-Table of subdivision stencils.\n-Definition: stencilTable.h:141\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n-Stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:273\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal\n-Table of limit subdivision stencils.\n-Definition: stencilTable.h:409\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n-Limit stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:583\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options\n-Definition: stencilTableFactory.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n-factorizeIntermediateLevels\n-unsigned int factorizeIntermediateLevels\n-Definition: stencilTableFactory.h:77\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::maxLevel\n-unsigned int maxLevel\n-generate stencils up to 'maxLevel'\n-Definition: stencilTableFactory.h:78\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::Options\n-Options()\n-Definition: stencilTableFactory.h:63\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n-generateIntermediateLevels\n-unsigned int generateIntermediateLevels\n-vertices at all levels or highest only\n-Definition: stencilTableFactory.h:74\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n-generateControlVerts\n-unsigned int generateControlVerts\n-generate stencils for control-vertices\n-Definition: stencilTableFactory.h:73\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n-generateOffsets\n-unsigned int generateOffsets\n-populate optional \"_offsets\" field\n-Definition: stencilTableFactory.h:72\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n-fvarChannel\n-unsigned int fvarChannel\n-Definition: stencilTableFactory.h:79\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n-interpolationMode\n-unsigned int interpolationMode\n-interpolation mode\n-Definition: stencilTableFactory.h:71\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options\n-Definition: stencilTableFactory.h:219\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::\n-Options\n-Options()\n-Definition: stencilTableFactory.h:221\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::\n-fvarChannel\n-unsigned int fvarChannel\n-face-varying channel to use\n-Definition: stencilTableFactory.h:229\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::\n-interpolationMode\n-unsigned int interpolationMode\n-interpolation mode\n-Definition: stencilTableFactory.h:226\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::\n-generate2ndDerivatives\n-unsigned int generate2ndDerivatives\n-Generate weights for 2nd derivatives.\n-Definition: stencilTableFactory.h:228\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::\n-generate1stDerivatives\n-unsigned int generate1stDerivatives\n-Generate weights for 1st derivatives.\n-Definition: stencilTableFactory.h:227\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n-LocationArray\n-Descriptor for limit surface locations.\n-Definition: stencilTableFactory.h:233\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n-LocationArray::t\n-REAL const * t\n-array of v coordinates\n-Definition: stencilTableFactory.h:241\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n-LocationArray::s\n-REAL const * s\n-array of u coordinates\n-Definition: stencilTableFactory.h:240\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n-LocationArray::LocationArray\n-LocationArray()\n-Definition: stencilTableFactory.h:235\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n-LocationArray::numLocations\n-int numLocations\n-number of (u,v) coordinates in the array\n-Definition: stencilTableFactory.h:238\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n-LocationArray::ptexIdx\n-int ptexIdx\n-ptex face index\n-Definition: stencilTableFactory.h:237\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory\n-Stencil table factory class wrapping the template for compatibility.\n-Definition: stencilTableFactory.h:287\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::\n-AppendLocalPointStencilTableVarying\n-static StencilTable const * AppendLocalPointStencilTableVarying(TopologyRefiner\n-const &refiner, StencilTable const *baseStencilTable, StencilTable const\n-*localPointStencilTable, bool factorize=true)\n-Definition: stencilTableFactory.h:321\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::Create\n-static StencilTable const * Create(TopologyRefiner const &refiner, Options\n-options=Options())\n-Definition: stencilTableFactory.h:293\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::\n-AppendLocalPointStencilTableFaceVarying\n-static StencilTable const * AppendLocalPointStencilTableFaceVarying\n-(TopologyRefiner const &refiner, StencilTable const *baseStencilTable,\n-StencilTable const *localPointStencilTable, int channel=0, bool factorize=true)\n-Definition: stencilTableFactory.h:334\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::\n-AppendLocalPointStencilTable\n-static StencilTable const * AppendLocalPointStencilTable(TopologyRefiner const\n-&refiner, StencilTable const *baseStencilTable, StencilTable const\n-*localPointStencilTable, bool factorize=true)\n-Definition: stencilTableFactory.h:308\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::Create\n-static StencilTable const * Create(int numTables, StencilTable const **tables)\n-Definition: stencilTableFactory.h:300\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactory\n-Stencil table factory class wrapping the template for compatibility.\n-Definition: stencilTableFactory.h:354\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactory::Create\n-static LimitStencilTable const * Create(TopologyRefiner const &refiner,\n-LocationArrayVec const &locationArrays, StencilTable const *cvStencils=0,\n-PatchTable const *patchTable=0, Options options=Options())\n-Definition: stencilTableFactory.h:360\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::Index\n+Vtr::Index Index\n+Definition: types.h:41\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor\n+A simple reference to raw topology data for use with TopologyRefinerFactory.\n+Definition: topologyDescriptor.h:49\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::fvarChannels\n+FVarChannel const * fvarChannels\n+Definition: topologyDescriptor.h:82\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numVertices\n+int numVertices\n+Definition: topologyDescriptor.h:51\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::cornerVertexIndices\n+Index const * cornerVertexIndices\n+Definition: topologyDescriptor.h:62\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::vertIndicesPerFace\n+Index const * vertIndicesPerFace\n+Definition: topologyDescriptor.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::creaseVertexIndexPairs\n+Index const * creaseVertexIndexPairs\n+Definition: topologyDescriptor.h:58\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::holeIndices\n+Index const * holeIndices\n+Definition: topologyDescriptor.h:66\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numFaces\n+int numFaces\n+Definition: topologyDescriptor.h:52\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numCorners\n+int numCorners\n+Definition: topologyDescriptor.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::cornerWeights\n+float const * cornerWeights\n+Definition: topologyDescriptor.h:63\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::isLeftHanded\n+bool isLeftHanded\n+Definition: topologyDescriptor.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numVertsPerFace\n+int const * numVertsPerFace\n+Definition: topologyDescriptor.h:54\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::TopologyDescriptor\n+TopologyDescriptor()\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::creaseWeights\n+float const * creaseWeights\n+Definition: topologyDescriptor.h:59\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numCreases\n+int numCreases\n+Definition: topologyDescriptor.h:57\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numHoles\n+int numHoles\n+Definition: topologyDescriptor.h:65\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numFVarChannels\n+int numFVarChannels\n+Definition: topologyDescriptor.h:81\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::FVarChannel\n+Definition: topologyDescriptor.h:73\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::FVarChannel::\n+FVarChannel\n+FVarChannel()\n+Definition: topologyDescriptor.h:78\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::FVarChannel::\n+valueIndices\n+Index const * valueIndices\n+Definition: topologyDescriptor.h:76\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::FVarChannel::numValues\n+int numValues\n+Definition: topologyDescriptor.h:75\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::\n+assignComponentTopology\n+static bool assignComponentTopology(TopologyRefiner &newRefiner, MESH const\n+&mesh)\n+Specify the relationships between vertices, faces, etc. ie the face-vertices,\n+vertex-faces,...\n+Definition: topologyRefinerFactory.h:621\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::\n+reportInvalidTopology\n+static void reportInvalidTopology(TopologyError errCode, char const *msg, MESH\n+const &mesh)\n+(Optional) Control run-time topology validation and error reporting\n+Definition: topologyRefinerFactory.h:707\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::\n+assignComponentTags\n+static bool assignComponentTags(TopologyRefiner &newRefiner, MESH const &mesh)\n+(Optional) Specify edge or vertex sharpness or face holes\n+Definition: topologyRefinerFactory.h:690\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::\n+resizeComponentTopology\n+static bool resizeComponentTopology(TopologyRefiner &newRefiner, MESH const\n+&mesh)\n+Specify the number of vertices, faces, face-vertices, etc.\n+Definition: topologyRefinerFactory.h:580\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::\n+assignFaceVaryingTopology\n+static bool assignFaceVaryingTopology(TopologyRefiner &newRefiner, MESH const\n+&mesh)\n+(Optional) Specify face-varying data per face\n+Definition: topologyRefinerFactory.h:674\n OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner\n Stores topology data for a specified set of refinement options.\n Definition: topologyRefiner.h:51\n * opensubdiv\n * far\n- * stencilTableFactory.h\n+ * topologyDescriptor.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00713.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00713.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/topologyDescriptor.h File Reference\n+OpenSubdiv: opensubdiv/far/stencilTable.h File Reference\n \n \n \n \n \n \n \n@@ -83,31 +83,51 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
topologyDescriptor.h File Reference
\n+
stencilTable.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../far/topologyRefiner.h"
\n-#include "../far/topologyRefinerFactory.h"
\n-#include "../far/error.h"
\n+#include "../far/types.h"
\n #include <cassert>
\n+#include <cstring>
\n+#include <vector>
\n+#include <iostream>
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n+\n+\n \n-\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n \n

\n Classes

struct  TopologyDescriptor
 A simple reference to raw topology data for use with TopologyRefinerFactory. More...
class  StencilReal< REAL >
 Vertex stencil descriptor. More...
 
struct  TopologyDescriptor::FVarChannel
class  Stencil
 Vertex stencil class wrapping the template for compatibility. More...
 
class  StencilTableReal< REAL >
 Table of subdivision stencils. More...
 
class  StencilTable
 Stencil table class wrapping the template for compatibility. More...
 
class  LimitStencilReal< REAL >
 Limit point stencil descriptor. More...
 
class  LimitStencil
 Limit point stencil class wrapping the template for compatibility. More...
 
class  LimitStencilTableReal< REAL >
 Table of limit subdivision stencils. More...
 
class  LimitStencilTable
 Limit stencil table class wrapping the template for compatibility. More...
 
\n \n \n \n \n@@ -116,13 +136,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,32 +5,53 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-topologyDescriptor.h File Reference\n+stencilTable.h File Reference\n #include \"../version.h\"\n-#include \"../far/topologyRefiner.h\"\n-#include \"../far/topologyRefinerFactory.h\"\n-#include \"../far/error.h\"\n+#include \"../far/types.h\"\n #include \n+#include \n+#include \n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n-struct \u00a0TopologyDescriptor\n-\u00a0 A simple reference to raw topology data for use with\n- TopologyRefinerFactory. More...\n+class \u00a0StencilReal<_REAL_>\n+\u00a0 Vertex stencil descriptor. More...\n \u00a0\n-struct \u00a0TopologyDescriptor::FVarChannel\n+class \u00a0Stencil\n+\u00a0 Vertex stencil class wrapping the template for compatibility. More...\n+\u00a0\n+class \u00a0StencilTableReal<_REAL_>\n+\u00a0 Table of subdivision stencils. More...\n+\u00a0\n+class \u00a0StencilTable\n+\u00a0 Stencil table class wrapping the template for compatibility. More...\n+\u00a0\n+class \u00a0LimitStencilReal<_REAL_>\n+\u00a0 Limit point stencil descriptor. More...\n+\u00a0\n+class \u00a0LimitStencil\n+\u00a0 Limit point stencil class wrapping the template for compatibility.\n+ More...\n+\u00a0\n+class \u00a0LimitStencilTableReal<_REAL_>\n+\u00a0 Table of limit subdivision stencils. More...\n+\u00a0\n+class \u00a0LimitStencilTable\n+\u00a0 Limit stencil table class wrapping the template for compatibility.\n+ More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n * opensubdiv\n * far\n- * topologyDescriptor.h\n+ * stencilTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00713.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00713.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,10 @@\n var a00713 = [\n- [\"TopologyDescriptor\", \"a01097.html\", \"a01097\"],\n- [\"TopologyDescriptor::FVarChannel\", \"a01101.html\", \"a01101\"]\n+ [\"StencilReal< REAL >\", \"a01045.html\", \"a01045\"],\n+ [\"Stencil\", \"a01049.html\", \"a01049\"],\n+ [\"StencilTableReal< REAL >\", \"a01053.html\", \"a01053\"],\n+ [\"StencilTable\", \"a01057.html\", \"a01057\"],\n+ [\"LimitStencilReal< REAL >\", \"a01061.html\", \"a01061\"],\n+ [\"LimitStencil\", \"a01065.html\", \"a01065\"],\n+ [\"LimitStencilTableReal< REAL >\", \"a01069.html\", \"a01069\"],\n+ [\"LimitStencilTable\", \"a01073.html\", \"a01073\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00713_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00713_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/topologyDescriptor.h Source File\n+OpenSubdiv: opensubdiv/far/stencilTable.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
topologyDescriptor.h
\n+
stencilTable.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2014 DreamWorks Animation LLC.
\n+
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -106,142 +106,747 @@\n
17//
\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n-
24#ifndef OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H
\n-
25#define OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H
\n-
26
\n-
27#include "../version.h"
\n-
28
\n-
29#include "../far/topologyRefiner.h"
\n-
30#include "../far/topologyRefinerFactory.h"
\n-
31#include "../far/error.h"
\n-
32
\n-
33#include <cassert>
\n-
34
\n-
35namespace OpenSubdiv {
\n-
36namespace OPENSUBDIV_VERSION {
\n-
37
\n-
38namespace Far {
\n+
24
\n+
25#ifndef OPENSUBDIV3_FAR_STENCILTABLE_H
\n+
26#define OPENSUBDIV3_FAR_STENCILTABLE_H
\n+
27
\n+
28#include "../version.h"
\n+
29
\n+
30#include "../far/types.h"
\n+
31
\n+
32#include <cassert>
\n+
33#include <cstring>
\n+
34#include <vector>
\n+
35#include <iostream>
\n+
36
\n+
37namespace OpenSubdiv {
\n+
38namespace OPENSUBDIV_VERSION {
\n
39
\n-
49struct TopologyDescriptor {
\n-
50
\n-
51 int numVertices,
\n-
52 numFaces;
\n-
53
\n-
54 int const * numVertsPerFace;
\n-
55 Index const * vertIndicesPerFace;
\n-
56
\n-
57 int numCreases;
\n-
58 Index const * creaseVertexIndexPairs;
\n-
59 float const * creaseWeights;
\n-
60
\n-
61 int numCorners;
\n-
62 Index const * cornerVertexIndices;
\n-
63 float const * cornerWeights;
\n-
64
\n-
65 int numHoles;
\n-
66 Index const * holeIndices;
\n-
67
\n-
68 bool isLeftHanded;
\n+
40namespace Far {
\n+
41
\n+
42// Forward declarations for friends:
\n+
43class PatchTableBuilder;
\n+
44
\n+
45template <typename REAL> class StencilTableFactoryReal;
\n+
46template <typename REAL> class LimitStencilTableFactoryReal;
\n+
47
\n+
52template <typename REAL>
\n+
53class StencilReal {
\n+
54public:
\n+
55
\n+
57 StencilReal() {}
\n+
58
\n+
67 StencilReal(int * size, Index * indices, REAL * weights)
\n+
68 : _size(size), _indices(indices), _weights(weights) { }
\n
69
\n-
70 // Face-varying data channel -- value indices correspond to vertex indices,
\n-
71 // i.e. one for every vertex of every face:
\n-
72 //
\n-
73 struct FVarChannel {
\n-
74
\n-
75 int numValues;
\n-
76 Index const * valueIndices;
\n-
77
\n-
78 FVarChannel() : numValues(0), valueIndices(0) { }
\n-
79 };
\n-
80
\n-
81 int numFVarChannels;
\n-
82 FVarChannel const * fvarChannels;
\n-
83
\n-
84 TopologyDescriptor();
\n-
85};
\n+
71 StencilReal(StencilReal const & other) {
\n+
72 _size = other._size;
\n+
73 _indices = other._indices;
\n+
74 _weights = other._weights;
\n+
75 }
\n+
76
\n+
78 int GetSize() const {
\n+
79 return *_size;
\n+
80 }
\n+
81
\n+
83 int * GetSizePtr() const {
\n+
84 return _size;
\n+
85 }
\n
86
\n-
87
\n-
88//
\n-
89// Forward declarations of required TopologyRefinerFactory<TopologyDescriptor>
\n-
90// specializations (defined internally):
\n-
91//
\n-
92// @cond EXCLUDE_DOXYGEN
\n-
93
\n-
94template <>
\n-
95bool
\n-
96TopologyRefinerFactory<TopologyDescriptor>::resizeComponentTopology(
\n-
97 TopologyRefiner & refiner, TopologyDescriptor const & desc);
\n-
98
\n-
99template <>
\n-
100bool
\n-
101TopologyRefinerFactory<TopologyDescriptor>::assignComponentTopology(
\n-
102 TopologyRefiner & refiner, TopologyDescriptor const & desc);
\n-
103
\n-
104template <>
\n-
105bool
\n-
106TopologyRefinerFactory<TopologyDescriptor>::assignComponentTags(
\n-
107 TopologyRefiner & refiner, TopologyDescriptor const & desc);
\n+
88 Index const * GetVertexIndices() const {
\n+
89 return _indices;
\n+
90 }
\n+
91
\n+
93 REAL const * GetWeights() const {
\n+
94 return _weights;
\n+
95 }
\n+
96
\n+
98 void Next() {
\n+
99 int stride = *_size;
\n+
100 ++_size;
\n+
101 _indices += stride;
\n+
102 _weights += stride;
\n+
103 }
\n+
104
\n+
105protected:
\n+
106 friend class StencilTableFactoryReal<REAL>;
\n+
107 friend class LimitStencilTableFactoryReal<REAL>;
\n
108
\n-
109template <>
\n-
110bool
\n-
111TopologyRefinerFactory<TopologyDescriptor>::assignFaceVaryingTopology(
\n-
112 TopologyRefiner & refiner, TopologyDescriptor const & desc);
\n+
109 int * _size;
\n+
110 Index * _indices;
\n+
111 REAL * _weights;
\n+
112};
\n
113
\n-
114template <>
\n-
115void
\n-
116TopologyRefinerFactory<TopologyDescriptor>::reportInvalidTopology(
\n-
117 TopologyError errCode, char const * msg, TopologyDescriptor const & desc);
\n-
118
\n-
119// @endcond
\n-
120
\n-
121} // end namespace Far
\n-
122
\n-
123} // end namespace OPENSUBDIV_VERSION
\n-
124using namespace OPENSUBDIV_VERSION;
\n-
125} // end namespace OpenSubdiv
\n+
116class Stencil : public StencilReal<float> {
\n+
117protected:
\n+
118 typedef StencilReal<float> BaseStencil;
\n+
119
\n+
120public:
\n+
121 Stencil() : BaseStencil() { }
\n+
122 Stencil(BaseStencil const & other) : BaseStencil(other) { }
\n+
123 Stencil(int * size, Index * indices, float * weights)
\n+
124 : BaseStencil(size, indices, weights) { }
\n+
125};
\n
126
\n-
127#endif /* OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H */
\n+
127
\n+
140template <typename REAL>
\n+
141class StencilTableReal {
\n+
142protected:
\n+
143 StencilTableReal(int numControlVerts,
\n+
144 std::vector<int> const& offsets,
\n+
145 std::vector<int> const& sizes,
\n+
146 std::vector<int> const& sources,
\n+
147 std::vector<REAL> const& weights,
\n+
148 bool includeCoarseVerts,
\n+
149 size_t firstOffset);
\n+
150
\n+
151public:
\n+
152
\n+
153 virtual ~StencilTableReal() {};
\n+
154
\n+
156 int GetNumStencils() const {
\n+
157 return (int)_sizes.size();
\n+
158 }
\n+
159
\n+
161 int GetNumControlVertices() const {
\n+
162 return _numControlVertices;
\n+
163 }
\n+
164
\n+
166 StencilReal<REAL> GetStencil(Index i) const;
\n+
167
\n+
169 std::vector<int> const & GetSizes() const {
\n+
170 return _sizes;
\n+
171 }
\n+
172
\n+
174 std::vector<Index> const & GetOffsets() const {
\n+
175 return _offsets;
\n+
176 }
\n+
177
\n+
179 std::vector<Index> const & GetControlIndices() const {
\n+
180 return _indices;
\n+
181 }
\n+
182
\n+
184 std::vector<REAL> const & GetWeights() const {
\n+
185 return _weights;
\n+
186 }
\n+
187
\n+
189 StencilReal<REAL> operator[] (Index index) const;
\n+
190
\n+
204 template <class T, class U>
\n+
205 void UpdateValues(T const &srcValues, U &dstValues, Index start=-1, Index end=-1) const {
\n+
206 this->update(srcValues, dstValues, _weights, start, end);
\n+
207 }
\n+
208
\n+
209 template <class T1, class T2, class U>
\n+
210 void UpdateValues(T1 const &srcBase, int numBase, T2 const &srcRef,
\n+
211 U &dstValues, Index start=-1, Index end=-1) const {
\n+
212 this->update(srcBase, numBase, srcRef, dstValues, _weights, start, end);
\n+
213 }
\n+
214
\n+
215 // Pointer interface for backward compatibility
\n+
216 template <class T, class U>
\n+
217 void UpdateValues(T const *src, U *dst, Index start=-1, Index end=-1) const {
\n+
218 this->update(src, dst, _weights, start, end);
\n+
219 }
\n+
220 template <class T1, class T2, class U>
\n+
221 void UpdateValues(T1 const *srcBase, int numBase, T2 const *srcRef,
\n+
222 U *dst, Index start=-1, Index end=-1) const {
\n+
223 this->update(srcBase, numBase, srcRef, dst, _weights, start, end);
\n+
224 }
\n+
225
\n+
227 void Clear();
\n+
228
\n+
229protected:
\n+
230
\n+
231 // Update values by applying cached stencil weights to new control values
\n+
232 template <class T, class U>
\n+
233 void update( T const &srcValues, U &dstValues,
\n+
234 std::vector<REAL> const & valueWeights, Index start, Index end) const;
\n+
235 template <class T1, class T2, class U>
\n+
236 void update( T1 const &srcBase, int numBase, T2 const &srcRef, U &dstValues,
\n+
237 std::vector<REAL> const & valueWeights, Index start, Index end) const;
\n+
238
\n+
239 // Populate the offsets table from the stencil sizes in _sizes (factory helper)
\n+
240 void generateOffsets();
\n+
241
\n+
242 // Resize the table arrays (factory helper)
\n+
243 void resize(int nstencils, int nelems);
\n+
244
\n+
245 // Reserves the table arrays (factory helper)
\n+
246 void reserve(int nstencils, int nelems);
\n+
247
\n+
248 // Reallocates the table arrays to remove excess capacity (factory helper)
\n+
249 void shrinkToFit();
\n+
250
\n+
251 // Performs any final operations on internal tables (factory helper)
\n+
252 void finalize();
\n+
253
\n+
254protected:
\n+
255 StencilTableReal() : _numControlVertices(0) {}
\n+
256 StencilTableReal(int numControlVerts)
\n+
257 : _numControlVertices(numControlVerts)
\n+
258 { }
\n+
259
\n+
260 friend class StencilTableFactoryReal<REAL>;
\n+
261 friend class Far::PatchTableBuilder;
\n+
262
\n+
263 int _numControlVertices; // number of control vertices
\n+
264
\n+
265 std::vector<int> _sizes; // number of coefficients for each stencil
\n+
266 std::vector<Index> _offsets, // offset to the start of each stencil
\n+
267 _indices; // indices of contributing coarse vertices
\n+
268 std::vector<REAL> _weights; // stencil weight coefficients
\n+
269};
\n+
270
\n+
273class StencilTable : public StencilTableReal<float> {
\n+
274protected:
\n+
275 typedef StencilTableReal<float> BaseTable;
\n+
276
\n+
277public:
\n+
278 Stencil GetStencil(Index index) const {
\n+
279 return Stencil(BaseTable::GetStencil(index));
\n+
280 }
\n+
281 Stencil operator[] (Index index) const {
\n+
282 return Stencil(BaseTable::GetStencil(index));
\n+
283 }
\n+
284
\n+
285protected:
\n+
286 StencilTable() : BaseTable() { }
\n+
287 StencilTable(int numControlVerts) : BaseTable(numControlVerts) { }
\n+
288 StencilTable(int numControlVerts,
\n+
289 std::vector<int> const& offsets,
\n+
290 std::vector<int> const& sizes,
\n+
291 std::vector<int> const& sources,
\n+
292 std::vector<float> const& weights,
\n+
293 bool includeCoarseVerts,
\n+
294 size_t firstOffset)
\n+
295 : BaseTable(numControlVerts, offsets,
\n+
296 sizes, sources, weights, includeCoarseVerts, firstOffset) { }
\n+
297};
\n+
298
\n+
299
\n+
302template <typename REAL>
\n+
303class LimitStencilReal : public StencilReal<REAL> {
\n+
304public:
\n+
305
\n+
324 LimitStencilReal( int* size,
\n+
325 Index * indices,
\n+
326 REAL * weights,
\n+
327 REAL * duWeights=0,
\n+
328 REAL * dvWeights=0,
\n+
329 REAL * duuWeights=0,
\n+
330 REAL * duvWeights=0,
\n+
331 REAL * dvvWeights=0)
\n+
332 : StencilReal<REAL>(size, indices, weights),
\n+
333 _duWeights(duWeights),
\n+
334 _dvWeights(dvWeights),
\n+
335 _duuWeights(duuWeights),
\n+
336 _duvWeights(duvWeights),
\n+
337 _dvvWeights(dvvWeights) {
\n+
338 }
\n+
339
\n+
341 REAL const * GetDuWeights() const {
\n+
342 return _duWeights;
\n+
343 }
\n+
344
\n+
346 REAL const * GetDvWeights() const {
\n+
347 return _dvWeights;
\n+
348 }
\n+
349
\n+
351 REAL const * GetDuuWeights() const {
\n+
352 return _duuWeights;
\n+
353 }
\n+
354
\n+
356 REAL const * GetDuvWeights() const {
\n+
357 return _duvWeights;
\n+
358 }
\n+
359
\n+
361 REAL const * GetDvvWeights() const {
\n+
362 return _dvvWeights;
\n+
363 }
\n+
364
\n+
366 void Next() {
\n+
367 int stride = *this->_size;
\n+
368 ++this->_size;
\n+
369 this->_indices += stride;
\n+
370 this->_weights += stride;
\n+
371 if (_duWeights) _duWeights += stride;
\n+
372 if (_dvWeights) _dvWeights += stride;
\n+
373 if (_duuWeights) _duuWeights += stride;
\n+
374 if (_duvWeights) _duvWeights += stride;
\n+
375 if (_dvvWeights) _dvvWeights += stride;
\n+
376 }
\n+
377
\n+
378private:
\n+
379
\n+
380 friend class StencilTableFactoryReal<REAL>;
\n+
381 friend class LimitStencilTableFactoryReal<REAL>;
\n+
382
\n+
383 REAL * _duWeights, // pointer to stencil u derivative limit weights
\n+
384 * _dvWeights, // pointer to stencil v derivative limit weights
\n+
385 * _duuWeights, // pointer to stencil uu derivative limit weights
\n+
386 * _duvWeights, // pointer to stencil uv derivative limit weights
\n+
387 * _dvvWeights; // pointer to stencil vv derivative limit weights
\n+
388};
\n+
389
\n+
392class LimitStencil : public LimitStencilReal<float> {
\n+
393protected:
\n+
394 typedef LimitStencilReal<float> BaseStencil;
\n+
395
\n+
396public:
\n+
397 LimitStencil(BaseStencil const & other) : BaseStencil(other) { }
\n+
398 LimitStencil(int* size, Index * indices, float * weights,
\n+
399 float * duWeights=0, float * dvWeights=0,
\n+
400 float * duuWeights=0, float * duvWeights=0, float * dvvWeights=0)
\n+
401 : BaseStencil(size, indices, weights,
\n+
402 duWeights, dvWeights, duuWeights, duvWeights, dvvWeights) { }
\n+
403};
\n+
404
\n+
405
\n+
408template <typename REAL>
\n+
409class LimitStencilTableReal : public StencilTableReal<REAL> {
\n+
410protected:
\n+
411 LimitStencilTableReal(
\n+
412 int numControlVerts,
\n+
413 std::vector<int> const& offsets,
\n+
414 std::vector<int> const& sizes,
\n+
415 std::vector<int> const& sources,
\n+
416 std::vector<REAL> const& weights,
\n+
417 std::vector<REAL> const& duWeights,
\n+
418 std::vector<REAL> const& dvWeights,
\n+
419 std::vector<REAL> const& duuWeights,
\n+
420 std::vector<REAL> const& duvWeights,
\n+
421 std::vector<REAL> const& dvvWeights,
\n+
422 bool includeCoarseVerts,
\n+
423 size_t firstOffset);
\n+
424
\n+
425public:
\n+
426
\n+
428 LimitStencilReal<REAL> GetLimitStencil(Index i) const;
\n+
429
\n+
431 LimitStencilReal<REAL> operator[] (Index index) const;
\n+
432
\n+
434 std::vector<REAL> const & GetDuWeights() const {
\n+
435 return _duWeights;
\n+
436 }
\n+
437
\n+
439 std::vector<REAL> const & GetDvWeights() const {
\n+
440 return _dvWeights;
\n+
441 }
\n+
442
\n+
444 std::vector<REAL> const & GetDuuWeights() const {
\n+
445 return _duuWeights;
\n+
446 }
\n+
447
\n+
449 std::vector<REAL> const & GetDuvWeights() const {
\n+
450 return _duvWeights;
\n+
451 }
\n+
452
\n+
454 std::vector<REAL> const & GetDvvWeights() const {
\n+
455 return _dvvWeights;
\n+
456 }
\n+
457
\n+
475 template <class T, class U>
\n+
476 void UpdateDerivs(T const & srcValues, U & uderivs, U & vderivs,
\n+
477 int start=-1, int end=-1) const {
\n+
478
\n+
479 this->update(srcValues, uderivs, _duWeights, start, end);
\n+
480 this->update(srcValues, vderivs, _dvWeights, start, end);
\n+
481 }
\n+
482
\n+
483 template <class T1, class T2, class U>
\n+
484 void UpdateDerivs(T1 const & srcBase, int numBase, T2 const & srcRef,
\n+
485 U & uderivs, U & vderivs, int start=-1, int end=-1) const {
\n+
486
\n+
487 this->update(srcBase, numBase, srcRef, uderivs, _duWeights, start, end);
\n+
488 this->update(srcBase, numBase, srcRef, vderivs, _dvWeights, start, end);
\n+
489 }
\n+
490
\n+
491 // Pointer interface for backward compatibility
\n+
492 template <class T, class U>
\n+
493 void UpdateDerivs(T const *src, U *uderivs, U *vderivs,
\n+
494 int start=-1, int end=-1) const {
\n+
495
\n+
496 this->update(src, uderivs, _duWeights, start, end);
\n+
497 this->update(src, vderivs, _dvWeights, start, end);
\n+
498 }
\n+
499 template <class T1, class T2, class U>
\n+
500 void UpdateDerivs(T1 const *srcBase, int numBase, T2 const *srcRef,
\n+
501 U *uderivs, U *vderivs, int start=-1, int end=-1) const {
\n+
502
\n+
503 this->update(srcBase, numBase, srcRef, uderivs, _duWeights, start, end);
\n+
504 this->update(srcBase, numBase, srcRef, vderivs, _dvWeights, start, end);
\n+
505 }
\n+
506
\n+
527 template <class T, class U>
\n+
528 void Update2ndDerivs(T const & srcValues,
\n+
529 U & uuderivs, U & uvderivs, U & vvderivs,
\n+
530 int start=-1, int end=-1) const {
\n+
531
\n+
532 this->update(srcValues, uuderivs, _duuWeights, start, end);
\n+
533 this->update(srcValues, uvderivs, _duvWeights, start, end);
\n+
534 this->update(srcValues, vvderivs, _dvvWeights, start, end);
\n+
535 }
\n+
536
\n+
537 template <class T1, class T2, class U>
\n+
538 void Update2ndDerivs(T1 const & srcBase, int numBase, T2 const & srcRef,
\n+
539 U & uuderivs, U & uvderivs, U & vvderivs, int start=-1, int end=-1) const {
\n+
540
\n+
541 this->update(srcBase, numBase, srcRef, uuderivs, _duuWeights, start, end);
\n+
542 this->update(srcBase, numBase, srcRef, uvderivs, _duvWeights, start, end);
\n+
543 this->update(srcBase, numBase, srcRef, vvderivs, _dvvWeights, start, end);
\n+
544 }
\n+
545
\n+
546 // Pointer interface for backward compatibility
\n+
547 template <class T, class U>
\n+
548 void Update2ndDerivs(T const *src, T *uuderivs, U *uvderivs, U *vvderivs,
\n+
549 int start=-1, int end=-1) const {
\n+
550
\n+
551 this->update(src, uuderivs, _duuWeights, start, end);
\n+
552 this->update(src, uvderivs, _duvWeights, start, end);
\n+
553 this->update(src, vvderivs, _dvvWeights, start, end);
\n+
554 }
\n+
555 template <class T1, class T2, class U>
\n+
556 void Update2ndDerivs(T1 const *srcBase, int numBase, T2 const *srcRef,
\n+
557 U *uuderivs, U *uvderivs, U *vvderivs, int start=-1, int end=-1) const {
\n+
558
\n+
559 this->update(srcBase, numBase, srcRef, uuderivs, _duuWeights, start, end);
\n+
560 this->update(srcBase, numBase, srcRef, uvderivs, _duvWeights, start, end);
\n+
561 this->update(srcBase, numBase, srcRef, vvderivs, _dvvWeights, start, end);
\n+
562 }
\n+
563
\n+
565 void Clear();
\n+
566
\n+
567private:
\n+
568 friend class LimitStencilTableFactoryReal<REAL>;
\n+
569
\n+
570 // Resize the table arrays (factory helper)
\n+
571 void resize(int nstencils, int nelems);
\n+
572
\n+
573private:
\n+
574 std::vector<REAL> _duWeights, // u derivative limit stencil weights
\n+
575 _dvWeights, // v derivative limit stencil weights
\n+
576 _duuWeights, // uu derivative limit stencil weights
\n+
577 _duvWeights, // uv derivative limit stencil weights
\n+
578 _dvvWeights; // vv derivative limit stencil weights
\n+
579};
\n+
580
\n+
583class LimitStencilTable : public LimitStencilTableReal<float> {
\n+
584protected:
\n+
585 typedef LimitStencilTableReal<float> BaseTable;
\n+
586
\n+
587public:
\n+
588 LimitStencil GetLimitStencil(Index index) const {
\n+
589 return LimitStencil(BaseTable::GetLimitStencil(index));
\n+
590 }
\n+
591 LimitStencil operator[] (Index index) const {
\n+
592 return LimitStencil(BaseTable::GetLimitStencil(index));
\n+
593 }
\n+
594
\n+
595protected:
\n+
596 LimitStencilTable(int numControlVerts,
\n+
597 std::vector<int> const& offsets,
\n+
598 std::vector<int> const& sizes,
\n+
599 std::vector<int> const& sources,
\n+
600 std::vector<float> const& weights,
\n+
601 std::vector<float> const& duWeights,
\n+
602 std::vector<float> const& dvWeights,
\n+
603 std::vector<float> const& duuWeights,
\n+
604 std::vector<float> const& duvWeights,
\n+
605 std::vector<float> const& dvvWeights,
\n+
606 bool includeCoarseVerts,
\n+
607 size_t firstOffset)
\n+
608 : BaseTable(numControlVerts,
\n+
609 offsets, sizes, sources, weights,
\n+
610 duWeights, dvWeights, duuWeights, duvWeights, dvvWeights,
\n+
611 includeCoarseVerts, firstOffset) { }
\n+
612};
\n+
613
\n+
614
\n+
615// Update values by applying cached stencil weights to new control values
\n+
616template <typename REAL>
\n+
617template <class T1, class T2, class U> void
\n+
618StencilTableReal<REAL>::update(T1 const &srcBase, int numBase,
\n+
619 T2 const &srcRef, U &dstValues,
\n+
620 std::vector<REAL> const &valueWeights, Index start, Index end) const {
\n+
621
\n+
622 int const * sizes = &_sizes.at(0);
\n+
623 Index const * indices = &_indices.at(0);
\n+
624 REAL const * weights = &valueWeights.at(0);
\n+
625
\n+
626 if (start > 0) {
\n+
627 assert(start < (Index)_offsets.size());
\n+
628 sizes += start;
\n+
629 indices += _offsets[start];
\n+
630 weights += _offsets[start];
\n+
631 } else {
\n+
632 start = 0;
\n+
633 }
\n+
634
\n+
635 int nstencils = ((end < start) ? GetNumStencils() : end) - start;
\n+
636
\n+
637 for (int i = 0; i < nstencils; ++i, ++sizes) {
\n+
638 dstValues[start + i].Clear();
\n+
639 for (int j = 0; j < *sizes; ++j, ++indices, ++weights) {
\n+
640 if (*indices < numBase) {
\n+
641 dstValues[start + i].AddWithWeight(srcBase[*indices], *weights);
\n+
642 } else {
\n+
643 dstValues[start + i].AddWithWeight(srcRef[*indices - numBase], *weights);
\n+
644 }
\n+
645 }
\n+
646 }
\n+
647}
\n+
648template <typename REAL>
\n+
649template <class T, class U> void
\n+
650StencilTableReal<REAL>::update(T const &srcValues, U &dstValues,
\n+
651 std::vector<REAL> const &valueWeights, Index start, Index end) const {
\n+
652
\n+
653 int const * sizes = &_sizes.at(0);
\n+
654 Index const * indices = &_indices.at(0);
\n+
655 REAL const * weights = &valueWeights.at(0);
\n+
656
\n+
657 if (start > 0) {
\n+
658 assert(start < (Index)_offsets.size());
\n+
659 sizes += start;
\n+
660 indices += _offsets[start];
\n+
661 weights += _offsets[start];
\n+
662 } else {
\n+
663 start = 0;
\n+
664 }
\n+
665
\n+
666 int nstencils = ((end < start) ? GetNumStencils() : end) - start;
\n+
667
\n+
668 for (int i = 0; i < nstencils; ++i, ++sizes) {
\n+
669 dstValues[start + i].Clear();
\n+
670 for (int j = 0; j < *sizes; ++j, ++indices, ++weights) {
\n+
671 dstValues[start + i].AddWithWeight(srcValues[*indices], *weights);
\n+
672 }
\n+
673 }
\n+
674}
\n+
675
\n+
676template <typename REAL>
\n+
677inline void
\n+
678StencilTableReal<REAL>::generateOffsets() {
\n+
679 Index offset=0;
\n+
680 int noffsets = (int)_sizes.size();
\n+
681 _offsets.resize(noffsets);
\n+
682 for (int i=0; i<(int)_sizes.size(); ++i ) {
\n+
683 _offsets[i]=offset;
\n+
684 offset+=_sizes[i];
\n+
685 }
\n+
686}
\n+
687
\n+
688template <typename REAL>
\n+
689inline void
\n+
690StencilTableReal<REAL>::resize(int nstencils, int nelems) {
\n+
691 _sizes.resize(nstencils);
\n+
692 _indices.resize(nelems);
\n+
693 _weights.resize(nelems);
\n+
694}
\n+
695
\n+
696template <typename REAL>
\n+
697inline void
\n+
698StencilTableReal<REAL>::reserve(int nstencils, int nelems) {
\n+
699 _sizes.reserve(nstencils);
\n+
700 _indices.reserve(nelems);
\n+
701 _weights.reserve(nelems);
\n+
702}
\n+
703
\n+
704template <typename REAL>
\n+
705inline void
\n+
706StencilTableReal<REAL>::shrinkToFit() {
\n+
707 std::vector<int>(_sizes).swap(_sizes);
\n+
708 std::vector<Index>(_indices).swap(_indices);
\n+
709 std::vector<REAL>(_weights).swap(_weights);
\n+
710}
\n+
711
\n+
712template <typename REAL>
\n+
713inline void
\n+
714StencilTableReal<REAL>::finalize() {
\n+
715 shrinkToFit();
\n+
716 generateOffsets();
\n+
717}
\n+
718
\n+
719// Returns a Stencil at index i in the table
\n+
720template <typename REAL>
\n+
721inline StencilReal<REAL>
\n+
722StencilTableReal<REAL>::GetStencil(Index i) const {
\n+
723 assert((! _offsets.empty()) && i<(int)_offsets.size());
\n+
724
\n+
725 Index ofs = _offsets[i];
\n+
726
\n+
727 return StencilReal<REAL>(const_cast<int*>(&_sizes[i]),
\n+
728 const_cast<Index*>(&_indices[ofs]),
\n+
729 const_cast<REAL*>(&_weights[ofs]));
\n+
730}
\n+
731
\n+
732template <typename REAL>
\n+
733inline StencilReal<REAL>
\n+
734StencilTableReal<REAL>::operator[] (Index index) const {
\n+
735 return GetStencil(index);
\n+
736}
\n+
737
\n+
738template <typename REAL>
\n+
739inline void
\n+
740LimitStencilTableReal<REAL>::resize(int nstencils, int nelems) {
\n+
741 StencilTableReal<REAL>::resize(nstencils, nelems);
\n+
742 _duWeights.resize(nelems);
\n+
743 _dvWeights.resize(nelems);
\n+
744}
\n+
745
\n+
746// Returns a LimitStencil at index i in the table
\n+
747template <typename REAL>
\n+
748inline LimitStencilReal<REAL>
\n+
749LimitStencilTableReal<REAL>::GetLimitStencil(Index i) const {
\n+
750 assert((! this->GetOffsets().empty()) && i<(int)this->GetOffsets().size());
\n+
751
\n+
752 Index ofs = this->GetOffsets()[i];
\n+
753
\n+
754 if (!_duWeights.empty() && !_dvWeights.empty() &&
\n+
755 !_duuWeights.empty() && !_duvWeights.empty() && !_dvvWeights.empty()) {
\n+
756 return LimitStencilReal<REAL>(
\n+
757 const_cast<int *>(&this->GetSizes()[i]),
\n+
758 const_cast<Index *>(&this->GetControlIndices()[ofs]),
\n+
759 const_cast<REAL *>(&this->GetWeights()[ofs]),
\n+
760 const_cast<REAL *>(&GetDuWeights()[ofs]),
\n+
761 const_cast<REAL *>(&GetDvWeights()[ofs]),
\n+
762 const_cast<REAL *>(&GetDuuWeights()[ofs]),
\n+
763 const_cast<REAL *>(&GetDuvWeights()[ofs]),
\n+
764 const_cast<REAL *>(&GetDvvWeights()[ofs]) );
\n+
765 } else if (!_duWeights.empty() && !_dvWeights.empty()) {
\n+
766 return LimitStencilReal<REAL>(
\n+
767 const_cast<int *>(&this->GetSizes()[i]),
\n+
768 const_cast<Index *>(&this->GetControlIndices()[ofs]),
\n+
769 const_cast<REAL *>(&this->GetWeights()[ofs]),
\n+
770 const_cast<REAL *>(&GetDuWeights()[ofs]),
\n+
771 const_cast<REAL *>(&GetDvWeights()[ofs]) );
\n+
772 } else {
\n+
773 return LimitStencilReal<REAL>(
\n+
774 const_cast<int *>(&this->GetSizes()[i]),
\n+
775 const_cast<Index *>(&this->GetControlIndices()[ofs]),
\n+
776 const_cast<REAL *>(&this->GetWeights()[ofs]) );
\n+
777 }
\n+
778}
\n+
779
\n+
780template <typename REAL>
\n+
781inline LimitStencilReal<REAL>
\n+
782LimitStencilTableReal<REAL>::operator[] (Index index) const {
\n+
783 return GetLimitStencil(index);
\n+
784}
\n+
785
\n+
786} // end namespace Far
\n+
787
\n+
788} // end namespace OPENSUBDIV_VERSION
\n+
789using namespace OPENSUBDIV_VERSION;
\n+
790
\n+
791} // end namespace OpenSubdiv
\n+
792
\n+
793#endif // OPENSUBDIV3_FAR_STENCILTABLE_H
\n \n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor
A simple reference to raw topology data for use with TopologyRefinerFactory.
Definition: topologyDescriptor.h:49
\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::assignComponentTopology
static bool assignComponentTopology(TopologyRefiner &newRefiner, MESH const &mesh)
Specify the relationships between vertices, faces, etc. ie the face-vertices, vertex-faces,...
Definition: topologyRefinerFactory.h:621
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::reportInvalidTopology
static void reportInvalidTopology(TopologyError errCode, char const *msg, MESH const &mesh)
(Optional) Control run-time topology validation and error reporting
Definition: topologyRefinerFactory.h:707
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::assignComponentTags
static bool assignComponentTags(TopologyRefiner &newRefiner, MESH const &mesh)
(Optional) Specify edge or vertex sharpness or face holes
Definition: topologyRefinerFactory.h:690
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::resizeComponentTopology
static bool resizeComponentTopology(TopologyRefiner &newRefiner, MESH const &mesh)
Specify the number of vertices, faces, face-vertices, etc.
Definition: topologyRefinerFactory.h:580
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::assignFaceVaryingTopology
static bool assignFaceVaryingTopology(TopologyRefiner &newRefiner, MESH const &mesh)
(Optional) Specify face-varying data per face
Definition: topologyRefinerFactory.h:674
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner
Stores topology data for a specified set of refinement options.
Definition: topologyRefiner.h:51
\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::StencilReal
StencilReal(int *size, Index *indices, REAL *weights)
Constructor.
Definition: stencilTable.h:67
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::Next
void Next()
Advance to the next stencil in the table.
Definition: stencilTable.h:98
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::StencilReal
StencilReal(StencilReal const &other)
Copy constructor.
Definition: stencilTable.h:71
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetVertexIndices
Index const * GetVertexIndices() const
Returns the control vertices' indices.
Definition: stencilTable.h:88
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetSize
int GetSize() const
Returns the size of the stencil.
Definition: stencilTable.h:78
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetSizePtr
int * GetSizePtr() const
Returns the size of the stencil as a pointer.
Definition: stencilTable.h:83
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetWeights
REAL const * GetWeights() const
Returns the interpolation weights.
Definition: stencilTable.h:93
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil
Vertex stencil class wrapping the template for compatibility.
Definition: stencilTable.h:116
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil::Stencil
Stencil(int *size, Index *indices, float *weights)
Definition: stencilTable.h:123
\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumControlVertices
int GetNumControlVertices() const
Returns the number of control vertices indexed in the table.
Definition: stencilTable.h:161
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::update
void update(T1 const &srcBase, int numBase, T2 const &srcRef, U &dstValues, std::vector< REAL > const &valueWeights, Index start, Index end) const
Definition: stencilTable.h:618
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues
void UpdateValues(T1 const &srcBase, int numBase, T2 const &srcRef, U &dstValues, Index start=-1, Index end=-1) const
Definition: stencilTable.h:210
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetControlIndices
std::vector< Index > const & GetControlIndices() const
Returns the indices of the control vertices.
Definition: stencilTable.h:179
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::operator[]
StencilReal< REAL > operator[](Index index) const
Returns the stencil at index i in the table.
Definition: stencilTable.h:734
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::update
void update(T const &srcValues, U &dstValues, std::vector< REAL > const &valueWeights, Index start, Index end) const
Definition: stencilTable.h:650
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetSizes
std::vector< int > const & GetSizes() const
Returns the number of control vertices of each stencil in the table.
Definition: stencilTable.h:169
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumStencils
int GetNumStencils() const
Returns the number of stencils in the table.
Definition: stencilTable.h:156
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetOffsets
std::vector< Index > const & GetOffsets() const
Returns the offset to a given stencil (factory may leave empty)
Definition: stencilTable.h:174
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetStencil
StencilReal< REAL > GetStencil(Index i) const
Returns a Stencil at index i in the table.
Definition: stencilTable.h:722
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::Clear
void Clear()
Clears the stencils from the table.
\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetWeights
std::vector< REAL > const & GetWeights() const
Returns the stencil interpolation weights.
Definition: stencilTable.h:184
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues
void UpdateValues(T1 const *srcBase, int numBase, T2 const *srcRef, U *dst, Index start=-1, Index end=-1) const
Definition: stencilTable.h:221
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::StencilTableReal
StencilTableReal(int numControlVerts, std::vector< int > const &offsets, std::vector< int > const &sizes, std::vector< int > const &sources, std::vector< REAL > const &weights, bool includeCoarseVerts, size_t firstOffset)
\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues
void UpdateValues(T const *src, U *dst, Index start=-1, Index end=-1) const
Definition: stencilTable.h:217
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues
void UpdateValues(T const &srcValues, U &dstValues, Index start=-1, Index end=-1) const
Updates point values based on the control values.
Definition: stencilTable.h:205
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::StencilTable
StencilTable(int numControlVerts, std::vector< int > const &offsets, std::vector< int > const &sizes, std::vector< int > const &sources, std::vector< float > const &weights, bool includeCoarseVerts, size_t firstOffset)
Definition: stencilTable.h:288
\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuvWeights
REAL const * GetDuvWeights() const
Returns the uv derivative weights.
Definition: stencilTable.h:356
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDvWeights
REAL const * GetDvWeights() const
Returns the v derivative weights.
Definition: stencilTable.h:346
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::Next
void Next()
Advance to the next stencil in the table.
Definition: stencilTable.h:366
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDvvWeights
REAL const * GetDvvWeights() const
Returns the vv derivative weights.
Definition: stencilTable.h:361
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::LimitStencilReal
LimitStencilReal(int *size, Index *indices, REAL *weights, REAL *duWeights=0, REAL *dvWeights=0, REAL *duuWeights=0, REAL *duvWeights=0, REAL *dvvWeights=0)
Constructor.
Definition: stencilTable.h:324
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuWeights
REAL const * GetDuWeights() const
Returns the u derivative weights.
Definition: stencilTable.h:341
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuuWeights
REAL const * GetDuuWeights() const
Returns the uu derivative weights.
Definition: stencilTable.h:351
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil
Limit point stencil class wrapping the template for compatibility.
Definition: stencilTable.h:392
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil::LimitStencil
LimitStencil(int *size, Index *indices, float *weights, float *duWeights=0, float *dvWeights=0, float *duuWeights=0, float *duvWeights=0, float *dvvWeights=0)
Definition: stencilTable.h:398
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuuWeights
std::vector< REAL > const & GetDuuWeights() const
Returns the 'uu' derivative stencil interpolation weights.
Definition: stencilTable.h:444
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::operator[]
LimitStencilReal< REAL > operator[](Index index) const
Returns the limit stencil at index i in the table.
Definition: stencilTable.h:782
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDvvWeights
std::vector< REAL > const & GetDvvWeights() const
Returns the 'vv' derivative stencil interpolation weights.
Definition: stencilTable.h:454
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs
void UpdateDerivs(T const *src, U *uderivs, U *vderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:493
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs
void Update2ndDerivs(T const *src, T *uuderivs, U *uvderivs, U *vvderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:548
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDvWeights
std::vector< REAL > const & GetDvWeights() const
Returns the 'v' derivative stencil interpolation weights.
Definition: stencilTable.h:439
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs
void Update2ndDerivs(T1 const *srcBase, int numBase, T2 const *srcRef, U *uuderivs, U *uvderivs, U *vvderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:556
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs
void UpdateDerivs(T const &srcValues, U &uderivs, U &vderivs, int start=-1, int end=-1) const
Updates derivative values based on the control values.
Definition: stencilTable.h:476
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetLimitStencil
LimitStencilReal< REAL > GetLimitStencil(Index i) const
Returns a LimitStencil at index i in the table.
Definition: stencilTable.h:749
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::LimitStencilTableReal
LimitStencilTableReal(int numControlVerts, std::vector< int > const &offsets, std::vector< int > const &sizes, std::vector< int > const &sources, std::vector< REAL > const &weights, std::vector< REAL > const &duWeights, std::vector< REAL > const &dvWeights, std::vector< REAL > const &duuWeights, std::vector< REAL > const &duvWeights, std::vector< REAL > const &dvvWeights, bool includeCoarseVerts, size_t firstOffset)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs
void Update2ndDerivs(T const &srcValues, U &uuderivs, U &uvderivs, U &vvderivs, int start=-1, int end=-1) const
Updates 2nd derivative values based on the control values.
Definition: stencilTable.h:528
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuWeights
std::vector< REAL > const & GetDuWeights() const
Returns the 'u' derivative stencil interpolation weights.
Definition: stencilTable.h:434
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Clear
void Clear()
Clears the stencils from the table.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs
void Update2ndDerivs(T1 const &srcBase, int numBase, T2 const &srcRef, U &uuderivs, U &uvderivs, U &vvderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:538
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuvWeights
std::vector< REAL > const & GetDuvWeights() const
Returns the 'uv' derivative stencil interpolation weights.
Definition: stencilTable.h:449
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs
void UpdateDerivs(T1 const &srcBase, int numBase, T2 const &srcRef, U &uderivs, U &vderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:484
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs
void UpdateDerivs(T1 const *srcBase, int numBase, T2 const *srcRef, U *uderivs, U *vderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:500
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable::LimitStencilTable
LimitStencilTable(int numControlVerts, std::vector< int > const &offsets, std::vector< int > const &sizes, std::vector< int > const &sources, std::vector< float > const &weights, std::vector< float > const &duWeights, std::vector< float > const &dvWeights, std::vector< float > const &duuWeights, std::vector< float > const &duvWeights, std::vector< float > const &dvvWeights, bool includeCoarseVerts, size_t firstOffset)
Definition: stencilTable.h:596
\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-topologyDescriptor.h\n+stencilTable.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2014 DreamWorks Animation LLC.\n+ 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -29,208 +29,993 @@\n 17//\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n- 24#ifndef OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H\n- 25#define OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H\n- 26\n- 27#include \"../version.h\"\n- 28\n- 29#include \"../far/topologyRefiner.h\"\n- 30#include \"../far/topologyRefinerFactory.h\"\n- 31#include \"../far/error.h\"\n- 32\n- 33#include \n- 34\n- 35namespace OpenSubdiv {\n- 36namespace OPENSUBDIV_VERSION {\n- 37\n- 38namespace Far {\n+ 24\n+ 25#ifndef OPENSUBDIV3_FAR_STENCILTABLE_H\n+ 26#define OPENSUBDIV3_FAR_STENCILTABLE_H\n+ 27\n+ 28#include \"../version.h\"\n+ 29\n+ 30#include \"../far/types.h\"\n+ 31\n+ 32#include \n+ 33#include \n+ 34#include \n+ 35#include \n+ 36\n+ 37namespace OpenSubdiv {\n+ 38namespace OPENSUBDIV_VERSION {\n 39\n-49struct TopologyDescriptor {\n- 50\n-51 int numVertices,\n-52 numFaces;\n- 53\n-54 int const * numVertsPerFace;\n-55 Index const * vertIndicesPerFace;\n- 56\n-57 int numCreases;\n-58 Index const * creaseVertexIndexPairs;\n-59 float const * creaseWeights;\n- 60\n-61 int numCorners;\n-62 Index const * cornerVertexIndices;\n-63 float const * cornerWeights;\n- 64\n-65 int numHoles;\n-66 Index const * holeIndices;\n- 67\n-68 bool isLeftHanded;\n+ 40namespace Far {\n+ 41\n+ 42// Forward declarations for friends:\n+ 43class PatchTableBuilder;\n+ 44\n+ 45template class StencilTableFactoryReal;\n+ 46template class LimitStencilTableFactoryReal;\n+ 47\n+ 52template \n+53class StencilReal {\n+ 54public:\n+ 55\n+57 StencilReal() {}\n+ 58\n+67 StencilReal(int * size, Index * indices, REAL * weights)\n+ 68 : _size(size), _indices(indices), _weights(weights) { }\n 69\n- 70 // Face-varying data channel -- value indices correspond to vertex indices,\n- 71 // i.e. one for every vertex of every face:\n- 72 //\n-73 struct FVarChannel {\n- 74\n-75 int numValues;\n-76 Index const * valueIndices;\n- 77\n-78 FVarChannel() : numValues(0), valueIndices(0) { }\n- 79 };\n- 80\n-81 int numFVarChannels;\n-82 FVarChannel const * fvarChannels;\n- 83\n-84 TopologyDescriptor();\n- 85};\n+71 StencilReal(StencilReal const & other) {\n+ 72 _size = other._size;\n+ 73 _indices = other._indices;\n+ 74 _weights = other._weights;\n+ 75 }\n+ 76\n+78 int GetSize() const {\n+ 79 return *_size;\n+ 80 }\n+ 81\n+83 int * GetSizePtr() const {\n+ 84 return _size;\n+ 85 }\n 86\n- 87\n- 88//\n- 89// Forward declarations of required\n-TopologyRefinerFactory\n- 90// specializations (defined internally):\n- 91//\n- 92// @cond EXCLUDE_DOXYGEN\n- 93\n- 94template <>\n- 95bool\n- 96TopologyRefinerFactory::resizeComponentTopology(\n- 97 TopologyRefiner & refiner, TopologyDescriptor const & desc);\n- 98\n- 99template <>\n- 100bool\n- 101TopologyRefinerFactory::assignComponentTopology(\n- 102 TopologyRefiner & refiner, TopologyDescriptor const & desc);\n- 103\n- 104template <>\n- 105bool\n- 106TopologyRefinerFactory::assignComponentTags(\n- 107 TopologyRefiner & refiner, TopologyDescriptor const & desc);\n+88 Index const * GetVertexIndices() const {\n+ 89 return _indices;\n+ 90 }\n+ 91\n+93 REAL const * GetWeights() const {\n+ 94 return _weights;\n+ 95 }\n+ 96\n+98 void Next() {\n+ 99 int stride = *_size;\n+ 100 ++_size;\n+ 101 _indices += stride;\n+ 102 _weights += stride;\n+ 103 }\n+ 104\n+ 105protected:\n+ 106 friend class StencilTableFactoryReal;\n+ 107 friend class LimitStencilTableFactoryReal;\n 108\n- 109template <>\n- 110bool\n- 111TopologyRefinerFactory::assignFaceVaryingTopology(\n- 112 TopologyRefiner & refiner, TopologyDescriptor const & desc);\n+109 int * _size;\n+110 Index * _indices;\n+111 REAL * _weights;\n+ 112};\n 113\n- 114template <>\n- 115void\n- 116TopologyRefinerFactory::reportInvalidTopology(\n- 117 TopologyError errCode, char const * msg, TopologyDescriptor const & desc);\n- 118\n- 119// @endcond\n- 120\n- 121} // end namespace Far\n- 122\n- 123} // end namespace OPENSUBDIV_VERSION\n- 124using namespace OPENSUBDIV_VERSION;\n- 125} // end namespace OpenSubdiv\n+116class Stencil : public StencilReal {\n+ 117protected:\n+118 typedef StencilReal BaseStencil;\n+ 119\n+ 120public:\n+121 Stencil() : BaseStencil() { }\n+122 Stencil(BaseStencil const & other) : BaseStencil(other) { }\n+123 Stencil(int * size, Index * indices, float * weights)\n+ 124 : BaseStencil(size, indices, weights) { }\n+ 125};\n 126\n- 127#endif /* OPENSUBDIV3_FAR_TOPOLOGY_DESCRIPTOR_H */\n+ 127\n+ 140template \n+141class StencilTableReal {\n+ 142protected:\n+143 StencilTableReal(int numControlVerts,\n+ 144 std::vector const& offsets,\n+ 145 std::vector const& sizes,\n+ 146 std::vector const& sources,\n+ 147 std::vector const& weights,\n+ 148 bool includeCoarseVerts,\n+ 149 size_t firstOffset);\n+ 150\n+ 151public:\n+ 152\n+153 virtual ~StencilTableReal() {};\n+ 154\n+156 int GetNumStencils() const {\n+ 157 return (int)_sizes.size();\n+ 158 }\n+ 159\n+161 int GetNumControlVertices() const {\n+ 162 return _numControlVertices;\n+ 163 }\n+ 164\n+166 StencilReal GetStencil(Index i) const;\n+ 167\n+169 std::vector const & GetSizes() const {\n+ 170 return _sizes;\n+ 171 }\n+ 172\n+174 std::vector const & GetOffsets() const {\n+ 175 return _offsets;\n+ 176 }\n+ 177\n+179 std::vector const & GetControlIndices() const {\n+ 180 return _indices;\n+ 181 }\n+ 182\n+184 std::vector const & GetWeights() const {\n+ 185 return _weights;\n+ 186 }\n+ 187\n+189 StencilReal operator[](Index index) const;\n+ 190\n+ 204 template \n+205 void UpdateValues(T const &srcValues, U &dstValues, Index start=-1, Index\n+end=-1) const {\n+ 206 this->update(srcValues, dstValues, _weights, start, end);\n+ 207 }\n+ 208\n+ 209 template \n+210 void UpdateValues(T1 const &srcBase, int numBase, T2 const &srcRef,\n+ 211 U &dstValues, Index start=-1, Index end=-1) const {\n+ 212 this->update(srcBase, numBase, srcRef, dstValues, _weights, start, end);\n+ 213 }\n+ 214\n+ 215 // Pointer interface for backward compatibility\n+ 216 template \n+217 void UpdateValues(T const *src, U *dst, Index start=-1, Index end=-1) const\n+{\n+ 218 this->update(src, dst, _weights, start, end);\n+ 219 }\n+ 220 template \n+221 void UpdateValues(T1 const *srcBase, int numBase, T2 const *srcRef,\n+ 222 U *dst, Index start=-1, Index end=-1) const {\n+ 223 this->update(srcBase, numBase, srcRef, dst, _weights, start, end);\n+ 224 }\n+ 225\n+227 void Clear();\n+ 228\n+ 229protected:\n+ 230\n+ 231 // Update values by applying cached stencil weights to new control values\n+ 232 template \n+233 void update( T const &srcValues, U &dstValues,\n+ 234 std::vector const & valueWeights, Index start, Index end) const;\n+ 235 template \n+236 void update( T1 const &srcBase, int numBase, T2 const &srcRef, U\n+&dstValues,\n+ 237 std::vector const & valueWeights, Index start, Index end) const;\n+ 238\n+ 239 // Populate the offsets table from the stencil sizes in _sizes (factory\n+helper)\n+240 void generateOffsets();\n+ 241\n+ 242 // Resize the table arrays (factory helper)\n+243 void resize(int nstencils, int nelems);\n+ 244\n+ 245 // Reserves the table arrays (factory helper)\n+246 void reserve(int nstencils, int nelems);\n+ 247\n+ 248 // Reallocates the table arrays to remove excess capacity (factory helper)\n+249 void shrinkToFit();\n+ 250\n+ 251 // Performs any final operations on internal tables (factory helper)\n+252 void finalize();\n+ 253\n+ 254protected:\n+255 StencilTableReal() : _numControlVertices(0) {}\n+256 StencilTableReal(int numControlVerts)\n+ 257 : _numControlVertices(numControlVerts)\n+ 258 { }\n+ 259\n+ 260 friend class StencilTableFactoryReal;\n+261 friend class Far::PatchTableBuilder;\n+ 262\n+263 int _numControlVertices; // number of control vertices\n+ 264\n+265 std::vector _sizes; // number of coefficients for each stencil\n+266 std::vector _offsets, // offset to the start of each stencil\n+267 _indices; // indices of contributing coarse vertices\n+268 std::vector _weights; // stencil weight coefficients\n+ 269};\n+ 270\n+273class StencilTable : public StencilTableReal {\n+ 274protected:\n+275 typedef StencilTableReal BaseTable;\n+ 276\n+ 277public:\n+278 Stencil GetStencil(Index index) const {\n+ 279 return Stencil(BaseTable::GetStencil(index));\n+ 280 }\n+281 Stencil operator[](Index index) const {\n+ 282 return Stencil(BaseTable::GetStencil(index));\n+ 283 }\n+ 284\n+ 285protected:\n+286 StencilTable() : BaseTable() { }\n+287 StencilTable(int numControlVerts) : BaseTable(numControlVerts) { }\n+288 StencilTable(int numControlVerts,\n+ 289 std::vector const& offsets,\n+ 290 std::vector const& sizes,\n+ 291 std::vector const& sources,\n+ 292 std::vector const& weights,\n+ 293 bool includeCoarseVerts,\n+ 294 size_t firstOffset)\n+ 295 : BaseTable(numControlVerts, offsets,\n+ 296 sizes, sources, weights, includeCoarseVerts, firstOffset) { }\n+ 297};\n+ 298\n+ 299\n+ 302template \n+303class LimitStencilReal : public StencilReal {\n+ 304public:\n+ 305\n+324 LimitStencilReal( int* size,\n+ 325 Index * indices,\n+ 326 REAL * weights,\n+ 327 REAL * duWeights=0,\n+ 328 REAL * dvWeights=0,\n+ 329 REAL * duuWeights=0,\n+ 330 REAL * duvWeights=0,\n+ 331 REAL * dvvWeights=0)\n+ 332 : StencilReal(size, indices, weights),\n+ 333 _duWeights(duWeights),\n+ 334 _dvWeights(dvWeights),\n+ 335 _duuWeights(duuWeights),\n+ 336 _duvWeights(duvWeights),\n+ 337 _dvvWeights(dvvWeights) {\n+ 338 }\n+ 339\n+341 REAL const * GetDuWeights() const {\n+ 342 return _duWeights;\n+ 343 }\n+ 344\n+346 REAL const * GetDvWeights() const {\n+ 347 return _dvWeights;\n+ 348 }\n+ 349\n+351 REAL const * GetDuuWeights() const {\n+ 352 return _duuWeights;\n+ 353 }\n+ 354\n+356 REAL const * GetDuvWeights() const {\n+ 357 return _duvWeights;\n+ 358 }\n+ 359\n+361 REAL const * GetDvvWeights() const {\n+ 362 return _dvvWeights;\n+ 363 }\n+ 364\n+366 void Next() {\n+ 367 int stride = *this->_size;\n+ 368 ++this->_size;\n+ 369 this->_indices += stride;\n+ 370 this->_weights += stride;\n+ 371 if (_duWeights) _duWeights += stride;\n+ 372 if (_dvWeights) _dvWeights += stride;\n+ 373 if (_duuWeights) _duuWeights += stride;\n+ 374 if (_duvWeights) _duvWeights += stride;\n+ 375 if (_dvvWeights) _dvvWeights += stride;\n+ 376 }\n+ 377\n+ 378private:\n+ 379\n+ 380 friend class StencilTableFactoryReal;\n+ 381 friend class LimitStencilTableFactoryReal;\n+ 382\n+ 383 REAL * _duWeights, // pointer to stencil u derivative limit weights\n+ 384 * _dvWeights, // pointer to stencil v derivative limit weights\n+ 385 * _duuWeights, // pointer to stencil uu derivative limit weights\n+ 386 * _duvWeights, // pointer to stencil uv derivative limit weights\n+ 387 * _dvvWeights; // pointer to stencil vv derivative limit weights\n+ 388};\n+ 389\n+392class LimitStencil : public LimitStencilReal {\n+ 393protected:\n+394 typedef LimitStencilReal BaseStencil;\n+ 395\n+ 396public:\n+397 LimitStencil(BaseStencil const & other) : BaseStencil(other) { }\n+398 LimitStencil(int* size, Index * indices, float * weights,\n+ 399 float * duWeights=0, float * dvWeights=0,\n+ 400 float * duuWeights=0, float * duvWeights=0, float * dvvWeights=0)\n+ 401 : BaseStencil(size, indices, weights,\n+ 402 duWeights, dvWeights, duuWeights, duvWeights, dvvWeights) { }\n+ 403};\n+ 404\n+ 405\n+ 408template \n+409class LimitStencilTableReal : public StencilTableReal {\n+ 410protected:\n+411 LimitStencilTableReal(\n+ 412 int numControlVerts,\n+ 413 std::vector const& offsets,\n+ 414 std::vector const& sizes,\n+ 415 std::vector const& sources,\n+ 416 std::vector const& weights,\n+ 417 std::vector const& duWeights,\n+ 418 std::vector const& dvWeights,\n+ 419 std::vector const& duuWeights,\n+ 420 std::vector const& duvWeights,\n+ 421 std::vector const& dvvWeights,\n+ 422 bool includeCoarseVerts,\n+ 423 size_t firstOffset);\n+ 424\n+ 425public:\n+ 426\n+428 LimitStencilReal GetLimitStencil(Index i) const;\n+ 429\n+431 LimitStencilReal operator[](Index index) const;\n+ 432\n+434 std::vector const & GetDuWeights() const {\n+ 435 return _duWeights;\n+ 436 }\n+ 437\n+439 std::vector const & GetDvWeights() const {\n+ 440 return _dvWeights;\n+ 441 }\n+ 442\n+444 std::vector const & GetDuuWeights() const {\n+ 445 return _duuWeights;\n+ 446 }\n+ 447\n+449 std::vector const & GetDuvWeights() const {\n+ 450 return _duvWeights;\n+ 451 }\n+ 452\n+454 std::vector const & GetDvvWeights() const {\n+ 455 return _dvvWeights;\n+ 456 }\n+ 457\n+ 475 template \n+476 void UpdateDerivs(T const & srcValues, U & uderivs, U & vderivs,\n+ 477 int start=-1, int end=-1) const {\n+ 478\n+ 479 this->update(srcValues, uderivs, _duWeights, start, end);\n+ 480 this->update(srcValues, vderivs, _dvWeights, start, end);\n+ 481 }\n+ 482\n+ 483 template \n+484 void UpdateDerivs(T1 const & srcBase, int numBase, T2 const & srcRef,\n+ 485 U & uderivs, U & vderivs, int start=-1, int end=-1) const {\n+ 486\n+ 487 this->update(srcBase, numBase, srcRef, uderivs, _duWeights, start, end);\n+ 488 this->update(srcBase, numBase, srcRef, vderivs, _dvWeights, start, end);\n+ 489 }\n+ 490\n+ 491 // Pointer interface for backward compatibility\n+ 492 template \n+493 void UpdateDerivs(T const *src, U *uderivs, U *vderivs,\n+ 494 int start=-1, int end=-1) const {\n+ 495\n+ 496 this->update(src, uderivs, _duWeights, start, end);\n+ 497 this->update(src, vderivs, _dvWeights, start, end);\n+ 498 }\n+ 499 template \n+500 void UpdateDerivs(T1 const *srcBase, int numBase, T2 const *srcRef,\n+ 501 U *uderivs, U *vderivs, int start=-1, int end=-1) const {\n+ 502\n+ 503 this->update(srcBase, numBase, srcRef, uderivs, _duWeights, start, end);\n+ 504 this->update(srcBase, numBase, srcRef, vderivs, _dvWeights, start, end);\n+ 505 }\n+ 506\n+ 527 template \n+528 void Update2ndDerivs(T const & srcValues,\n+ 529 U & uuderivs, U & uvderivs, U & vvderivs,\n+ 530 int start=-1, int end=-1) const {\n+ 531\n+ 532 this->update(srcValues, uuderivs, _duuWeights, start, end);\n+ 533 this->update(srcValues, uvderivs, _duvWeights, start, end);\n+ 534 this->update(srcValues, vvderivs, _dvvWeights, start, end);\n+ 535 }\n+ 536\n+ 537 template \n+538 void Update2ndDerivs(T1 const & srcBase, int numBase, T2 const & srcRef,\n+ 539 U & uuderivs, U & uvderivs, U & vvderivs, int start=-1, int end=-1) const\n+{\n+ 540\n+ 541 this->update(srcBase, numBase, srcRef, uuderivs, _duuWeights, start, end);\n+ 542 this->update(srcBase, numBase, srcRef, uvderivs, _duvWeights, start, end);\n+ 543 this->update(srcBase, numBase, srcRef, vvderivs, _dvvWeights, start, end);\n+ 544 }\n+ 545\n+ 546 // Pointer interface for backward compatibility\n+ 547 template \n+548 void Update2ndDerivs(T const *src, T *uuderivs, U *uvderivs, U *vvderivs,\n+ 549 int start=-1, int end=-1) const {\n+ 550\n+ 551 this->update(src, uuderivs, _duuWeights, start, end);\n+ 552 this->update(src, uvderivs, _duvWeights, start, end);\n+ 553 this->update(src, vvderivs, _dvvWeights, start, end);\n+ 554 }\n+ 555 template \n+556 void Update2ndDerivs(T1 const *srcBase, int numBase, T2 const *srcRef,\n+ 557 U *uuderivs, U *uvderivs, U *vvderivs, int start=-1, int end=-1) const {\n+ 558\n+ 559 this->update(srcBase, numBase, srcRef, uuderivs, _duuWeights, start, end);\n+ 560 this->update(srcBase, numBase, srcRef, uvderivs, _duvWeights, start, end);\n+ 561 this->update(srcBase, numBase, srcRef, vvderivs, _dvvWeights, start, end);\n+ 562 }\n+ 563\n+565 void Clear();\n+ 566\n+ 567private:\n+ 568 friend class LimitStencilTableFactoryReal;\n+ 569\n+ 570 // Resize the table arrays (factory helper)\n+ 571 void resize(int nstencils, int nelems);\n+ 572\n+ 573private:\n+ 574 std::vector _duWeights, // u derivative limit stencil weights\n+ 575 _dvWeights, // v derivative limit stencil weights\n+ 576 _duuWeights, // uu derivative limit stencil weights\n+ 577 _duvWeights, // uv derivative limit stencil weights\n+ 578 _dvvWeights; // vv derivative limit stencil weights\n+ 579};\n+ 580\n+583class LimitStencilTable : public LimitStencilTableReal {\n+ 584protected:\n+585 typedef LimitStencilTableReal BaseTable;\n+ 586\n+ 587public:\n+588 LimitStencil GetLimitStencil(Index index) const {\n+ 589 return LimitStencil(BaseTable::GetLimitStencil(index));\n+ 590 }\n+591 LimitStencil operator[](Index index) const {\n+ 592 return LimitStencil(BaseTable::GetLimitStencil(index));\n+ 593 }\n+ 594\n+ 595protected:\n+596 LimitStencilTable(int numControlVerts,\n+ 597 std::vector const& offsets,\n+ 598 std::vector const& sizes,\n+ 599 std::vector const& sources,\n+ 600 std::vector const& weights,\n+ 601 std::vector const& duWeights,\n+ 602 std::vector const& dvWeights,\n+ 603 std::vector const& duuWeights,\n+ 604 std::vector const& duvWeights,\n+ 605 std::vector const& dvvWeights,\n+ 606 bool includeCoarseVerts,\n+ 607 size_t firstOffset)\n+ 608 : BaseTable(numControlVerts,\n+ 609 offsets, sizes, sources, weights,\n+ 610 duWeights, dvWeights, duuWeights, duvWeights, dvvWeights,\n+ 611 includeCoarseVerts, firstOffset) { }\n+ 612};\n+ 613\n+ 614\n+ 615// Update values by applying cached stencil weights to new control values\n+ 616template \n+ 617template void\n+618StencilTableReal::update(T1 const &srcBase, int numBase,\n+ 619 T2 const &srcRef, U &dstValues,\n+ 620 std::vector const &valueWeights, Index start, Index end) const {\n+ 621\n+ 622 int const * sizes = &_sizes.at(0);\n+ 623 Index const * indices = &_indices.at(0);\n+ 624 REAL const * weights = &valueWeights.at(0);\n+ 625\n+ 626 if (start > 0) {\n+ 627 assert(start < (Index)_offsets.size());\n+ 628 sizes += start;\n+ 629 indices += _offsets[start];\n+ 630 weights += _offsets[start];\n+ 631 } else {\n+ 632 start = 0;\n+ 633 }\n+ 634\n+ 635 int nstencils = ((end < start) ? GetNumStencils() : end) - start;\n+ 636\n+ 637 for (int i = 0; i < nstencils; ++i, ++sizes) {\n+ 638 dstValues[start + i].Clear();\n+ 639 for (int j = 0; j < *sizes; ++j, ++indices, ++weights) {\n+ 640 if (*indices < numBase) {\n+ 641 dstValues[start + i].AddWithWeight(srcBase[*indices], *weights);\n+ 642 } else {\n+ 643 dstValues[start + i].AddWithWeight(srcRef[*indices - numBase], *weights);\n+ 644 }\n+ 645 }\n+ 646 }\n+ 647}\n+ 648template \n+ 649template void\n+650StencilTableReal::update(T const &srcValues, U &dstValues,\n+ 651 std::vector const &valueWeights, Index start, Index end) const {\n+ 652\n+ 653 int const * sizes = &_sizes.at(0);\n+ 654 Index const * indices = &_indices.at(0);\n+ 655 REAL const * weights = &valueWeights.at(0);\n+ 656\n+ 657 if (start > 0) {\n+ 658 assert(start < (Index)_offsets.size());\n+ 659 sizes += start;\n+ 660 indices += _offsets[start];\n+ 661 weights += _offsets[start];\n+ 662 } else {\n+ 663 start = 0;\n+ 664 }\n+ 665\n+ 666 int nstencils = ((end < start) ? GetNumStencils() : end) - start;\n+ 667\n+ 668 for (int i = 0; i < nstencils; ++i, ++sizes) {\n+ 669 dstValues[start + i].Clear();\n+ 670 for (int j = 0; j < *sizes; ++j, ++indices, ++weights) {\n+ 671 dstValues[start + i].AddWithWeight(srcValues[*indices], *weights);\n+ 672 }\n+ 673 }\n+ 674}\n+ 675\n+ 676template \n+ 677inline void\n+678StencilTableReal::generateOffsets() {\n+ 679 Index offset=0;\n+ 680 int noffsets = (int)_sizes.size();\n+ 681 _offsets.resize(noffsets);\n+ 682 for (int i=0; i<(int)_sizes.size(); ++i ) {\n+ 683 _offsets[i]=offset;\n+ 684 offset+=_sizes[i];\n+ 685 }\n+ 686}\n+ 687\n+ 688template \n+ 689inline void\n+690StencilTableReal::resize(int nstencils, int nelems) {\n+ 691 _sizes.resize(nstencils);\n+ 692 _indices.resize(nelems);\n+ 693 _weights.resize(nelems);\n+ 694}\n+ 695\n+ 696template \n+ 697inline void\n+698StencilTableReal::reserve(int nstencils, int nelems) {\n+ 699 _sizes.reserve(nstencils);\n+ 700 _indices.reserve(nelems);\n+ 701 _weights.reserve(nelems);\n+ 702}\n+ 703\n+ 704template \n+ 705inline void\n+706StencilTableReal::shrinkToFit() {\n+ 707 std::vector(_sizes).swap(_sizes);\n+ 708 std::vector(_indices).swap(_indices);\n+ 709 std::vector(_weights).swap(_weights);\n+ 710}\n+ 711\n+ 712template \n+ 713inline void\n+714StencilTableReal::finalize() {\n+ 715 shrinkToFit();\n+ 716 generateOffsets();\n+ 717}\n+ 718\n+ 719// Returns a Stencil at index i in the table\n+ 720template \n+ 721inline StencilReal\n+722StencilTableReal::GetStencil(Index i) const {\n+ 723 assert((! _offsets.empty()) && i<(int)_offsets.size());\n+ 724\n+ 725 Index ofs = _offsets[i];\n+ 726\n+ 727 return StencilReal(const_cast(&_sizes[i]),\n+ 728 const_cast(&_indices[ofs]),\n+ 729 const_cast(&_weights[ofs]));\n+ 730}\n+ 731\n+ 732template \n+ 733inline StencilReal\n+734StencilTableReal::operator[](Index index) const {\n+ 735 return GetStencil(index);\n+ 736}\n+ 737\n+ 738template \n+ 739inline void\n+ 740LimitStencilTableReal::resize(int nstencils, int nelems) {\n+ 741 StencilTableReal::resize(nstencils, nelems);\n+ 742 _duWeights.resize(nelems);\n+ 743 _dvWeights.resize(nelems);\n+ 744}\n+ 745\n+ 746// Returns a LimitStencil at index i in the table\n+ 747template \n+ 748inline LimitStencilReal\n+749LimitStencilTableReal::GetLimitStencil(Index i) const {\n+ 750 assert((! this->GetOffsets().empty()) && i<(int)this->GetOffsets().size\n+());\n+ 751\n+ 752 Index ofs = this->GetOffsets()[i];\n+ 753\n+ 754 if (!_duWeights.empty() && !_dvWeights.empty() &&\n+ 755 !_duuWeights.empty() && !_duvWeights.empty() && !_dvvWeights.empty()) {\n+ 756 return LimitStencilReal(\n+ 757 const_cast(&this->GetSizes()[i]),\n+ 758 const_cast(&this->GetControlIndices()[ofs]),\n+ 759 const_cast(&this->GetWeights()[ofs]),\n+ 760 const_cast(&GetDuWeights()[ofs]),\n+ 761 const_cast(&GetDvWeights()[ofs]),\n+ 762 const_cast(&GetDuuWeights()[ofs]),\n+ 763 const_cast(&GetDuvWeights()[ofs]),\n+ 764 const_cast(&GetDvvWeights()[ofs]) );\n+ 765 } else if (!_duWeights.empty() && !_dvWeights.empty()) {\n+ 766 return LimitStencilReal(\n+ 767 const_cast(&this->GetSizes()[i]),\n+ 768 const_cast(&this->GetControlIndices()[ofs]),\n+ 769 const_cast(&this->GetWeights()[ofs]),\n+ 770 const_cast(&GetDuWeights()[ofs]),\n+ 771 const_cast(&GetDvWeights()[ofs]) );\n+ 772 } else {\n+ 773 return LimitStencilReal(\n+ 774 const_cast(&this->GetSizes()[i]),\n+ 775 const_cast(&this->GetControlIndices()[ofs]),\n+ 776 const_cast(&this->GetWeights()[ofs]) );\n+ 777 }\n+ 778}\n+ 779\n+ 780template \n+ 781inline LimitStencilReal\n+782LimitStencilTableReal::operator[](Index index) const {\n+ 783 return GetLimitStencil(index);\n+ 784}\n+ 785\n+ 786} // end namespace Far\n+ 787\n+ 788} // end namespace OPENSUBDIV_VERSION\n+ 789using namespace OPENSUBDIV_VERSION;\n+ 790\n+ 791} // end namespace OpenSubdiv\n+ 792\n+ 793#endif // OPENSUBDIV3_FAR_STENCILTABLE_H\n OpenSubdiv\n Definition: limits.h:32\n OpenSubdiv::OPENSUBDIV_VERSION::Far::Index\n Vtr::Index Index\n Definition: types.h:41\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor\n-A simple reference to raw topology data for use with TopologyRefinerFactory.\n-Definition: topologyDescriptor.h:49\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::fvarChannels\n-FVarChannel const * fvarChannels\n-Definition: topologyDescriptor.h:82\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numVertices\n-int numVertices\n-Definition: topologyDescriptor.h:51\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::cornerVertexIndices\n-Index const * cornerVertexIndices\n-Definition: topologyDescriptor.h:62\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::vertIndicesPerFace\n-Index const * vertIndicesPerFace\n-Definition: topologyDescriptor.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::creaseVertexIndexPairs\n-Index const * creaseVertexIndexPairs\n-Definition: topologyDescriptor.h:58\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::holeIndices\n-Index const * holeIndices\n-Definition: topologyDescriptor.h:66\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numFaces\n-int numFaces\n-Definition: topologyDescriptor.h:52\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numCorners\n-int numCorners\n-Definition: topologyDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::cornerWeights\n-float const * cornerWeights\n-Definition: topologyDescriptor.h:63\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::isLeftHanded\n-bool isLeftHanded\n-Definition: topologyDescriptor.h:68\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numVertsPerFace\n-int const * numVertsPerFace\n-Definition: topologyDescriptor.h:54\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::TopologyDescriptor\n-TopologyDescriptor()\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::creaseWeights\n-float const * creaseWeights\n-Definition: topologyDescriptor.h:59\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numCreases\n-int numCreases\n-Definition: topologyDescriptor.h:57\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numHoles\n-int numHoles\n-Definition: topologyDescriptor.h:65\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::numFVarChannels\n-int numFVarChannels\n-Definition: topologyDescriptor.h:81\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::FVarChannel\n-Definition: topologyDescriptor.h:73\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::FVarChannel::\n-FVarChannel\n-FVarChannel()\n-Definition: topologyDescriptor.h:78\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::FVarChannel::\n-valueIndices\n-Index const * valueIndices\n-Definition: topologyDescriptor.h:76\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::FVarChannel::numValues\n-int numValues\n-Definition: topologyDescriptor.h:75\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::\n-assignComponentTopology\n-static bool assignComponentTopology(TopologyRefiner &newRefiner, MESH const\n-&mesh)\n-Specify the relationships between vertices, faces, etc. ie the face-vertices,\n-vertex-faces,...\n-Definition: topologyRefinerFactory.h:621\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::\n-reportInvalidTopology\n-static void reportInvalidTopology(TopologyError errCode, char const *msg, MESH\n-const &mesh)\n-(Optional) Control run-time topology validation and error reporting\n-Definition: topologyRefinerFactory.h:707\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::\n-assignComponentTags\n-static bool assignComponentTags(TopologyRefiner &newRefiner, MESH const &mesh)\n-(Optional) Specify edge or vertex sharpness or face holes\n-Definition: topologyRefinerFactory.h:690\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::\n-resizeComponentTopology\n-static bool resizeComponentTopology(TopologyRefiner &newRefiner, MESH const\n-&mesh)\n-Specify the number of vertices, faces, face-vertices, etc.\n-Definition: topologyRefinerFactory.h:580\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory::\n-assignFaceVaryingTopology\n-static bool assignFaceVaryingTopology(TopologyRefiner &newRefiner, MESH const\n-&mesh)\n-(Optional) Specify face-varying data per face\n-Definition: topologyRefinerFactory.h:674\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner\n-Stores topology data for a specified set of refinement options.\n-Definition: topologyRefiner.h:51\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal\n+A specialized factory for StencilTable.\n+Definition: stencilTableFactory.h:51\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal\n+A specialized factory for LimitStencilTable.\n+Definition: stencilTableFactory.h:209\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal\n+Vertex stencil descriptor.\n+Definition: stencilTable.h:53\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::StencilReal\n+StencilReal()\n+Default constructor.\n+Definition: stencilTable.h:57\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::_size\n+int * _size\n+Definition: stencilTable.h:109\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::StencilReal\n+StencilReal(int *size, Index *indices, REAL *weights)\n+Constructor.\n+Definition: stencilTable.h:67\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::Next\n+void Next()\n+Advance to the next stencil in the table.\n+Definition: stencilTable.h:98\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::StencilReal\n+StencilReal(StencilReal const &other)\n+Copy constructor.\n+Definition: stencilTable.h:71\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::_indices\n+Index * _indices\n+Definition: stencilTable.h:110\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetVertexIndices\n+Index const * GetVertexIndices() const\n+Returns the control vertices' indices.\n+Definition: stencilTable.h:88\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetSize\n+int GetSize() const\n+Returns the size of the stencil.\n+Definition: stencilTable.h:78\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::_weights\n+REAL * _weights\n+Definition: stencilTable.h:111\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetSizePtr\n+int * GetSizePtr() const\n+Returns the size of the stencil as a pointer.\n+Definition: stencilTable.h:83\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetWeights\n+REAL const * GetWeights() const\n+Returns the interpolation weights.\n+Definition: stencilTable.h:93\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil\n+Vertex stencil class wrapping the template for compatibility.\n+Definition: stencilTable.h:116\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil::Stencil\n+Stencil(BaseStencil const &other)\n+Definition: stencilTable.h:122\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil::Stencil\n+Stencil(int *size, Index *indices, float *weights)\n+Definition: stencilTable.h:123\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil::BaseStencil\n+StencilReal< float > BaseStencil\n+Definition: stencilTable.h:118\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil::Stencil\n+Stencil()\n+Definition: stencilTable.h:121\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal\n+Table of subdivision stencils.\n+Definition: stencilTable.h:141\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumControlVertices\n+int GetNumControlVertices() const\n+Returns the number of control vertices indexed in the table.\n+Definition: stencilTable.h:161\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::_sizes\n+std::vector< int > _sizes\n+Definition: stencilTable.h:265\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::PatchTableBuilder\n+friend class Far::PatchTableBuilder\n+Definition: stencilTable.h:261\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::update\n+void update(T1 const &srcBase, int numBase, T2 const &srcRef, U &dstValues,\n+std::vector< REAL > const &valueWeights, Index start, Index end) const\n+Definition: stencilTable.h:618\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues\n+void UpdateValues(T1 const &srcBase, int numBase, T2 const &srcRef, U\n+&dstValues, Index start=-1, Index end=-1) const\n+Definition: stencilTable.h:210\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetControlIndices\n+std::vector< Index > const & GetControlIndices() const\n+Returns the indices of the control vertices.\n+Definition: stencilTable.h:179\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::finalize\n+void finalize()\n+Definition: stencilTable.h:714\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::operator[]\n+StencilReal< REAL > operator[](Index index) const\n+Returns the stencil at index i in the table.\n+Definition: stencilTable.h:734\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::_numControlVertices\n+int _numControlVertices\n+Definition: stencilTable.h:263\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::update\n+void update(T const &srcValues, U &dstValues, std::vector< REAL > const\n+&valueWeights, Index start, Index end) const\n+Definition: stencilTable.h:650\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetSizes\n+std::vector< int > const & GetSizes() const\n+Returns the number of control vertices of each stencil in the table.\n+Definition: stencilTable.h:169\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumStencils\n+int GetNumStencils() const\n+Returns the number of stencils in the table.\n+Definition: stencilTable.h:156\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::~StencilTableReal\n+virtual ~StencilTableReal()\n+Definition: stencilTable.h:153\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetOffsets\n+std::vector< Index > const & GetOffsets() const\n+Returns the offset to a given stencil (factory may leave empty)\n+Definition: stencilTable.h:174\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetStencil\n+StencilReal< REAL > GetStencil(Index i) const\n+Returns a Stencil at index i in the table.\n+Definition: stencilTable.h:722\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::shrinkToFit\n+void shrinkToFit()\n+Definition: stencilTable.h:706\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::generateOffsets\n+void generateOffsets()\n+Definition: stencilTable.h:678\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::Clear\n+void Clear()\n+Clears the stencils from the table.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::_offsets\n+std::vector< Index > _offsets\n+Definition: stencilTable.h:266\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::_weights\n+std::vector< REAL > _weights\n+Definition: stencilTable.h:268\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::resize\n+void resize(int nstencils, int nelems)\n+Definition: stencilTable.h:690\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetWeights\n+std::vector< REAL > const & GetWeights() const\n+Returns the stencil interpolation weights.\n+Definition: stencilTable.h:184\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues\n+void UpdateValues(T1 const *srcBase, int numBase, T2 const *srcRef, U *dst,\n+Index start=-1, Index end=-1) const\n+Definition: stencilTable.h:221\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::StencilTableReal\n+StencilTableReal(int numControlVerts, std::vector< int > const &offsets, std::\n+vector< int > const &sizes, std::vector< int > const &sources, std::vector<\n+REAL > const &weights, bool includeCoarseVerts, size_t firstOffset)\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::_indices\n+std::vector< Index > _indices\n+Definition: stencilTable.h:267\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::StencilTableReal\n+StencilTableReal()\n+Definition: stencilTable.h:255\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::reserve\n+void reserve(int nstencils, int nelems)\n+Definition: stencilTable.h:698\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues\n+void UpdateValues(T const *src, U *dst, Index start=-1, Index end=-1) const\n+Definition: stencilTable.h:217\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues\n+void UpdateValues(T const &srcValues, U &dstValues, Index start=-1, Index end=-\n+1) const\n+Updates point values based on the control values.\n+Definition: stencilTable.h:205\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n+Stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:273\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::StencilTable\n+StencilTable(int numControlVerts, std::vector< int > const &offsets, std::\n+vector< int > const &sizes, std::vector< int > const &sources, std::vector<\n+float > const &weights, bool includeCoarseVerts, size_t firstOffset)\n+Definition: stencilTable.h:288\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::StencilTable\n+StencilTable(int numControlVerts)\n+Definition: stencilTable.h:287\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::BaseTable\n+StencilTableReal< float > BaseTable\n+Definition: stencilTable.h:275\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::GetStencil\n+Stencil GetStencil(Index index) const\n+Definition: stencilTable.h:278\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::operator[]\n+Stencil operator[](Index index) const\n+Definition: stencilTable.h:281\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::StencilTable\n+StencilTable()\n+Definition: stencilTable.h:286\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal\n+Limit point stencil descriptor.\n+Definition: stencilTable.h:303\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuvWeights\n+REAL const * GetDuvWeights() const\n+Returns the uv derivative weights.\n+Definition: stencilTable.h:356\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDvWeights\n+REAL const * GetDvWeights() const\n+Returns the v derivative weights.\n+Definition: stencilTable.h:346\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::Next\n+void Next()\n+Advance to the next stencil in the table.\n+Definition: stencilTable.h:366\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDvvWeights\n+REAL const * GetDvvWeights() const\n+Returns the vv derivative weights.\n+Definition: stencilTable.h:361\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::LimitStencilReal\n+LimitStencilReal(int *size, Index *indices, REAL *weights, REAL *duWeights=0,\n+REAL *dvWeights=0, REAL *duuWeights=0, REAL *duvWeights=0, REAL *dvvWeights=0)\n+Constructor.\n+Definition: stencilTable.h:324\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuWeights\n+REAL const * GetDuWeights() const\n+Returns the u derivative weights.\n+Definition: stencilTable.h:341\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuuWeights\n+REAL const * GetDuuWeights() const\n+Returns the uu derivative weights.\n+Definition: stencilTable.h:351\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil\n+Limit point stencil class wrapping the template for compatibility.\n+Definition: stencilTable.h:392\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil::LimitStencil\n+LimitStencil(BaseStencil const &other)\n+Definition: stencilTable.h:397\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil::BaseStencil\n+LimitStencilReal< float > BaseStencil\n+Definition: stencilTable.h:394\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil::LimitStencil\n+LimitStencil(int *size, Index *indices, float *weights, float *duWeights=0,\n+float *dvWeights=0, float *duuWeights=0, float *duvWeights=0, float\n+*dvvWeights=0)\n+Definition: stencilTable.h:398\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal\n+Table of limit subdivision stencils.\n+Definition: stencilTable.h:409\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuuWeights\n+std::vector< REAL > const & GetDuuWeights() const\n+Returns the 'uu' derivative stencil interpolation weights.\n+Definition: stencilTable.h:444\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::operator[]\n+LimitStencilReal< REAL > operator[](Index index) const\n+Returns the limit stencil at index i in the table.\n+Definition: stencilTable.h:782\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDvvWeights\n+std::vector< REAL > const & GetDvvWeights() const\n+Returns the 'vv' derivative stencil interpolation weights.\n+Definition: stencilTable.h:454\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs\n+void UpdateDerivs(T const *src, U *uderivs, U *vderivs, int start=-1, int end=-\n+1) const\n+Definition: stencilTable.h:493\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs\n+void Update2ndDerivs(T const *src, T *uuderivs, U *uvderivs, U *vvderivs, int\n+start=-1, int end=-1) const\n+Definition: stencilTable.h:548\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDvWeights\n+std::vector< REAL > const & GetDvWeights() const\n+Returns the 'v' derivative stencil interpolation weights.\n+Definition: stencilTable.h:439\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs\n+void Update2ndDerivs(T1 const *srcBase, int numBase, T2 const *srcRef, U\n+*uuderivs, U *uvderivs, U *vvderivs, int start=-1, int end=-1) const\n+Definition: stencilTable.h:556\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs\n+void UpdateDerivs(T const &srcValues, U &uderivs, U &vderivs, int start=-1, int\n+end=-1) const\n+Updates derivative values based on the control values.\n+Definition: stencilTable.h:476\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetLimitStencil\n+LimitStencilReal< REAL > GetLimitStencil(Index i) const\n+Returns a LimitStencil at index i in the table.\n+Definition: stencilTable.h:749\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::\n+LimitStencilTableReal\n+LimitStencilTableReal(int numControlVerts, std::vector< int > const &offsets,\n+std::vector< int > const &sizes, std::vector< int > const &sources, std::\n+vector< REAL > const &weights, std::vector< REAL > const &duWeights, std::\n+vector< REAL > const &dvWeights, std::vector< REAL > const &duuWeights, std::\n+vector< REAL > const &duvWeights, std::vector< REAL > const &dvvWeights, bool\n+includeCoarseVerts, size_t firstOffset)\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs\n+void Update2ndDerivs(T const &srcValues, U &uuderivs, U &uvderivs, U &vvderivs,\n+int start=-1, int end=-1) const\n+Updates 2nd derivative values based on the control values.\n+Definition: stencilTable.h:528\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuWeights\n+std::vector< REAL > const & GetDuWeights() const\n+Returns the 'u' derivative stencil interpolation weights.\n+Definition: stencilTable.h:434\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Clear\n+void Clear()\n+Clears the stencils from the table.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs\n+void Update2ndDerivs(T1 const &srcBase, int numBase, T2 const &srcRef, U\n+&uuderivs, U &uvderivs, U &vvderivs, int start=-1, int end=-1) const\n+Definition: stencilTable.h:538\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuvWeights\n+std::vector< REAL > const & GetDuvWeights() const\n+Returns the 'uv' derivative stencil interpolation weights.\n+Definition: stencilTable.h:449\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs\n+void UpdateDerivs(T1 const &srcBase, int numBase, T2 const &srcRef, U &uderivs,\n+U &vderivs, int start=-1, int end=-1) const\n+Definition: stencilTable.h:484\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs\n+void UpdateDerivs(T1 const *srcBase, int numBase, T2 const *srcRef, U *uderivs,\n+U *vderivs, int start=-1, int end=-1) const\n+Definition: stencilTable.h:500\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n+Limit stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:583\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable::BaseTable\n+LimitStencilTableReal< float > BaseTable\n+Definition: stencilTable.h:585\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable::GetLimitStencil\n+LimitStencil GetLimitStencil(Index index) const\n+Definition: stencilTable.h:588\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable::LimitStencilTable\n+LimitStencilTable(int numControlVerts, std::vector< int > const &offsets, std::\n+vector< int > const &sizes, std::vector< int > const &sources, std::vector<\n+float > const &weights, std::vector< float > const &duWeights, std::vector<\n+float > const &dvWeights, std::vector< float > const &duuWeights, std::vector<\n+float > const &duvWeights, std::vector< float > const &dvvWeights, bool\n+includeCoarseVerts, size_t firstOffset)\n+Definition: stencilTable.h:596\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable::operator[]\n+LimitStencil operator[](Index index) const\n+Definition: stencilTable.h:591\n * opensubdiv\n * far\n- * topologyDescriptor.h\n+ * stencilTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00716.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00716.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/ptexIndices.h File Reference\n+OpenSubdiv: opensubdiv/far/stencilTableFactory.h File Reference\n \n \n \n \n \n \n \n@@ -83,28 +83,43 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
ptexIndices.h File Reference
\n+
stencilTableFactory.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../far/topologyRefiner.h"
\n-#include "../far/types.h"
\n+#include "../far/patchTable.h"
\n #include <vector>
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n \n

\n Classes

class  PtexIndices
 Object used to compute and query ptex face indices. More...
class  StencilTableFactoryReal< REAL >
 A specialized factory for StencilTable. More...
 
struct  StencilTableFactoryReal< REAL >::Options
 
class  LimitStencilTableFactoryReal< REAL >
 A specialized factory for LimitStencilTable. More...
 
struct  LimitStencilTableFactoryReal< REAL >::Options
 
struct  LimitStencilTableFactoryReal< REAL >::LocationArray
 Descriptor for limit surface locations. More...
 
class  StencilTableFactory
 Stencil table factory class wrapping the template for compatibility. More...
 
class  LimitStencilTableFactory
 Stencil table factory class wrapping the template for compatibility. More...
 
\n \n \n \n \n@@ -113,13 +128,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,28 +5,45 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-ptexIndices.h File Reference\n+stencilTableFactory.h File Reference\n #include \"../version.h\"\n-#include \"../far/topologyRefiner.h\"\n-#include \"../far/types.h\"\n+#include \"../far/patchTable.h\"\n #include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0PtexIndices\n-\u00a0 Object used to compute and query ptex face indices. More...\n+ class \u00a0StencilTableFactoryReal<_REAL_>\n+\u00a0 A specialized factory for StencilTable. More...\n+\u00a0\n+struct \u00a0StencilTableFactoryReal<_REAL_>::Options\n+\u00a0\n+ class \u00a0LimitStencilTableFactoryReal<_REAL_>\n+\u00a0 A specialized factory for LimitStencilTable. More...\n+\u00a0\n+struct \u00a0LimitStencilTableFactoryReal<_REAL_>::Options\n+\u00a0\n+struct \u00a0LimitStencilTableFactoryReal<_REAL_>::LocationArray\n+\u00a0 Descriptor for limit surface locations. More...\n+\u00a0\n+ class \u00a0StencilTableFactory\n+\u00a0 Stencil table factory class wrapping the template for compatibility.\n+ More...\n+\u00a0\n+ class \u00a0LimitStencilTableFactory\n+\u00a0 Stencil table factory class wrapping the template for compatibility.\n+ More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n * opensubdiv\n * far\n- * ptexIndices.h\n+ * stencilTableFactory.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00716.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00716.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,9 @@\n var a00716 = [\n- [\"PtexIndices\", \"a01033.html\", \"a01033\"]\n+ [\"StencilTableFactoryReal< REAL >\", \"a01037.html\", \"a01037\"],\n+ [\"StencilTableFactoryReal< REAL >::Options\", \"a01077.html\", \"a01077\"],\n+ [\"LimitStencilTableFactoryReal< REAL >\", \"a01041.html\", \"a01041\"],\n+ [\"LimitStencilTableFactoryReal< REAL >::Options\", \"a01081.html\", \"a01081\"],\n+ [\"LimitStencilTableFactoryReal< REAL >::LocationArray\", \"a01085.html\", \"a01085\"],\n+ [\"StencilTableFactory\", \"a01089.html\", null],\n+ [\"LimitStencilTableFactory\", \"a01093.html\", null]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00716_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00716_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/ptexIndices.h Source File\n+OpenSubdiv: opensubdiv/far/stencilTableFactory.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
ptexIndices.h
\n+
stencilTableFactory.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2015 Pixar
\n+
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -106,79 +106,322 @@\n
17//
\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n-
24#ifndef OPENSUBDIV3_FAR_PTEX_INDICES_H
\n-
25#define OPENSUBDIV3_FAR_PTEX_INDICES_H
\n-
26
\n-
27#include "../version.h"
\n-
28
\n-
29#include "../far/topologyRefiner.h"
\n-
30#include "../far/types.h"
\n+
24
\n+
25#ifndef OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H
\n+
26#define OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H
\n+
27
\n+
28#include "../version.h"
\n+
29
\n+
30#include "../far/patchTable.h"
\n
31
\n
32#include <vector>
\n
33
\n
34namespace OpenSubdiv {
\n
35namespace OPENSUBDIV_VERSION {
\n
36
\n
37namespace Far {
\n
38
\n-
46class PtexIndices {
\n+
39class TopologyRefiner;
\n+
40
\n+
41template <typename REAL> class StencilReal;
\n+
42template <typename REAL> class StencilTableReal;
\n+
43
\n+
44template <typename REAL> class LimitStencilReal;
\n+
45template <typename REAL> class LimitStencilTableReal;
\n+
46
\n
47
\n-
48public:
\n-
49
\n-
51 PtexIndices(TopologyRefiner const &refiner);
\n+
50template <typename REAL>
\n+
51class StencilTableFactoryReal {
\n
52
\n-
54 ~PtexIndices();
\n-
55
\n-
57
\n+
53public:
\n+
54
\n+
55 enum Mode {
\n+
56 INTERPOLATE_VERTEX=0,
\n+
57 INTERPOLATE_VARYING,
\n+
58 INTERPOLATE_FACE_VARYING
\n+
59 };
\n
60
\n-
63 int GetNumFaces() const;
\n-
64
\n-
67 int GetFaceId(Index f) const;
\n-
68
\n-
83 void GetAdjacency(
\n-
84 TopologyRefiner const &refiner,
\n-
85 int face, int quadrant,
\n-
86 int adjFaces[4], int adjEdges[4]) const;
\n-
87
\n-
89
\n-
90private:
\n-
91
\n-
92 void initializePtexIndices(TopologyRefiner const &refiner);
\n-
93
\n-
94private:
\n-
95
\n-
96 std::vector<Index> _ptexIndices;
\n-
97};
\n-
98
\n-
99
\n-
100} // end namespace Far
\n-
101
\n-
102} // end namespace OPENSUBDIV_VERSION
\n-
103using namespace OPENSUBDIV_VERSION;
\n-
104} // end namespace OpenSubdiv
\n-
105
\n-
106#endif /* OPENSUBDIV3_FAR_PTEX_INDICES_H */
\n+
61 struct Options {
\n+
62
\n+
63 Options() : interpolationMode(INTERPOLATE_VERTEX),
\n+
64 generateOffsets(false),
\n+
65 generateControlVerts(false),
\n+
66 generateIntermediateLevels(true),
\n+
67 factorizeIntermediateLevels(true),
\n+
68 maxLevel(10),
\n+
69 fvarChannel(0) { }
\n+
70
\n+
71 unsigned int interpolationMode : 2,
\n+
72 generateOffsets : 1,
\n+
73 generateControlVerts : 1,
\n+
74 generateIntermediateLevels : 1,
\n+
75 factorizeIntermediateLevels : 1,
\n+
78 maxLevel : 4;
\n+
79 unsigned int fvarChannel;
\n+
81 };
\n+
82
\n+
94 static StencilTableReal<REAL> const * Create(
\n+
95 TopologyRefiner const & refiner, Options options = Options());
\n+
96
\n+
97
\n+
110 static StencilTableReal<REAL> const * Create(
\n+
111 int numTables, StencilTableReal<REAL> const ** tables);
\n+
112
\n+
113
\n+
128 static StencilTableReal<REAL> const * AppendLocalPointStencilTable(
\n+
129 TopologyRefiner const &refiner,
\n+
130 StencilTableReal<REAL> const *baseStencilTable,
\n+
131 StencilTableReal<REAL> const *localPointStencilTable,
\n+
132 bool factorize = true);
\n+
133
\n+
148 static StencilTableReal<REAL> const * AppendLocalPointStencilTableVarying(
\n+
149 TopologyRefiner const &refiner,
\n+
150 StencilTableReal<REAL> const *baseStencilTable,
\n+
151 StencilTableReal<REAL> const *localPointStencilTable,
\n+
152 bool factorize = true) {
\n+
153 return AppendLocalPointStencilTable(
\n+
154 refiner, baseStencilTable, localPointStencilTable, factorize);
\n+
155 }
\n+
156
\n+
174 static StencilTableReal<REAL> const * AppendLocalPointStencilTableFaceVarying(
\n+
175 TopologyRefiner const &refiner,
\n+
176 StencilTableReal<REAL> const *baseStencilTable,
\n+
177 StencilTableReal<REAL> const *localPointStencilTable,
\n+
178 int channel = 0,
\n+
179 bool factorize = true);
\n+
180
\n+
181private:
\n+
182
\n+
183 // Generate stencils for the coarse control-vertices (single weight = 1.0f)
\n+
184 static void generateControlVertStencils(
\n+
185 int numControlVerts,
\n+
186 StencilReal<REAL> & dst);
\n+
187
\n+
188 // Internal method to splice local point stencils
\n+
189 static StencilTableReal<REAL> const * appendLocalPointStencilTable(
\n+
190 TopologyRefiner const &refiner,
\n+
191 StencilTableReal<REAL> const * baseStencilTable,
\n+
192 StencilTableReal<REAL> const * localPointStencilTable,
\n+
193 int channel,
\n+
194 bool factorize);
\n+
195};
\n+
196
\n+
208template <typename REAL>
\n+
209class LimitStencilTableFactoryReal {
\n+
210
\n+
211public:
\n+
212
\n+
213 enum Mode {
\n+
214 INTERPOLATE_VERTEX=0,
\n+
215 INTERPOLATE_VARYING,
\n+
216 INTERPOLATE_FACE_VARYING
\n+
217 };
\n+
218
\n+
219 struct Options {
\n+
220
\n+
221 Options() : interpolationMode(INTERPOLATE_VERTEX),
\n+
222 generate1stDerivatives(true),
\n+
223 generate2ndDerivatives(false),
\n+
224 fvarChannel(0) { }
\n+
225
\n+
226 unsigned int interpolationMode : 2,
\n+
227 generate1stDerivatives : 1,
\n+
228 generate2ndDerivatives : 1;
\n+
229 unsigned int fvarChannel;
\n+
230 };
\n+
231
\n+
233 struct LocationArray {
\n+
234
\n+
235 LocationArray() : ptexIdx(-1), numLocations(0), s(0), t(0) { }
\n+
236
\n+
237 int ptexIdx,
\n+
238 numLocations;
\n+
239
\n+
240 REAL const * s,
\n+
241 * t;
\n+
242 };
\n+
243
\n+
244 typedef std::vector<LocationArray> LocationArrayVec;
\n+
245
\n+
269 static LimitStencilTableReal<REAL> const * Create(
\n+
270 TopologyRefiner const & refiner,
\n+
271 LocationArrayVec const & locationArrays,
\n+
272 StencilTableReal<REAL> const * cvStencils = 0,
\n+
273 PatchTable const * patchTable = 0,
\n+
274 Options options = Options());
\n+
275
\n+
276};
\n+
277
\n+
278
\n+
279//
\n+
280// Public wrapper classes for the templates
\n+
281//
\n+
282class Stencil;
\n+
283class StencilTable;
\n+
284
\n+
287class StencilTableFactory : public StencilTableFactoryReal<float> {
\n+
288private:
\n+
289 typedef StencilTableFactoryReal<float> BaseFactory;
\n+
290 typedef StencilTableReal<float> BaseTable;
\n+
291
\n+
292public:
\n+
293 static StencilTable const * Create(
\n+
294 TopologyRefiner const & refiner, Options options = Options()) {
\n+
295
\n+
296 return static_cast<StencilTable const *>(
\n+
297 BaseFactory::Create(refiner, options));
\n+
298 }
\n+
299
\n+
300 static StencilTable const * Create(
\n+
301 int numTables, StencilTable const ** tables) {
\n+
302
\n+
303 return static_cast<StencilTable const *>(
\n+
304 BaseFactory::Create(numTables,
\n+
305 reinterpret_cast<BaseTable const **>(tables)));
\n+
306 }
\n+
307
\n+
308 static StencilTable const * AppendLocalPointStencilTable(
\n+
309 TopologyRefiner const &refiner,
\n+
310 StencilTable const *baseStencilTable,
\n+
311 StencilTable const *localPointStencilTable,
\n+
312 bool factorize = true) {
\n+
313
\n+
314 return static_cast<StencilTable const *>(
\n+
315 BaseFactory::AppendLocalPointStencilTable(refiner,
\n+
316 static_cast<BaseTable const *>(baseStencilTable),
\n+
317 static_cast<BaseTable const *>(localPointStencilTable),
\n+
318 factorize));
\n+
319 }
\n+
320
\n+
321 static StencilTable const * AppendLocalPointStencilTableVarying(
\n+
322 TopologyRefiner const &refiner,
\n+
323 StencilTable const *baseStencilTable,
\n+
324 StencilTable const *localPointStencilTable,
\n+
325 bool factorize = true) {
\n+
326
\n+
327 return static_cast<StencilTable const *>(
\n+
328 BaseFactory::AppendLocalPointStencilTableVarying(refiner,
\n+
329 static_cast<BaseTable const *>(baseStencilTable),
\n+
330 static_cast<BaseTable const *>(localPointStencilTable),
\n+
331 factorize));
\n+
332 }
\n+
333
\n+
334 static StencilTable const * AppendLocalPointStencilTableFaceVarying(
\n+
335 TopologyRefiner const &refiner,
\n+
336 StencilTable const *baseStencilTable,
\n+
337 StencilTable const *localPointStencilTable,
\n+
338 int channel = 0,
\n+
339 bool factorize = true) {
\n+
340
\n+
341 return static_cast<StencilTable const *>(
\n+
342 BaseFactory::AppendLocalPointStencilTableFaceVarying(refiner,
\n+
343 static_cast<BaseTable const *>(baseStencilTable),
\n+
344 static_cast<BaseTable const *>(localPointStencilTable),
\n+
345 channel, factorize));
\n+
346 }
\n+
347};
\n+
348
\n+
349class LimitStencil;
\n+
350class LimitStencilTable;
\n+
351
\n+
354class LimitStencilTableFactory : public LimitStencilTableFactoryReal<float> {
\n+
355private:
\n+
356 typedef LimitStencilTableFactoryReal<float> BaseFactory;
\n+
357 typedef StencilTableReal<float> BaseTable;
\n+
358
\n+
359public:
\n+
360 static LimitStencilTable const * Create(
\n+
361 TopologyRefiner const & refiner,
\n+
362 LocationArrayVec const & locationArrays,
\n+
363 StencilTable const * cvStencils = 0,
\n+
364 PatchTable const * patchTable = 0,
\n+
365 Options options = Options()) {
\n+
366
\n+
367 return static_cast<LimitStencilTable const *>(
\n+
368 BaseFactory::Create(
\n+
369 refiner,
\n+
370 locationArrays,
\n+
371 static_cast<BaseTable const *>(cvStencils),
\n+
372 patchTable,
\n+
373 options));
\n+
374 }
\n+
375};
\n+
376
\n+
377} // end namespace Far
\n+
378
\n+
379} // end namespace OPENSUBDIV_VERSION
\n+
380using namespace OPENSUBDIV_VERSION;
\n+
381
\n+
382} // end namespace OpenSubdiv
\n+
383
\n+
384#endif // OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H
\n \n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices
Object used to compute and query ptex face indices.
Definition: ptexIndices.h:46
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetFaceId
int GetFaceId(Index f) const
Returns the ptex face index given a coarse face 'f' or -1.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::PtexIndices
PtexIndices(TopologyRefiner const &refiner)
Constructor.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetAdjacency
void GetAdjacency(TopologyRefiner const &refiner, int face, int quadrant, int adjFaces[4], int adjEdges[4]) const
Returns ptex face adjacency information for a given coarse face.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetNumFaces
int GetNumFaces() const
Returns the number of ptex faces in the mesh.
\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Create
static StencilTableReal< REAL > const * Create(TopologyRefiner const &refiner, Options options=Options())
Instantiates StencilTable from TopologyRefiner that have been refined uniformly or adaptively.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::AppendLocalPointStencilTableFaceVarying
static StencilTableReal< REAL > const * AppendLocalPointStencilTableFaceVarying(TopologyRefiner const &refiner, StencilTableReal< REAL > const *baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, int channel=0, bool factorize=true)
Utility function for stencil splicing for local point face-varying stencils.
\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::AppendLocalPointStencilTableVarying
static StencilTableReal< REAL > const * AppendLocalPointStencilTableVarying(TopologyRefiner const &refiner, StencilTableReal< REAL > const *baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, bool factorize=true)
Utility function for stencil splicing for local point varying stencils.
Definition: stencilTableFactory.h:148
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::AppendLocalPointStencilTable
static StencilTableReal< REAL > const * AppendLocalPointStencilTable(TopologyRefiner const &refiner, StencilTableReal< REAL > const *baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, bool factorize=true)
Utility function for stencil splicing for local point stencils.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Create
static StencilTableReal< REAL > const * Create(int numTables, StencilTableReal< REAL > const **tables)
Instantiates StencilTable by concatenating an array of existing stencil tables.
\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Create
static LimitStencilTableReal< REAL > const * Create(TopologyRefiner const &refiner, LocationArrayVec const &locationArrays, StencilTableReal< REAL > const *cvStencils=0, PatchTable const *patchTable=0, Options options=Options())
Instantiates LimitStencilTable from a TopologyRefiner that has been refined either uniformly or adapt...
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil
Vertex stencil class wrapping the template for compatibility.
Definition: stencilTable.h:116
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::generateIntermediateLevels
unsigned int generateIntermediateLevels
vertices at all levels or highest only
Definition: stencilTableFactory.h:74
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::generateControlVerts
unsigned int generateControlVerts
generate stencils for control-vertices
Definition: stencilTableFactory.h:73
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::generateOffsets
unsigned int generateOffsets
populate optional "_offsets" field
Definition: stencilTableFactory.h:72
\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::generate2ndDerivatives
unsigned int generate2ndDerivatives
Generate weights for 2nd derivatives.
Definition: stencilTableFactory.h:228
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::generate1stDerivatives
unsigned int generate1stDerivatives
Generate weights for 1st derivatives.
Definition: stencilTableFactory.h:227
\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory
Stencil table factory class wrapping the template for compatibility.
Definition: stencilTableFactory.h:287
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::AppendLocalPointStencilTableVarying
static StencilTable const * AppendLocalPointStencilTableVarying(TopologyRefiner const &refiner, StencilTable const *baseStencilTable, StencilTable const *localPointStencilTable, bool factorize=true)
Definition: stencilTableFactory.h:321
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::Create
static StencilTable const * Create(TopologyRefiner const &refiner, Options options=Options())
Definition: stencilTableFactory.h:293
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::AppendLocalPointStencilTableFaceVarying
static StencilTable const * AppendLocalPointStencilTableFaceVarying(TopologyRefiner const &refiner, StencilTable const *baseStencilTable, StencilTable const *localPointStencilTable, int channel=0, bool factorize=true)
Definition: stencilTableFactory.h:334
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::AppendLocalPointStencilTable
static StencilTable const * AppendLocalPointStencilTable(TopologyRefiner const &refiner, StencilTable const *baseStencilTable, StencilTable const *localPointStencilTable, bool factorize=true)
Definition: stencilTableFactory.h:308
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::Create
static StencilTable const * Create(int numTables, StencilTable const **tables)
Definition: stencilTableFactory.h:300
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactory
Stencil table factory class wrapping the template for compatibility.
Definition: stencilTableFactory.h:354
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactory::Create
static LimitStencilTable const * Create(TopologyRefiner const &refiner, LocationArrayVec const &locationArrays, StencilTable const *cvStencils=0, PatchTable const *patchTable=0, Options options=Options())
Definition: stencilTableFactory.h:360
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner
Stores topology data for a specified set of refinement options.
Definition: topologyRefiner.h:51
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-ptexIndices.h\n+stencilTableFactory.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2015 Pixar\n+ 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -29,90 +29,499 @@\n 17//\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n- 24#ifndef OPENSUBDIV3_FAR_PTEX_INDICES_H\n- 25#define OPENSUBDIV3_FAR_PTEX_INDICES_H\n- 26\n- 27#include \"../version.h\"\n- 28\n- 29#include \"../far/topologyRefiner.h\"\n- 30#include \"../far/types.h\"\n+ 24\n+ 25#ifndef OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H\n+ 26#define OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H\n+ 27\n+ 28#include \"../version.h\"\n+ 29\n+ 30#include \"../far/patchTable.h\"\n 31\n 32#include \n 33\n 34namespace OpenSubdiv {\n 35namespace OPENSUBDIV_VERSION {\n 36\n 37namespace Far {\n 38\n-46class PtexIndices {\n+ 39class TopologyRefiner;\n+ 40\n+ 41template class StencilReal;\n+ 42template class StencilTableReal;\n+ 43\n+ 44template class LimitStencilReal;\n+ 45template class LimitStencilTableReal;\n+ 46\n 47\n- 48public:\n- 49\n-51 PtexIndices(TopologyRefiner const &refiner);\n+ 50template \n+51class StencilTableFactoryReal {\n 52\n-54 ~PtexIndices();\n- 55\n- 57\n+ 53public:\n+ 54\n+55 enum Mode {\n+56 INTERPOLATE_VERTEX=0,\n+57 INTERPOLATE_VARYING,\n+ 58 INTERPOLATE_FACE_VARYING\n+59 };\n 60\n-63 int GetNumFaces() const;\n- 64\n-67 int GetFaceId(Index f) const;\n- 68\n-83 void GetAdjacency(\n- 84 TopologyRefiner const &refiner,\n- 85 int face, int quadrant,\n- 86 int adjFaces[4], int adjEdges[4]) const;\n- 87\n- 89\n- 90private:\n- 91\n- 92 void initializePtexIndices(TopologyRefiner const &refiner);\n- 93\n- 94private:\n- 95\n- 96 std::vector _ptexIndices;\n- 97};\n- 98\n- 99\n- 100} // end namespace Far\n- 101\n- 102} // end namespace OPENSUBDIV_VERSION\n- 103using namespace OPENSUBDIV_VERSION;\n- 104} // end namespace OpenSubdiv\n- 105\n- 106#endif /* OPENSUBDIV3_FAR_PTEX_INDICES_H */\n+61 struct Options {\n+ 62\n+63 Options() : interpolationMode(INTERPOLATE_VERTEX),\n+ 64 generateOffsets(false),\n+ 65 generateControlVerts(false),\n+ 66 generateIntermediateLevels(true),\n+ 67 factorizeIntermediateLevels(true),\n+ 68 maxLevel(10),\n+ 69 fvarChannel(0) { }\n+ 70\n+71 unsigned int interpolationMode : 2,\n+72 generateOffsets : 1,\n+73 generateControlVerts : 1,\n+74 generateIntermediateLevels : 1,\n+75 factorizeIntermediateLevels : 1,\n+78 maxLevel : 4;\n+79 unsigned int fvarChannel;\n+ 81 };\n+ 82\n+94 static StencilTableReal const * Create(\n+ 95 TopologyRefiner const & refiner, Options options = Options());\n+ 96\n+ 97\n+110 static StencilTableReal const * Create(\n+ 111 int numTables, StencilTableReal const ** tables);\n+ 112\n+ 113\n+128 static StencilTableReal const * AppendLocalPointStencilTable(\n+ 129 TopologyRefiner const &refiner,\n+ 130 StencilTableReal const *baseStencilTable,\n+ 131 StencilTableReal const *localPointStencilTable,\n+ 132 bool factorize = true);\n+ 133\n+148 static StencilTableReal const * AppendLocalPointStencilTableVarying(\n+ 149 TopologyRefiner const &refiner,\n+ 150 StencilTableReal const *baseStencilTable,\n+ 151 StencilTableReal const *localPointStencilTable,\n+ 152 bool factorize = true) {\n+ 153 return AppendLocalPointStencilTable(\n+ 154 refiner, baseStencilTable, localPointStencilTable, factorize);\n+ 155 }\n+ 156\n+174 static StencilTableReal const *\n+AppendLocalPointStencilTableFaceVarying(\n+ 175 TopologyRefiner const &refiner,\n+ 176 StencilTableReal const *baseStencilTable,\n+ 177 StencilTableReal const *localPointStencilTable,\n+ 178 int channel = 0,\n+ 179 bool factorize = true);\n+ 180\n+ 181private:\n+ 182\n+ 183 // Generate stencils for the coarse control-vertices (single weight =\n+1.0f)\n+ 184 static void generateControlVertStencils(\n+ 185 int numControlVerts,\n+ 186 StencilReal & dst);\n+ 187\n+ 188 // Internal method to splice local point stencils\n+ 189 static StencilTableReal const * appendLocalPointStencilTable(\n+ 190 TopologyRefiner const &refiner,\n+ 191 StencilTableReal const * baseStencilTable,\n+ 192 StencilTableReal const * localPointStencilTable,\n+ 193 int channel,\n+ 194 bool factorize);\n+ 195};\n+ 196\n+ 208template \n+209class LimitStencilTableFactoryReal {\n+ 210\n+ 211public:\n+ 212\n+213 enum Mode {\n+214 INTERPOLATE_VERTEX=0,\n+215 INTERPOLATE_VARYING,\n+ 216 INTERPOLATE_FACE_VARYING\n+217 };\n+ 218\n+219 struct Options {\n+ 220\n+221 Options() : interpolationMode(INTERPOLATE_VERTEX),\n+ 222 generate1stDerivatives(true),\n+ 223 generate2ndDerivatives(false),\n+ 224 fvarChannel(0) { }\n+ 225\n+226 unsigned int interpolationMode : 2,\n+227 generate1stDerivatives : 1,\n+228 generate2ndDerivatives : 1;\n+229 unsigned int fvarChannel;\n+ 230 };\n+ 231\n+233 struct LocationArray {\n+ 234\n+235 LocationArray() : ptexIdx(-1), numLocations(0), s(0), t(0) { }\n+ 236\n+237 int ptexIdx,\n+238 numLocations;\n+ 239\n+240 REAL const * s,\n+241 * t;\n+ 242 };\n+ 243\n+244 typedef std::vector LocationArrayVec;\n+ 245\n+269 static LimitStencilTableReal const * Create(\n+ 270 TopologyRefiner const & refiner,\n+ 271 LocationArrayVec const & locationArrays,\n+ 272 StencilTableReal const * cvStencils = 0,\n+ 273 PatchTable const * patchTable = 0,\n+ 274 Options options = Options());\n+ 275\n+ 276};\n+ 277\n+ 278\n+ 279//\n+ 280// Public wrapper classes for the templates\n+ 281//\n+ 282class Stencil;\n+ 283class StencilTable;\n+ 284\n+287class StencilTableFactory : public StencilTableFactoryReal {\n+ 288private:\n+ 289 typedef StencilTableFactoryReal BaseFactory;\n+ 290 typedef StencilTableReal BaseTable;\n+ 291\n+ 292public:\n+293 static StencilTable const * Create(\n+ 294 TopologyRefiner const & refiner, Options options = Options()) {\n+ 295\n+ 296 return static_cast(\n+ 297 BaseFactory::Create(refiner, options));\n+ 298 }\n+ 299\n+300 static StencilTable const * Create(\n+ 301 int numTables, StencilTable const ** tables) {\n+ 302\n+ 303 return static_cast(\n+ 304 BaseFactory::Create(numTables,\n+ 305 reinterpret_cast(tables)));\n+ 306 }\n+ 307\n+308 static StencilTable const * AppendLocalPointStencilTable(\n+ 309 TopologyRefiner const &refiner,\n+ 310 StencilTable const *baseStencilTable,\n+ 311 StencilTable const *localPointStencilTable,\n+ 312 bool factorize = true) {\n+ 313\n+ 314 return static_cast(\n+ 315 BaseFactory::AppendLocalPointStencilTable(refiner,\n+ 316 static_cast(baseStencilTable),\n+ 317 static_cast(localPointStencilTable),\n+ 318 factorize));\n+ 319 }\n+ 320\n+321 static StencilTable const * AppendLocalPointStencilTableVarying(\n+ 322 TopologyRefiner const &refiner,\n+ 323 StencilTable const *baseStencilTable,\n+ 324 StencilTable const *localPointStencilTable,\n+ 325 bool factorize = true) {\n+ 326\n+ 327 return static_cast(\n+ 328 BaseFactory::AppendLocalPointStencilTableVarying(refiner,\n+ 329 static_cast(baseStencilTable),\n+ 330 static_cast(localPointStencilTable),\n+ 331 factorize));\n+ 332 }\n+ 333\n+334 static StencilTable const * AppendLocalPointStencilTableFaceVarying(\n+ 335 TopologyRefiner const &refiner,\n+ 336 StencilTable const *baseStencilTable,\n+ 337 StencilTable const *localPointStencilTable,\n+ 338 int channel = 0,\n+ 339 bool factorize = true) {\n+ 340\n+ 341 return static_cast(\n+ 342 BaseFactory::AppendLocalPointStencilTableFaceVarying(refiner,\n+ 343 static_cast(baseStencilTable),\n+ 344 static_cast(localPointStencilTable),\n+ 345 channel, factorize));\n+ 346 }\n+ 347};\n+ 348\n+ 349class LimitStencil;\n+ 350class LimitStencilTable;\n+ 351\n+354class LimitStencilTableFactory : public LimitStencilTableFactoryReal\n+{\n+ 355private:\n+ 356 typedef LimitStencilTableFactoryReal BaseFactory;\n+ 357 typedef StencilTableReal BaseTable;\n+ 358\n+ 359public:\n+360 static LimitStencilTable const * Create(\n+ 361 TopologyRefiner const & refiner,\n+ 362 LocationArrayVec const & locationArrays,\n+ 363 StencilTable const * cvStencils = 0,\n+ 364 PatchTable const * patchTable = 0,\n+ 365 Options options = Options()) {\n+ 366\n+ 367 return static_cast(\n+ 368 BaseFactory::Create(\n+ 369 refiner,\n+ 370 locationArrays,\n+ 371 static_cast(cvStencils),\n+ 372 patchTable,\n+ 373 options));\n+ 374 }\n+ 375};\n+ 376\n+ 377} // end namespace Far\n+ 378\n+ 379} // end namespace OPENSUBDIV_VERSION\n+ 380using namespace OPENSUBDIV_VERSION;\n+ 381\n+ 382} // end namespace OpenSubdiv\n+ 383\n+ 384#endif // OPENSUBDIV3_FAR_STENCILTABLE_FACTORY_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::Index\n-Vtr::Index Index\n-Definition: types.h:41\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices\n-Object used to compute and query ptex face indices.\n-Definition: ptexIndices.h:46\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetFaceId\n-int GetFaceId(Index f) const\n-Returns the ptex face index given a coarse face 'f' or -1.\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::PtexIndices\n-PtexIndices(TopologyRefiner const &refiner)\n-Constructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetAdjacency\n-void GetAdjacency(TopologyRefiner const &refiner, int face, int quadrant, int\n-adjFaces[4], int adjEdges[4]) const\n-Returns ptex face adjacency information for a given coarse face.\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetNumFaces\n-int GetNumFaces() const\n-Returns the number of ptex faces in the mesh.\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::~PtexIndices\n-~PtexIndices()\n-Destructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n+Container for arrays of parametric patches.\n+Definition: patchTable.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal\n+A specialized factory for StencilTable.\n+Definition: stencilTableFactory.h:51\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Create\n+static StencilTableReal< REAL > const * Create(TopologyRefiner const &refiner,\n+Options options=Options())\n+Instantiates StencilTable from TopologyRefiner that have been refined uniformly\n+or adaptively.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n+AppendLocalPointStencilTableFaceVarying\n+static StencilTableReal< REAL > const * AppendLocalPointStencilTableFaceVarying\n+(TopologyRefiner const &refiner, StencilTableReal< REAL > const\n+*baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, int\n+channel=0, bool factorize=true)\n+Utility function for stencil splicing for local point face-varying stencils.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Mode\n+Mode\n+Definition: stencilTableFactory.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n+INTERPOLATE_FACE_VARYING\n+@ INTERPOLATE_FACE_VARYING\n+face-varying primvar stencils\n+Definition: stencilTableFactory.h:58\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n+INTERPOLATE_VARYING\n+@ INTERPOLATE_VARYING\n+varying primvar stencils\n+Definition: stencilTableFactory.h:57\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n+INTERPOLATE_VERTEX\n+@ INTERPOLATE_VERTEX\n+vertex primvar stencils\n+Definition: stencilTableFactory.h:56\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n+AppendLocalPointStencilTableVarying\n+static StencilTableReal< REAL > const * AppendLocalPointStencilTableVarying\n+(TopologyRefiner const &refiner, StencilTableReal< REAL > const\n+*baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, bool\n+factorize=true)\n+Utility function for stencil splicing for local point varying stencils.\n+Definition: stencilTableFactory.h:148\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::\n+AppendLocalPointStencilTable\n+static StencilTableReal< REAL > const * AppendLocalPointStencilTable\n+(TopologyRefiner const &refiner, StencilTableReal< REAL > const\n+*baseStencilTable, StencilTableReal< REAL > const *localPointStencilTable, bool\n+factorize=true)\n+Utility function for stencil splicing for local point stencils.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Create\n+static StencilTableReal< REAL > const * Create(int numTables, StencilTableReal<\n+REAL > const **tables)\n+Instantiates StencilTable by concatenating an array of existing stencil tables.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal\n+A specialized factory for LimitStencilTable.\n+Definition: stencilTableFactory.h:209\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Mode\n+Mode\n+Definition: stencilTableFactory.h:213\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n+INTERPOLATE_FACE_VARYING\n+@ INTERPOLATE_FACE_VARYING\n+face-varying primvar stencils\n+Definition: stencilTableFactory.h:216\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n+INTERPOLATE_VARYING\n+@ INTERPOLATE_VARYING\n+varying primvar stencils\n+Definition: stencilTableFactory.h:215\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n+INTERPOLATE_VERTEX\n+@ INTERPOLATE_VERTEX\n+vertex primvar stencils\n+Definition: stencilTableFactory.h:214\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n+LocationArrayVec\n+std::vector< LocationArray > LocationArrayVec\n+Definition: stencilTableFactory.h:244\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Create\n+static LimitStencilTableReal< REAL > const * Create(TopologyRefiner const\n+&refiner, LocationArrayVec const &locationArrays, StencilTableReal< REAL >\n+const *cvStencils=0, PatchTable const *patchTable=0, Options options=Options())\n+Instantiates LimitStencilTable from a TopologyRefiner that has been refined\n+either uniformly or adapt...\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal\n+Vertex stencil descriptor.\n+Definition: stencilTable.h:53\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil\n+Vertex stencil class wrapping the template for compatibility.\n+Definition: stencilTable.h:116\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal\n+Table of subdivision stencils.\n+Definition: stencilTable.h:141\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n+Stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:273\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal\n+Table of limit subdivision stencils.\n+Definition: stencilTable.h:409\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n+Limit stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:583\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options\n+Definition: stencilTableFactory.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n+factorizeIntermediateLevels\n+unsigned int factorizeIntermediateLevels\n+Definition: stencilTableFactory.h:77\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::maxLevel\n+unsigned int maxLevel\n+generate stencils up to 'maxLevel'\n+Definition: stencilTableFactory.h:78\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::Options\n+Options()\n+Definition: stencilTableFactory.h:63\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n+generateIntermediateLevels\n+unsigned int generateIntermediateLevels\n+vertices at all levels or highest only\n+Definition: stencilTableFactory.h:74\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n+generateControlVerts\n+unsigned int generateControlVerts\n+generate stencils for control-vertices\n+Definition: stencilTableFactory.h:73\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n+generateOffsets\n+unsigned int generateOffsets\n+populate optional \"_offsets\" field\n+Definition: stencilTableFactory.h:72\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n+fvarChannel\n+unsigned int fvarChannel\n+Definition: stencilTableFactory.h:79\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Options::\n+interpolationMode\n+unsigned int interpolationMode\n+interpolation mode\n+Definition: stencilTableFactory.h:71\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options\n+Definition: stencilTableFactory.h:219\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::\n+Options\n+Options()\n+Definition: stencilTableFactory.h:221\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::\n+fvarChannel\n+unsigned int fvarChannel\n+face-varying channel to use\n+Definition: stencilTableFactory.h:229\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::\n+interpolationMode\n+unsigned int interpolationMode\n+interpolation mode\n+Definition: stencilTableFactory.h:226\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::\n+generate2ndDerivatives\n+unsigned int generate2ndDerivatives\n+Generate weights for 2nd derivatives.\n+Definition: stencilTableFactory.h:228\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Options::\n+generate1stDerivatives\n+unsigned int generate1stDerivatives\n+Generate weights for 1st derivatives.\n+Definition: stencilTableFactory.h:227\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n+LocationArray\n+Descriptor for limit surface locations.\n+Definition: stencilTableFactory.h:233\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n+LocationArray::t\n+REAL const * t\n+array of v coordinates\n+Definition: stencilTableFactory.h:241\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n+LocationArray::s\n+REAL const * s\n+array of u coordinates\n+Definition: stencilTableFactory.h:240\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n+LocationArray::LocationArray\n+LocationArray()\n+Definition: stencilTableFactory.h:235\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n+LocationArray::numLocations\n+int numLocations\n+number of (u,v) coordinates in the array\n+Definition: stencilTableFactory.h:238\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::\n+LocationArray::ptexIdx\n+int ptexIdx\n+ptex face index\n+Definition: stencilTableFactory.h:237\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory\n+Stencil table factory class wrapping the template for compatibility.\n+Definition: stencilTableFactory.h:287\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::\n+AppendLocalPointStencilTableVarying\n+static StencilTable const * AppendLocalPointStencilTableVarying(TopologyRefiner\n+const &refiner, StencilTable const *baseStencilTable, StencilTable const\n+*localPointStencilTable, bool factorize=true)\n+Definition: stencilTableFactory.h:321\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::Create\n+static StencilTable const * Create(TopologyRefiner const &refiner, Options\n+options=Options())\n+Definition: stencilTableFactory.h:293\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::\n+AppendLocalPointStencilTableFaceVarying\n+static StencilTable const * AppendLocalPointStencilTableFaceVarying\n+(TopologyRefiner const &refiner, StencilTable const *baseStencilTable,\n+StencilTable const *localPointStencilTable, int channel=0, bool factorize=true)\n+Definition: stencilTableFactory.h:334\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::\n+AppendLocalPointStencilTable\n+static StencilTable const * AppendLocalPointStencilTable(TopologyRefiner const\n+&refiner, StencilTable const *baseStencilTable, StencilTable const\n+*localPointStencilTable, bool factorize=true)\n+Definition: stencilTableFactory.h:308\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::Create\n+static StencilTable const * Create(int numTables, StencilTable const **tables)\n+Definition: stencilTableFactory.h:300\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactory\n+Stencil table factory class wrapping the template for compatibility.\n+Definition: stencilTableFactory.h:354\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactory::Create\n+static LimitStencilTable const * Create(TopologyRefiner const &refiner,\n+LocationArrayVec const &locationArrays, StencilTable const *cvStencils=0,\n+PatchTable const *patchTable=0, Options options=Options())\n+Definition: stencilTableFactory.h:360\n OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner\n Stores topology data for a specified set of refinement options.\n Definition: topologyRefiner.h:51\n * opensubdiv\n * far\n- * ptexIndices.h\n+ * stencilTableFactory.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00719.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00719.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/primvarRefiner.h File Reference\n+OpenSubdiv: opensubdiv/far/topologyLevel.h File Reference\n \n \n \n \n \n \n \n@@ -83,43 +83,29 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
primvarRefiner.h File Reference
\n+
topologyLevel.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../sdc/types.h"
\n-#include "../sdc/options.h"
\n-#include "../sdc/bilinearScheme.h"
\n-#include "../sdc/catmarkScheme.h"
\n-#include "../sdc/loopScheme.h"
\n-#include "../vtr/level.h"
\n-#include "../vtr/fvarLevel.h"
\n-#include "../vtr/refinement.h"
\n-#include "../vtr/fvarRefinement.h"
\n-#include "../vtr/stackBuffer.h"
\n-#include "../vtr/componentInterfaces.h"
\n+#include "../vtr/level.h"
\n+#include "../vtr/refinement.h"
\n #include "../far/types.h"
\n-#include "../far/error.h"
\n-#include "../far/topologyLevel.h"
\n-#include "../far/topologyRefiner.h"
\n-#include <cassert>
\n+#include <vector>
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n-\n-\n+\n+\n \n

\n Classes

class  PrimvarRefinerReal< REAL >
 Applies refinement operations to generic primvar data. More...
 
class  PrimvarRefiner
class  TopologyLevel
 An interface for accessing data in a specific level of a refined topology hierarchy. More...
 
\n \n \n \n \n@@ -128,13 +114,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,43 +5,30 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-primvarRefiner.h File Reference\n+topologyLevel.h File Reference\n #include \"../version.h\"\n-#include \"../sdc/types.h\"\n-#include \"../sdc/options.h\"\n-#include \"../sdc/bilinearScheme.h\"\n-#include \"../sdc/catmarkScheme.h\"\n-#include \"../sdc/loopScheme.h\"\n #include \"../vtr/level.h\"\n-#include \"../vtr/fvarLevel.h\"\n #include \"../vtr/refinement.h\"\n-#include \"../vtr/fvarRefinement.h\"\n-#include \"../vtr/stackBuffer.h\"\n-#include \"../vtr/componentInterfaces.h\"\n #include \"../far/types.h\"\n-#include \"../far/error.h\"\n-#include \"../far/topologyLevel.h\"\n-#include \"../far/topologyRefiner.h\"\n-#include \n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0PrimvarRefinerReal<_REAL_>\n-\u00a0 Applies refinement operations to generic primvar data. More...\n-\u00a0\n-class \u00a0PrimvarRefiner\n+class \u00a0TopologyLevel\n+\u00a0 An interface for accessing data in a specific level of a refined\n+ topology hierarchy. More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n * opensubdiv\n * far\n- * primvarRefiner.h\n+ * topologyLevel.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00719.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00719.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,3 @@\n var a00719 = [\n- [\"PrimvarRefinerReal< REAL >\", \"a01021.html\", \"a01021\"],\n- [\"PrimvarRefiner\", \"a01029.html\", \"a01029\"]\n+ [\"TopologyLevel\", \"a01105.html\", \"a01105\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00719_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00719_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/primvarRefiner.h Source File\n+OpenSubdiv: opensubdiv/far/topologyLevel.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
primvarRefiner.h
\n+
topologyLevel.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 DreamWorks Animation LLC.
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -106,1200 +106,218 @@\n
17//
\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n-
24#ifndef OPENSUBDIV3_FAR_PRIMVAR_REFINER_H
\n-
25#define OPENSUBDIV3_FAR_PRIMVAR_REFINER_H
\n+
24#ifndef OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H
\n+
25#define OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H
\n
26
\n
27#include "../version.h"
\n
28
\n-
29#include "../sdc/types.h"
\n-
30#include "../sdc/options.h"
\n-
31#include "../sdc/bilinearScheme.h"
\n-
32#include "../sdc/catmarkScheme.h"
\n-
33#include "../sdc/loopScheme.h"
\n-
34#include "../vtr/level.h"
\n-
35#include "../vtr/fvarLevel.h"
\n-
36#include "../vtr/refinement.h"
\n-
37#include "../vtr/fvarRefinement.h"
\n-
38#include "../vtr/stackBuffer.h"
\n-
39#include "../vtr/componentInterfaces.h"
\n-
40#include "../far/types.h"
\n-
41#include "../far/error.h"
\n-
42#include "../far/topologyLevel.h"
\n-
43#include "../far/topologyRefiner.h"
\n-
44
\n-
45#include <cassert>
\n-
46
\n-
47namespace OpenSubdiv {
\n-
48namespace OPENSUBDIV_VERSION {
\n-
49
\n-
50namespace Far {
\n-
51
\n-
55template <typename REAL>
\n-
56class PrimvarRefinerReal {
\n-
57
\n-
58public:
\n-
59 PrimvarRefinerReal(TopologyRefiner const & refiner) : _refiner(refiner) { }
\n-
60 ~PrimvarRefinerReal() { }
\n-
61
\n-
62 TopologyRefiner const & GetTopologyRefiner() const { return _refiner; }
\n+
29#include "../vtr/level.h"
\n+
30#include "../vtr/refinement.h"
\n+
31#include "../far/types.h"
\n+
32
\n+
33#include <vector>
\n+
34
\n+
35namespace OpenSubdiv {
\n+
36namespace OPENSUBDIV_VERSION {
\n+
37
\n+
38namespace Far {
\n+
39
\n+
49class TopologyLevel {
\n+
50
\n+
51public:
\n+
53
\n+
60
\n+
62 int GetNumVertices() const { return _level->getNumVertices(); }
\n
63
\n-
65
\n+
65 int GetNumFaces() const { return _level->getNumFaces(); }
\n+
66
\n+
68 int GetNumEdges() const { return _level->getNumEdges(); }
\n+
69
\n+
71 int GetNumFaceVertices() const { return _level->getNumFaceVerticesTotal(); }
\n+
73
\n+
75
\n+
90 //
\n+
91
\n+
93 ConstIndexArray GetFaceVertices(Index f) const { return _level->getFaceVertices(f); }
\n+
94
\n+
96 ConstIndexArray GetFaceEdges(Index f) const { return _level->getFaceEdges(f); }
\n+
97
\n+
99 ConstIndexArray GetEdgeVertices(Index e) const { return _level->getEdgeVertices(e); }
\n
100
\n-
113 template <class T, class U> void Interpolate(int level, T const & src, U & dst) const;
\n-
114
\n-
130 template <class T, class U> void InterpolateVarying(int level, T const & src, U & dst) const;
\n-
131
\n-
148 template <class T, class U> void InterpolateFaceUniform(int level, T const & src, U & dst) const;
\n-
149
\n-
161 template <class T, class U> void InterpolateFaceVarying(int level, T const & src, U & dst, int channel = 0) const;
\n-
162
\n-
163
\n-
175 template <class T, class U> void Limit(T const & src, U & dstPos) const;
\n-
176
\n-
177 template <class T, class U, class U1, class U2>
\n-
178 void Limit(T const & src, U & dstPos, U1 & dstTan1, U2 & dstTan2) const;
\n-
179
\n-
180 template <class T, class U> void LimitFaceVarying(T const & src, U & dst, int channel = 0) const;
\n-
181
\n+
102 ConstIndexArray GetEdgeFaces(Index e) const { return _level->getEdgeFaces(e); }
\n+
103
\n+
105 ConstIndexArray GetVertexFaces(Index v) const { return _level->getVertexFaces(v); }
\n+
106
\n+
108 ConstIndexArray GetVertexEdges(Index v) const { return _level->getVertexEdges(v); }
\n+
109
\n+
111 ConstLocalIndexArray GetVertexFaceLocalIndices(Index v) const { return _level->getVertexFaceLocalIndices(v); }
\n+
112
\n+
114 ConstLocalIndexArray GetVertexEdgeLocalIndices(Index v) const { return _level->getVertexEdgeLocalIndices(v); }
\n+
115
\n+
117 ConstLocalIndexArray GetEdgeFaceLocalIndices(Index e) const { return _level->getEdgeFaceLocalIndices(e); }
\n+
118
\n+
120 Index FindEdge(Index v0, Index v1) const { return _level->findEdge(v0, v1); }
\n+
122
\n+
124
\n+
126
\n+
128 bool IsEdgeNonManifold(Index e) const { return _level->isEdgeNonManifold(e); }
\n+
129
\n+
131 bool IsVertexNonManifold(Index v) const { return _level->isVertexNonManifold(v); }
\n+
132
\n+
134 bool IsEdgeBoundary(Index e) const { return _level->getEdgeTag(e)._boundary; }
\n+
135
\n+
137 bool IsVertexBoundary(Index v) const { return _level->getVertexTag(v)._boundary; }
\n+
138
\n+
140 bool IsVertexCorner(Index v) const { return (_level->getNumVertexFaces(v) == 1); }
\n+
141
\n+
162 bool IsVertexValenceRegular(Index v) const { return !_level->getVertexTag(v)._xordinary || IsVertexCorner(v); }
\n+
164
\n+
166
\n+
171
\n+
173 float GetEdgeSharpness(Index e) const { return _level->getEdgeSharpness(e); }
\n+
174
\n+
176 float GetVertexSharpness(Index v) const { return _level->getVertexSharpness(v); }
\n+
177
\n+
179 bool IsEdgeInfSharp(Index e) const { return _level->getEdgeTag(e)._infSharp; }
\n+
180
\n+
182 bool IsVertexInfSharp(Index v) const { return _level->getVertexTag(v)._infSharp; }
\n
183
\n-
184private:
\n-
185 typedef REAL Weight;
\n+
185 bool IsEdgeSemiSharp(Index e) const { return _level->getEdgeTag(e)._semiSharp; }
\n
186
\n-
187 // Non-copyable:
\n-
188 PrimvarRefinerReal(PrimvarRefinerReal const & src) : _refiner(src._refiner) { }
\n-
189 PrimvarRefinerReal & operator=(PrimvarRefinerReal const &) { return *this; }
\n-
190
\n-
191 template <Sdc::SchemeType SCHEME, class T, class U> void interpFromFaces(int, T const &, U &) const;
\n-
192 template <Sdc::SchemeType SCHEME, class T, class U> void interpFromEdges(int, T const &, U &) const;
\n-
193 template <Sdc::SchemeType SCHEME, class T, class U> void interpFromVerts(int, T const &, U &) const;
\n-
194
\n-
195 template <Sdc::SchemeType SCHEME, class T, class U> void interpFVarFromFaces(int, T const &, U &, int) const;
\n-
196 template <Sdc::SchemeType SCHEME, class T, class U> void interpFVarFromEdges(int, T const &, U &, int) const;
\n-
197 template <Sdc::SchemeType SCHEME, class T, class U> void interpFVarFromVerts(int, T const &, U &, int) const;
\n+
188 bool IsVertexSemiSharp(Index v) const { return _level->getVertexTag(v)._semiSharp; }
\n+
189
\n+
191 bool IsFaceHole(Index f) const { return _level->isFaceHole(f); }
\n+
192
\n+
194 Sdc::Crease::Rule GetVertexRule(Index v) const { return _level->getVertexRule(v); }
\n+
196
\n
198
\n-
199 template <Sdc::SchemeType SCHEME, class T, class U, class U1, class U2>
\n-
200 void limit(T const & src, U & pos, U1 * tan1, U2 * tan2) const;
\n-
201
\n-
202 template <Sdc::SchemeType SCHEME, class T, class U>
\n-
203 void limitFVar(T const & src, U & dst, int channel) const;
\n-
204
\n-
205private:
\n-
206 TopologyRefiner const & _refiner;
\n-
207
\n-
208private:
\n-
209 //
\n-
210 // Local class to fulfill interface for <typename MASK> in the Scheme mask queries:
\n-
211 //
\n-
212 class Mask {
\n-
213 public:
\n-
214 typedef REAL Weight; // Also part of the expected interface
\n-
215
\n-
216 public:
\n-
217 Mask(Weight* v, Weight* e, Weight* f) :
\n-
218 _vertWeights(v), _edgeWeights(e), _faceWeights(f),
\n-
219 _vertCount(0), _edgeCount(0), _faceCount(0),
\n-
220 _faceWeightsForFaceCenters(false)
\n-
221 { }
\n-
222
\n-
223 ~Mask() { }
\n-
224
\n-
225 public: // Generic interface expected of <typename MASK>:
\n-
226 int GetNumVertexWeights() const { return _vertCount; }
\n-
227 int GetNumEdgeWeights() const { return _edgeCount; }
\n-
228 int GetNumFaceWeights() const { return _faceCount; }
\n-
229
\n-
230 void SetNumVertexWeights(int count) { _vertCount = count; }
\n-
231 void SetNumEdgeWeights( int count) { _edgeCount = count; }
\n-
232 void SetNumFaceWeights( int count) { _faceCount = count; }
\n-
233
\n-
234 Weight const& VertexWeight(int index) const { return _vertWeights[index]; }
\n-
235 Weight const& EdgeWeight( int index) const { return _edgeWeights[index]; }
\n-
236 Weight const& FaceWeight( int index) const { return _faceWeights[index]; }
\n-
237
\n-
238 Weight& VertexWeight(int index) { return _vertWeights[index]; }
\n-
239 Weight& EdgeWeight( int index) { return _edgeWeights[index]; }
\n-
240 Weight& FaceWeight( int index) { return _faceWeights[index]; }
\n-
241
\n-
242 bool AreFaceWeightsForFaceCenters() const { return _faceWeightsForFaceCenters; }
\n-
243 void SetFaceWeightsForFaceCenters(bool on) { _faceWeightsForFaceCenters = on; }
\n-
244
\n-
245 private:
\n-
246 Weight* _vertWeights;
\n-
247 Weight* _edgeWeights;
\n-
248 Weight* _faceWeights;
\n-
249
\n-
250 int _vertCount;
\n-
251 int _edgeCount;
\n-
252 int _faceCount;
\n-
253
\n-
254 bool _faceWeightsForFaceCenters;
\n-
255 };
\n-
256};
\n+
228
\n+
230 int GetNumFVarChannels() const { return _level->getNumFVarChannels(); }
\n+
231
\n+
234 int GetNumFVarValues(int channel = 0) const { return _level->getNumFVarValues(channel); }
\n+
235
\n+
237 ConstIndexArray GetFaceFVarValues(Index f, int channel = 0) const {
\n+
238 return _level->getFaceFVarValues(f, channel);
\n+
239 }
\n+
240
\n+
242 bool DoesVertexFVarTopologyMatch(Index v, int channel = 0) const {
\n+
243 return _level->doesVertexFVarTopologyMatch(v, channel);
\n+
244 }
\n+
245
\n+
247 bool DoesEdgeFVarTopologyMatch(Index e, int channel = 0) const {
\n+
248 return _level->doesEdgeFVarTopologyMatch(e, channel);
\n+
249 }
\n+
250
\n+
252 bool DoesFaceFVarTopologyMatch(Index f, int channel = 0) const {
\n+
253 return _level->doesFaceFVarTopologyMatch(f, channel);
\n+
254 }
\n+
255
\n
257
\n-
258
\n-
259//
\n-
260// Public entry points to the methods. Queries of the scheme type and its
\n-
261// use as a template parameter in subsequent implementation will be factored
\n-
262// out of a later release:
\n-
263//
\n-
264template <typename REAL>
\n-
265template <class T, class U>
\n-
266inline void
\n-
267PrimvarRefinerReal<REAL>::Interpolate(int level, T const & src, U & dst) const {
\n-
268
\n-
269 assert(level>0 && level<=(int)_refiner._refinements.size());
\n-
270
\n-
271 switch (_refiner._subdivType) {
\n-
272 case Sdc::SCHEME_CATMARK:
\n-
273 interpFromFaces<Sdc::SCHEME_CATMARK>(level, src, dst);
\n-
274 interpFromEdges<Sdc::SCHEME_CATMARK>(level, src, dst);
\n-
275 interpFromVerts<Sdc::SCHEME_CATMARK>(level, src, dst);
\n-
276 break;
\n-
277 case Sdc::SCHEME_LOOP:
\n-
278 interpFromFaces<Sdc::SCHEME_LOOP>(level, src, dst);
\n-
279 interpFromEdges<Sdc::SCHEME_LOOP>(level, src, dst);
\n-
280 interpFromVerts<Sdc::SCHEME_LOOP>(level, src, dst);
\n-
281 break;
\n-
282 case Sdc::SCHEME_BILINEAR:
\n-
283 interpFromFaces<Sdc::SCHEME_BILINEAR>(level, src, dst);
\n-
284 interpFromEdges<Sdc::SCHEME_BILINEAR>(level, src, dst);
\n-
285 interpFromVerts<Sdc::SCHEME_BILINEAR>(level, src, dst);
\n-
286 break;
\n-
287 }
\n-
288}
\n+
259
\n+
260
\n+
262 ConstIndexArray GetFaceChildFaces(Index f) const { return _refToChild->getFaceChildFaces(f); }
\n+
263
\n+
265 ConstIndexArray GetFaceChildEdges(Index f) const { return _refToChild->getFaceChildEdges(f); }
\n+
266
\n+
268 ConstIndexArray GetEdgeChildEdges(Index e) const { return _refToChild->getEdgeChildEdges(e); }
\n+
269
\n+
271 Index GetFaceChildVertex( Index f) const { return _refToChild->getFaceChildVertex(f); }
\n+
272
\n+
274 Index GetEdgeChildVertex( Index e) const { return _refToChild->getEdgeChildVertex(e); }
\n+
275
\n+
277 Index GetVertexChildVertex(Index v) const { return _refToChild->getVertexChildVertex(v); }
\n+
278
\n+
280 Index GetFaceParentFace(Index f) const { return _refToParent->getChildFaceParentFace(f); }
\n+
282
\n+
284
\n+
285
\n+
286 bool ValidateTopology() const { return _level->validateTopology(); }
\n+
287 void PrintTopology(bool children = true) const { _level->print((children && _refToChild) ? _refToChild : 0); }
\n
289
\n-
290template <typename REAL>
\n-
291template <class T, class U>
\n-
292inline void
\n-
293PrimvarRefinerReal<REAL>::InterpolateFaceVarying(int level, T const & src, U & dst, int channel) const {
\n-
294
\n-
295 assert(level>0 && level<=(int)_refiner._refinements.size());
\n-
296
\n-
297 switch (_refiner._subdivType) {
\n-
298 case Sdc::SCHEME_CATMARK:
\n-
299 interpFVarFromFaces<Sdc::SCHEME_CATMARK>(level, src, dst, channel);
\n-
300 interpFVarFromEdges<Sdc::SCHEME_CATMARK>(level, src, dst, channel);
\n-
301 interpFVarFromVerts<Sdc::SCHEME_CATMARK>(level, src, dst, channel);
\n-
302 break;
\n-
303 case Sdc::SCHEME_LOOP:
\n-
304 interpFVarFromFaces<Sdc::SCHEME_LOOP>(level, src, dst, channel);
\n-
305 interpFVarFromEdges<Sdc::SCHEME_LOOP>(level, src, dst, channel);
\n-
306 interpFVarFromVerts<Sdc::SCHEME_LOOP>(level, src, dst, channel);
\n-
307 break;
\n-
308 case Sdc::SCHEME_BILINEAR:
\n-
309 interpFVarFromFaces<Sdc::SCHEME_BILINEAR>(level, src, dst, channel);
\n-
310 interpFVarFromEdges<Sdc::SCHEME_BILINEAR>(level, src, dst, channel);
\n-
311 interpFVarFromVerts<Sdc::SCHEME_BILINEAR>(level, src, dst, channel);
\n-
312 break;
\n-
313 }
\n-
314}
\n-
315
\n-
316template <typename REAL>
\n-
317template <class T, class U>
\n-
318inline void
\n-
319PrimvarRefinerReal<REAL>::Limit(T const & src, U & dst) const {
\n-
320
\n-
321 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() == 0) {
\n-
322 Error(FAR_RUNTIME_ERROR,
\n-
323 "Failure in PrimvarRefiner::Limit() -- "
\n-
324 "last level of refinement does not include full topology.");
\n-
325 return;
\n-
326 }
\n-
327
\n-
328 switch (_refiner._subdivType) {
\n-
329 case Sdc::SCHEME_CATMARK:
\n-
330 limit<Sdc::SCHEME_CATMARK>(src, dst, (U*)0, (U*)0);
\n-
331 break;
\n-
332 case Sdc::SCHEME_LOOP:
\n-
333 limit<Sdc::SCHEME_LOOP>(src, dst, (U*)0, (U*)0);
\n-
334 break;
\n-
335 case Sdc::SCHEME_BILINEAR:
\n-
336 limit<Sdc::SCHEME_BILINEAR>(src, dst, (U*)0, (U*)0);
\n-
337 break;
\n-
338 }
\n-
339}
\n-
340
\n-
341template <typename REAL>
\n-
342template <class T, class U, class U1, class U2>
\n-
343inline void
\n-
344PrimvarRefinerReal<REAL>::Limit(T const & src, U & dstPos, U1 & dstTan1, U2 & dstTan2) const {
\n-
345
\n-
346 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() == 0) {
\n-
347 Error(FAR_RUNTIME_ERROR,
\n-
348 "Failure in PrimvarRefiner::Limit() -- "
\n-
349 "last level of refinement does not include full topology.");
\n-
350 return;
\n-
351 }
\n-
352
\n-
353 switch (_refiner._subdivType) {
\n-
354 case Sdc::SCHEME_CATMARK:
\n-
355 limit<Sdc::SCHEME_CATMARK>(src, dstPos, &dstTan1, &dstTan2);
\n-
356 break;
\n-
357 case Sdc::SCHEME_LOOP:
\n-
358 limit<Sdc::SCHEME_LOOP>(src, dstPos, &dstTan1, &dstTan2);
\n-
359 break;
\n-
360 case Sdc::SCHEME_BILINEAR:
\n-
361 limit<Sdc::SCHEME_BILINEAR>(src, dstPos, &dstTan1, &dstTan2);
\n-
362 break;
\n-
363 }
\n-
364}
\n-
365
\n-
366template <typename REAL>
\n-
367template <class T, class U>
\n-
368inline void
\n-
369PrimvarRefinerReal<REAL>::LimitFaceVarying(T const & src, U & dst, int channel) const {
\n-
370
\n-
371 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() == 0) {
\n-
372 Error(FAR_RUNTIME_ERROR,
\n-
373 "Failure in PrimvarRefiner::LimitFaceVarying() -- "
\n-
374 "last level of refinement does not include full topology.");
\n-
375 return;
\n-
376 }
\n-
377
\n-
378 switch (_refiner._subdivType) {
\n-
379 case Sdc::SCHEME_CATMARK:
\n-
380 limitFVar<Sdc::SCHEME_CATMARK>(src, dst, channel);
\n-
381 break;
\n-
382 case Sdc::SCHEME_LOOP:
\n-
383 limitFVar<Sdc::SCHEME_LOOP>(src, dst, channel);
\n-
384 break;
\n-
385 case Sdc::SCHEME_BILINEAR:
\n-
386 limitFVar<Sdc::SCHEME_BILINEAR>(src, dst, channel);
\n-
387 break;
\n-
388 }
\n-
389}
\n-
390
\n-
391template <typename REAL>
\n-
392template <class T, class U>
\n-
393inline void
\n-
394PrimvarRefinerReal<REAL>::InterpolateFaceUniform(int level, T const & src, U & dst) const {
\n-
395
\n-
396 assert(level>0 && level<=(int)_refiner._refinements.size());
\n-
397
\n-
398 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n-
399 Vtr::internal::Level const & child = refinement.child();
\n-
400
\n-
401 for (int cFace = 0; cFace < child.getNumFaces(); ++cFace) {
\n-
402
\n-
403 Vtr::Index pFace = refinement.getChildFaceParentFace(cFace);
\n-
404
\n-
405 dst[cFace] = src[pFace];
\n-
406 }
\n-
407}
\n-
408
\n-
409template <typename REAL>
\n-
410template <class T, class U>
\n-
411inline void
\n-
412PrimvarRefinerReal<REAL>::InterpolateVarying(int level, T const & src, U & dst) const {
\n-
413
\n-
414 assert(level>0 && level<=(int)_refiner._refinements.size());
\n-
415
\n-
416 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n-
417 Vtr::internal::Level const & parent = refinement.parent();
\n-
418
\n-
419 //
\n-
420 // Group values to interpolate based on origin -- note that there may
\n-
421 // be none originating from faces:
\n-
422 //
\n-
423 if (refinement.getNumChildVerticesFromFaces() > 0) {
\n-
424
\n-
425 for (int face = 0; face < parent.getNumFaces(); ++face) {
\n-
426
\n-
427 Vtr::Index cVert = refinement.getFaceChildVertex(face);
\n-
428 if (Vtr::IndexIsValid(cVert)) {
\n-
429
\n-
430 // Apply the weights to the parent face's vertices:
\n-
431 ConstIndexArray fVerts = parent.getFaceVertices(face);
\n-
432
\n-
433 Weight fVaryingWeight = 1.0f / (Weight) fVerts.size();
\n-
434
\n-
435 dst[cVert].Clear();
\n-
436 for (int i = 0; i < fVerts.size(); ++i) {
\n-
437 dst[cVert].AddWithWeight(src[fVerts[i]], fVaryingWeight);
\n-
438 }
\n-
439 }
\n-
440 }
\n-
441 }
\n-
442 for (int edge = 0; edge < parent.getNumEdges(); ++edge) {
\n-
443
\n-
444 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);
\n-
445 if (Vtr::IndexIsValid(cVert)) {
\n-
446
\n-
447 // Apply the weights to the parent edges's vertices
\n-
448 ConstIndexArray eVerts = parent.getEdgeVertices(edge);
\n-
449
\n-
450 dst[cVert].Clear();
\n-
451 dst[cVert].AddWithWeight(src[eVerts[0]], 0.5f);
\n-
452 dst[cVert].AddWithWeight(src[eVerts[1]], 0.5f);
\n-
453 }
\n-
454 }
\n-
455 for (int vert = 0; vert < parent.getNumVertices(); ++vert) {
\n-
456
\n-
457 Vtr::Index cVert = refinement.getVertexChildVertex(vert);
\n-
458 if (Vtr::IndexIsValid(cVert)) {
\n-
459
\n-
460 // Essentially copy the parent vertex:
\n-
461 dst[cVert].Clear();
\n-
462 dst[cVert].AddWithWeight(src[vert], 1.0f);
\n-
463 }
\n-
464 }
\n-
465}
\n-
466
\n-
467
\n-
468//
\n-
469// Internal implementation methods -- grouping vertices to be interpolated
\n-
470// based on the type of parent component from which they originated:
\n-
471//
\n-
472template <typename REAL>
\n-
473template <Sdc::SchemeType SCHEME, class T, class U>
\n-
474inline void
\n-
475PrimvarRefinerReal<REAL>::interpFromFaces(int level, T const & src, U & dst) const {
\n-
476
\n-
477 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n-
478 Vtr::internal::Level const & parent = refinement.parent();
\n-
479
\n-
480 if (refinement.getNumChildVerticesFromFaces() == 0) return;
\n-
481
\n-
482 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n-
483
\n-
484 Vtr::internal::StackBuffer<Weight,16> fVertWeights(parent.getMaxValence());
\n-
485
\n-
486 for (int face = 0; face < parent.getNumFaces(); ++face) {
\n-
487
\n-
488 Vtr::Index cVert = refinement.getFaceChildVertex(face);
\n-
489 if (!Vtr::IndexIsValid(cVert))
\n-
490 continue;
\n-
491
\n-
492 // Declare and compute mask weights for this vertex relative to its parent face:
\n-
493 ConstIndexArray fVerts = parent.getFaceVertices(face);
\n-
494
\n-
495 Mask fMask(fVertWeights, 0, 0);
\n-
496 Vtr::internal::FaceInterface fHood(fVerts.size());
\n-
497
\n-
498 scheme.ComputeFaceVertexMask(fHood, fMask);
\n-
499
\n-
500 // Apply the weights to the parent face's vertices:
\n-
501 dst[cVert].Clear();
\n-
502
\n-
503 for (int i = 0; i < fVerts.size(); ++i) {
\n-
504
\n-
505 dst[cVert].AddWithWeight(src[fVerts[i]], fVertWeights[i]);
\n-
506 }
\n-
507 }
\n-
508}
\n-
509
\n-
510template <typename REAL>
\n-
511template <Sdc::SchemeType SCHEME, class T, class U>
\n-
512inline void
\n-
513PrimvarRefinerReal<REAL>::interpFromEdges(int level, T const & src, U & dst) const {
\n-
514
\n-
515 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n-
516 Vtr::internal::Level const & parent = refinement.parent();
\n-
517 Vtr::internal::Level const & child = refinement.child();
\n-
518
\n-
519 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n-
520
\n-
521 Vtr::internal::EdgeInterface eHood(parent);
\n-
522
\n-
523 Weight eVertWeights[2];
\n-
524 Vtr::internal::StackBuffer<Weight,8> eFaceWeights(parent.getMaxEdgeFaces());
\n-
525
\n-
526 for (int edge = 0; edge < parent.getNumEdges(); ++edge) {
\n-
527
\n-
528 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);
\n-
529 if (!Vtr::IndexIsValid(cVert))
\n-
530 continue;
\n-
531
\n-
532 // Declare and compute mask weights for this vertex relative to its parent edge:
\n-
533 ConstIndexArray eVerts = parent.getEdgeVertices(edge),
\n-
534 eFaces = parent.getEdgeFaces(edge);
\n-
535
\n-
536 Mask eMask(eVertWeights, 0, eFaceWeights);
\n-
537
\n-
538 eHood.SetIndex(edge);
\n-
539
\n-
540 Sdc::Crease::Rule pRule = (parent.getEdgeSharpness(edge) > 0.0f) ? Sdc::Crease::RULE_CREASE : Sdc::Crease::RULE_SMOOTH;
\n-
541 Sdc::Crease::Rule cRule = child.getVertexRule(cVert);
\n-
542
\n-
543 scheme.ComputeEdgeVertexMask(eHood, eMask, pRule, cRule);
\n-
544
\n-
545 // Apply the weights to the parent edges's vertices and (if applicable) to
\n-
546 // the child vertices of its incident faces:
\n-
547 dst[cVert].Clear();
\n-
548 dst[cVert].AddWithWeight(src[eVerts[0]], eVertWeights[0]);
\n-
549 dst[cVert].AddWithWeight(src[eVerts[1]], eVertWeights[1]);
\n-
550
\n-
551 if (eMask.GetNumFaceWeights() > 0) {
\n-
552
\n-
553 for (int i = 0; i < eFaces.size(); ++i) {
\n-
554
\n-
555 if (eMask.AreFaceWeightsForFaceCenters()) {
\n-
556 assert(refinement.getNumChildVerticesFromFaces() > 0);
\n-
557 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(eFaces[i]);
\n-
558
\n-
559 assert(Vtr::IndexIsValid(cVertOfFace));
\n-
560 dst[cVert].AddWithWeight(dst[cVertOfFace], eFaceWeights[i]);
\n-
561 } else {
\n-
562 Vtr::Index pFace = eFaces[i];
\n-
563 ConstIndexArray pFaceEdges = parent.getFaceEdges(pFace),
\n-
564 pFaceVerts = parent.getFaceVertices(pFace);
\n-
565
\n-
566 int eInFace = 0;
\n-
567 for ( ; pFaceEdges[eInFace] != edge; ++eInFace ) ;
\n-
568
\n-
569 int vInFace = eInFace + 2;
\n-
570 if (vInFace >= pFaceVerts.size()) vInFace -= pFaceVerts.size();
\n-
571
\n-
572 Vtr::Index pVertNext = pFaceVerts[vInFace];
\n-
573 dst[cVert].AddWithWeight(src[pVertNext], eFaceWeights[i]);
\n-
574 }
\n-
575 }
\n-
576 }
\n-
577 }
\n-
578}
\n-
579
\n-
580template <typename REAL>
\n-
581template <Sdc::SchemeType SCHEME, class T, class U>
\n-
582inline void
\n-
583PrimvarRefinerReal<REAL>::interpFromVerts(int level, T const & src, U & dst) const {
\n-
584
\n-
585 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n-
586 Vtr::internal::Level const & parent = refinement.parent();
\n-
587 Vtr::internal::Level const & child = refinement.child();
\n-
588
\n-
589 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n-
590
\n-
591 Vtr::internal::VertexInterface vHood(parent, child);
\n-
592
\n-
593 Vtr::internal::StackBuffer<Weight,32> weightBuffer(2*parent.getMaxValence());
\n-
594
\n-
595 for (int vert = 0; vert < parent.getNumVertices(); ++vert) {
\n-
596
\n-
597 Vtr::Index cVert = refinement.getVertexChildVertex(vert);
\n-
598 if (!Vtr::IndexIsValid(cVert))
\n-
599 continue;
\n-
600
\n-
601 // Declare and compute mask weights for this vertex relative to its parent edge:
\n-
602 ConstIndexArray vEdges = parent.getVertexEdges(vert),
\n-
603 vFaces = parent.getVertexFaces(vert);
\n-
604
\n-
605 Weight vVertWeight,
\n-
606 * vEdgeWeights = weightBuffer,
\n-
607 * vFaceWeights = vEdgeWeights + vEdges.size();
\n-
608
\n-
609 Mask vMask(&vVertWeight, vEdgeWeights, vFaceWeights);
\n-
610
\n-
611 vHood.SetIndex(vert, cVert);
\n-
612
\n-
613 Sdc::Crease::Rule pRule = parent.getVertexRule(vert);
\n-
614 Sdc::Crease::Rule cRule = child.getVertexRule(cVert);
\n-
615
\n-
616 scheme.ComputeVertexVertexMask(vHood, vMask, pRule, cRule);
\n-
617
\n-
618 // Apply the weights to the parent vertex, the vertices opposite its incident
\n-
619 // edges, and the child vertices of its incident faces:
\n-
620 //
\n-
621 // In order to improve numerical precision, it's better to apply smaller weights
\n-
622 // first, so begin with the face-weights followed by the edge-weights and the
\n-
623 // vertex weight last.
\n-
624 dst[cVert].Clear();
\n-
625
\n-
626 if (vMask.GetNumFaceWeights() > 0) {
\n-
627 assert(vMask.AreFaceWeightsForFaceCenters());
\n-
628
\n-
629 for (int i = 0; i < vFaces.size(); ++i) {
\n-
630
\n-
631 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(vFaces[i]);
\n-
632 assert(Vtr::IndexIsValid(cVertOfFace));
\n-
633 dst[cVert].AddWithWeight(dst[cVertOfFace], vFaceWeights[i]);
\n-
634 }
\n-
635 }
\n-
636 if (vMask.GetNumEdgeWeights() > 0) {
\n-
637
\n-
638 for (int i = 0; i < vEdges.size(); ++i) {
\n-
639
\n-
640 ConstIndexArray eVerts = parent.getEdgeVertices(vEdges[i]);
\n-
641 Vtr::Index pVertOppositeEdge = (eVerts[0] == vert) ? eVerts[1] : eVerts[0];
\n-
642
\n-
643 dst[cVert].AddWithWeight(src[pVertOppositeEdge], vEdgeWeights[i]);
\n-
644 }
\n-
645 }
\n-
646 dst[cVert].AddWithWeight(src[vert], vVertWeight);
\n-
647 }
\n-
648}
\n-
649
\n-
650
\n-
651//
\n-
652// Internal face-varying implementation details:
\n-
653//
\n-
654template <typename REAL>
\n-
655template <Sdc::SchemeType SCHEME, class T, class U>
\n-
656inline void
\n-
657PrimvarRefinerReal<REAL>::interpFVarFromFaces(int level, T const & src, U & dst, int channel) const {
\n-
658
\n-
659 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n-
660
\n-
661 if (refinement.getNumChildVerticesFromFaces() == 0) return;
\n-
662
\n-
663 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n-
664
\n-
665 Vtr::internal::Level const & parentLevel = refinement.parent();
\n-
666 Vtr::internal::Level const & childLevel = refinement.child();
\n-
667
\n-
668 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel(channel);
\n-
669 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel(channel);
\n-
670
\n-
671 Vtr::internal::StackBuffer<Weight,16> fValueWeights(parentLevel.getMaxValence());
\n-
672
\n-
673 for (int face = 0; face < parentLevel.getNumFaces(); ++face) {
\n-
674
\n-
675 Vtr::Index cVert = refinement.getFaceChildVertex(face);
\n-
676 if (!Vtr::IndexIsValid(cVert))
\n-
677 continue;
\n-
678
\n-
679 Vtr::Index cVertValue = childFVar.getVertexValueOffset(cVert);
\n-
680
\n-
681 // The only difference for face-varying here is that we get the values associated
\n-
682 // with each face-vertex directly from the FVarLevel, rather than using the parent
\n-
683 // face-vertices directly. If any face-vertex has any sibling values, then we may
\n-
684 // get the wrong one using the face-vertex index directly.
\n-
685
\n-
686 // Declare and compute mask weights for this vertex relative to its parent face:
\n-
687 ConstIndexArray fValues = parentFVar.getFaceValues(face);
\n-
688
\n-
689 Mask fMask(fValueWeights, 0, 0);
\n-
690 Vtr::internal::FaceInterface fHood(fValues.size());
\n-
691
\n-
692 scheme.ComputeFaceVertexMask(fHood, fMask);
\n-
693
\n-
694 // Apply the weights to the parent face's vertices:
\n-
695 dst[cVertValue].Clear();
\n-
696
\n-
697 for (int i = 0; i < fValues.size(); ++i) {
\n-
698 dst[cVertValue].AddWithWeight(src[fValues[i]], fValueWeights[i]);
\n-
699 }
\n-
700 }
\n-
701}
\n-
702
\n-
703template <typename REAL>
\n-
704template <Sdc::SchemeType SCHEME, class T, class U>
\n-
705inline void
\n-
706PrimvarRefinerReal<REAL>::interpFVarFromEdges(int level, T const & src, U & dst, int channel) const {
\n-
707
\n-
708 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n-
709
\n-
710 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n-
711
\n-
712 Vtr::internal::Level const & parentLevel = refinement.parent();
\n-
713 Vtr::internal::Level const & childLevel = refinement.child();
\n-
714
\n-
715 Vtr::internal::FVarRefinement const & refineFVar = refinement.getFVarRefinement(channel);
\n-
716 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel(channel);
\n-
717 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel(channel);
\n-
718
\n-
719 //
\n-
720 // Allocate and initialize (if linearly interpolated) interpolation weights for
\n-
721 // the edge mask:
\n-
722 //
\n-
723 Weight eVertWeights[2];
\n-
724 Vtr::internal::StackBuffer<Weight,8> eFaceWeights(parentLevel.getMaxEdgeFaces());
\n-
725
\n-
726 Mask eMask(eVertWeights, 0, eFaceWeights);
\n-
727
\n-
728 bool isLinearFVar = parentFVar.isLinear() || (_refiner._subdivType == Sdc::SCHEME_BILINEAR);
\n-
729 if (isLinearFVar) {
\n-
730 eMask.SetNumVertexWeights(2);
\n-
731 eMask.SetNumEdgeWeights(0);
\n-
732 eMask.SetNumFaceWeights(0);
\n-
733
\n-
734 eVertWeights[0] = 0.5f;
\n-
735 eVertWeights[1] = 0.5f;
\n-
736 }
\n-
737
\n-
738 Vtr::internal::EdgeInterface eHood(parentLevel);
\n-
739
\n-
740 for (int edge = 0; edge < parentLevel.getNumEdges(); ++edge) {
\n-
741
\n-
742 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);
\n-
743 if (!Vtr::IndexIsValid(cVert))
\n-
744 continue;
\n-
745
\n-
746 ConstIndexArray cVertValues = childFVar.getVertexValues(cVert);
\n-
747
\n-
748 bool fvarEdgeVertMatchesVertex = childFVar.valueTopologyMatches(cVertValues[0]);
\n-
749 if (fvarEdgeVertMatchesVertex) {
\n-
750 //
\n-
751 // If smoothly interpolated, compute new weights for the edge mask:
\n-
752 //
\n-
753 if (!isLinearFVar) {
\n-
754 eHood.SetIndex(edge);
\n-
755
\n-
756 Sdc::Crease::Rule pRule = (parentLevel.getEdgeSharpness(edge) > 0.0f)
\n-
757 ? Sdc::Crease::RULE_CREASE : Sdc::Crease::RULE_SMOOTH;
\n-
758 Sdc::Crease::Rule cRule = childLevel.getVertexRule(cVert);
\n-
759
\n-
760 scheme.ComputeEdgeVertexMask(eHood, eMask, pRule, cRule);
\n-
761 }
\n-
762
\n-
763 // Apply the weights to the parent edge's vertices and (if applicable) to
\n-
764 // the child vertices of its incident faces:
\n-
765 //
\n-
766 // Even though the face-varying topology matches the vertex topology, we need
\n-
767 // to be careful here when getting values corresponding to the two end-vertices.
\n-
768 // While the edge may be continuous, the vertices at their ends may have
\n-
769 // discontinuities elsewhere in their neighborhood (i.e. on the "other side"
\n-
770 // of the end-vertex) and so have sibling values associated with them. In most
\n-
771 // cases the topology for an end-vertex will match and we can use it directly,
\n-
772 // but we must still check and retrieve as needed.
\n-
773 //
\n-
774 // Indices for values corresponding to face-vertices are guaranteed to match,
\n-
775 // so we can use the child-vertex indices directly.
\n-
776 //
\n-
777 // And by "directly", we always use getVertexValue(vertexIndex) to reference
\n-
778 // values in the "src" to account for the possible indirection that may exist at
\n-
779 // level 0 -- where there may be fewer values than vertices and an additional
\n-
780 // indirection is necessary. We can use a vertex index directly for "dst" when
\n-
781 // it matches.
\n-
782 //
\n-
783 Vtr::Index eVertValues[2];
\n-
784
\n-
785 parentFVar.getEdgeFaceValues(edge, 0, eVertValues);
\n-
786
\n-
787 Index cVertValue = cVertValues[0];
\n-
788
\n-
789 dst[cVertValue].Clear();
\n-
790 dst[cVertValue].AddWithWeight(src[eVertValues[0]], eVertWeights[0]);
\n-
791 dst[cVertValue].AddWithWeight(src[eVertValues[1]], eVertWeights[1]);
\n-
792
\n-
793 if (eMask.GetNumFaceWeights() > 0) {
\n-
794
\n-
795 ConstIndexArray eFaces = parentLevel.getEdgeFaces(edge);
\n-
796
\n-
797 for (int i = 0; i < eFaces.size(); ++i) {
\n-
798 if (eMask.AreFaceWeightsForFaceCenters()) {
\n-
799
\n-
800 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(eFaces[i]);
\n-
801 assert(Vtr::IndexIsValid(cVertOfFace));
\n-
802
\n-
803 Vtr::Index cValueOfFace = childFVar.getVertexValueOffset(cVertOfFace);
\n-
804 dst[cVertValue].AddWithWeight(dst[cValueOfFace], eFaceWeights[i]);
\n-
805 } else {
\n-
806 Vtr::Index pFace = eFaces[i];
\n-
807 ConstIndexArray pFaceEdges = parentLevel.getFaceEdges(pFace),
\n-
808 pFaceVerts = parentLevel.getFaceVertices(pFace);
\n-
809
\n-
810 int eInFace = 0;
\n-
811 for ( ; pFaceEdges[eInFace] != edge; ++eInFace ) ;
\n-
812
\n-
813 // Edge "i" spans vertices [i,i+1] so we want i+2...
\n-
814 int vInFace = eInFace + 2;
\n-
815 if (vInFace >= pFaceVerts.size()) vInFace -= pFaceVerts.size();
\n-
816
\n-
817 Vtr::Index pValueNext = parentFVar.getFaceValues(pFace)[vInFace];
\n-
818 dst[cVertValue].AddWithWeight(src[pValueNext], eFaceWeights[i]);
\n-
819 }
\n-
820 }
\n-
821 }
\n-
822 } else {
\n-
823 //
\n-
824 // Mismatched edge-verts should just be linearly interpolated between the pairs of
\n-
825 // values for each sibling of the child edge-vertex -- the question is: which face
\n-
826 // holds that pair of values for a given sibling?
\n-
827 //
\n-
828 // In the manifold case, the sibling and edge-face indices will correspond. We
\n-
829 // will eventually need to update this to account for > 3 incident faces.
\n-
830 //
\n-
831 for (int i = 0; i < cVertValues.size(); ++i) {
\n-
832 Vtr::Index eVertValues[2];
\n-
833 int eFaceIndex = refineFVar.getChildValueParentSource(cVert, i);
\n-
834 assert(eFaceIndex == i);
\n-
835
\n-
836 parentFVar.getEdgeFaceValues(edge, eFaceIndex, eVertValues);
\n-
837
\n-
838 Index cVertValue = cVertValues[i];
\n-
839
\n-
840 dst[cVertValue].Clear();
\n-
841 dst[cVertValue].AddWithWeight(src[eVertValues[0]], 0.5);
\n-
842 dst[cVertValue].AddWithWeight(src[eVertValues[1]], 0.5);
\n-
843 }
\n-
844 }
\n-
845 }
\n-
846}
\n-
847
\n-
848template <typename REAL>
\n-
849template <Sdc::SchemeType SCHEME, class T, class U>
\n-
850inline void
\n-
851PrimvarRefinerReal<REAL>::interpFVarFromVerts(int level, T const & src, U & dst, int channel) const {
\n-
852
\n-
853 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n-
854
\n-
855 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n-
856
\n-
857 Vtr::internal::Level const & parentLevel = refinement.parent();
\n-
858 Vtr::internal::Level const & childLevel = refinement.child();
\n-
859
\n-
860 Vtr::internal::FVarRefinement const & refineFVar = refinement.getFVarRefinement(channel);
\n-
861 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel(channel);
\n-
862 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel(channel);
\n-
863
\n-
864 bool isLinearFVar = parentFVar.isLinear() || (_refiner._subdivType == Sdc::SCHEME_BILINEAR);
\n-
865
\n-
866 Vtr::internal::StackBuffer<Weight,32> weightBuffer(2*parentLevel.getMaxValence());
\n-
867
\n-
868 Vtr::internal::StackBuffer<Vtr::Index,16> vEdgeValues(parentLevel.getMaxValence());
\n-
869
\n-
870 Vtr::internal::VertexInterface vHood(parentLevel, childLevel);
\n-
871
\n-
872 for (int vert = 0; vert < parentLevel.getNumVertices(); ++vert) {
\n-
873
\n-
874 Vtr::Index cVert = refinement.getVertexChildVertex(vert);
\n-
875 if (!Vtr::IndexIsValid(cVert))
\n-
876 continue;
\n-
877
\n-
878 ConstIndexArray pVertValues = parentFVar.getVertexValues(vert),
\n-
879 cVertValues = childFVar.getVertexValues(cVert);
\n-
880
\n-
881 bool fvarVertVertMatchesVertex = childFVar.valueTopologyMatches(cVertValues[0]);
\n-
882 if (isLinearFVar && fvarVertVertMatchesVertex) {
\n-
883 dst[cVertValues[0]].Clear();
\n-
884 dst[cVertValues[0]].AddWithWeight(src[pVertValues[0]], 1.0f);
\n-
885 continue;
\n-
886 }
\n-
887
\n-
888 if (fvarVertVertMatchesVertex) {
\n-
889 //
\n-
890 // Declare and compute mask weights for this vertex relative to its parent edge:
\n-
891 //
\n-
892 // (We really need to encapsulate this somewhere else for use here and in the
\n-
893 // general case)
\n-
894 //
\n-
895 ConstIndexArray vEdges = parentLevel.getVertexEdges(vert);
\n-
896
\n-
897 Weight vVertWeight;
\n-
898 Weight * vEdgeWeights = weightBuffer;
\n-
899 Weight * vFaceWeights = vEdgeWeights + vEdges.size();
\n-
900
\n-
901 Mask vMask(&vVertWeight, vEdgeWeights, vFaceWeights);
\n-
902
\n-
903 vHood.SetIndex(vert, cVert);
\n-
904
\n-
905 Sdc::Crease::Rule pRule = parentLevel.getVertexRule(vert);
\n-
906 Sdc::Crease::Rule cRule = childLevel.getVertexRule(cVert);
\n-
907
\n-
908 scheme.ComputeVertexVertexMask(vHood, vMask, pRule, cRule);
\n-
909
\n-
910 // Apply the weights to the parent vertex, the vertices opposite its incident
\n-
911 // edges, and the child vertices of its incident faces:
\n-
912 //
\n-
913 // Even though the face-varying topology matches the vertex topology, we need
\n-
914 // to be careful here when getting values corresponding to vertices at the
\n-
915 // ends of edges. While the edge may be continuous, the end vertex may have
\n-
916 // discontinuities elsewhere in their neighborhood (i.e. on the "other side"
\n-
917 // of the end-vertex) and so have sibling values associated with them. In most
\n-
918 // cases the topology for an end-vertex will match and we can use it directly,
\n-
919 // but we must still check and retrieve as needed.
\n-
920 //
\n-
921 // Indices for values corresponding to face-vertices are guaranteed to match,
\n-
922 // so we can use the child-vertex indices directly.
\n-
923 //
\n-
924 // And by "directly", we always use getVertexValue(vertexIndex) to reference
\n-
925 // values in the "src" to account for the possible indirection that may exist at
\n-
926 // level 0 -- where there may be fewer values than vertices and an additional
\n-
927 // indirection is necessary. We can use a vertex index directly for "dst" when
\n-
928 // it matches.
\n-
929 //
\n-
930 // As with applying the mask to vertex data, in order to improve numerical
\n-
931 // precision, it's better to apply smaller weights first, so begin with the
\n-
932 // face-weights followed by the edge-weights and the vertex weight last.
\n-
933 //
\n-
934 Vtr::Index pVertValue = pVertValues[0];
\n-
935 Vtr::Index cVertValue = cVertValues[0];
\n-
936
\n-
937 dst[cVertValue].Clear();
\n-
938 if (vMask.GetNumFaceWeights() > 0) {
\n-
939 assert(vMask.AreFaceWeightsForFaceCenters());
\n-
940
\n-
941 ConstIndexArray vFaces = parentLevel.getVertexFaces(vert);
\n-
942
\n-
943 for (int i = 0; i < vFaces.size(); ++i) {
\n-
944
\n-
945 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(vFaces[i]);
\n-
946 assert(Vtr::IndexIsValid(cVertOfFace));
\n-
947
\n-
948 Vtr::Index cValueOfFace = childFVar.getVertexValueOffset(cVertOfFace);
\n-
949 dst[cVertValue].AddWithWeight(dst[cValueOfFace], vFaceWeights[i]);
\n-
950 }
\n-
951 }
\n-
952 if (vMask.GetNumEdgeWeights() > 0) {
\n-
953
\n-
954 parentFVar.getVertexEdgeValues(vert, vEdgeValues);
\n-
955
\n-
956 for (int i = 0; i < vEdges.size(); ++i) {
\n-
957 dst[cVertValue].AddWithWeight(src[vEdgeValues[i]], vEdgeWeights[i]);
\n-
958 }
\n-
959 }
\n-
960 dst[cVertValue].AddWithWeight(src[pVertValue], vVertWeight);
\n-
961 } else {
\n-
962 //
\n-
963 // Each FVar value associated with a vertex will be either a corner or a crease,
\n-
964 // or potentially in transition from corner to crease:
\n-
965 // - if the CHILD is a corner, there can be no transition so we have a corner
\n-
966 // - otherwise if the PARENT is a crease, both will be creases (no transition)
\n-
967 // - otherwise the parent must be a corner and the child a crease (transition)
\n-
968 //
\n-
969 Vtr::internal::FVarLevel::ConstValueTagArray pValueTags = parentFVar.getVertexValueTags(vert);
\n-
970 Vtr::internal::FVarLevel::ConstValueTagArray cValueTags = childFVar.getVertexValueTags(cVert);
\n-
971
\n-
972 for (int cSiblingIndex = 0; cSiblingIndex < cVertValues.size(); ++cSiblingIndex) {
\n-
973 int pSiblingIndex = refineFVar.getChildValueParentSource(cVert, cSiblingIndex);
\n-
974 assert(pSiblingIndex == cSiblingIndex);
\n-
975
\n-
976 typedef Vtr::internal::FVarLevel::Sibling SiblingIntType;
\n-
977
\n-
978 SiblingIntType cSibling = (SiblingIntType) cSiblingIndex;
\n-
979 SiblingIntType pSibling = (SiblingIntType) pSiblingIndex;
\n-
980
\n-
981 Vtr::Index pVertValue = pVertValues[pSibling];
\n-
982 Vtr::Index cVertValue = cVertValues[cSibling];
\n-
983
\n-
984 dst[cVertValue].Clear();
\n-
985 if (isLinearFVar || cValueTags[cSibling].isCorner()) {
\n-
986 dst[cVertValue].AddWithWeight(src[pVertValue], 1.0f);
\n-
987 } else {
\n-
988 //
\n-
989 // We have either a crease or a transition from corner to crease -- in
\n-
990 // either case, we need the end values for the full/fractional crease:
\n-
991 //
\n-
992 Index pEndValues[2];
\n-
993 parentFVar.getVertexCreaseEndValues(vert, pSibling, pEndValues);
\n-
994
\n-
995 Weight vWeight = 0.75f;
\n-
996 Weight eWeight = 0.125f;
\n-
997
\n-
998 //
\n-
999 // If semi-sharp we need to apply fractional weighting -- if made sharp because
\n-
1000 // of the other sibling (dependent-sharp) use the fractional weight from that
\n-
1001 // other sibling (should only occur when there are 2):
\n-
1002 //
\n-
1003 if (pValueTags[pSibling].isSemiSharp()) {
\n-
1004 Weight wCorner = pValueTags[pSibling].isDepSharp()
\n-
1005 ? refineFVar.getFractionalWeight(vert, !pSibling, cVert, !cSibling)
\n-
1006 : refineFVar.getFractionalWeight(vert, pSibling, cVert, cSibling);
\n-
1007 Weight wCrease = 1.0f - wCorner;
\n-
1008
\n-
1009 vWeight = wCrease * 0.75f + wCorner;
\n-
1010 eWeight = wCrease * 0.125f;
\n-
1011 }
\n-
1012 dst[cVertValue].AddWithWeight(src[pEndValues[0]], eWeight);
\n-
1013 dst[cVertValue].AddWithWeight(src[pEndValues[1]], eWeight);
\n-
1014 dst[cVertValue].AddWithWeight(src[pVertValue], vWeight);
\n-
1015 }
\n-
1016 }
\n-
1017 }
\n-
1018 }
\n-
1019}
\n-
1020
\n-
1021template <typename REAL>
\n-
1022template <Sdc::SchemeType SCHEME, class T, class U, class U1, class U2>
\n-
1023inline void
\n-
1024PrimvarRefinerReal<REAL>::limit(T const & src, U & dstPos, U1 * dstTan1Ptr, U2 * dstTan2Ptr) const {
\n-
1025
\n-
1026 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n-
1027
\n-
1028 Vtr::internal::Level const & level = _refiner.getLevel(_refiner.GetMaxLevel());
\n-
1029
\n-
1030 int maxWeightsPerMask = 1 + 2 * level.getMaxValence();
\n-
1031 bool hasTangents = (dstTan1Ptr && dstTan2Ptr);
\n-
1032 int numMasks = 1 + (hasTangents ? 2 : 0);
\n-
1033
\n-
1034 Vtr::internal::StackBuffer<Index,33> indexBuffer(maxWeightsPerMask);
\n-
1035 Vtr::internal::StackBuffer<Weight,99> weightBuffer(numMasks * maxWeightsPerMask);
\n-
1036
\n-
1037 Weight * vPosWeights = weightBuffer,
\n-
1038 * ePosWeights = vPosWeights + 1,
\n-
1039 * fPosWeights = ePosWeights + level.getMaxValence();
\n-
1040 Weight * vTan1Weights = vPosWeights + maxWeightsPerMask,
\n-
1041 * eTan1Weights = ePosWeights + maxWeightsPerMask,
\n-
1042 * fTan1Weights = fPosWeights + maxWeightsPerMask;
\n-
1043 Weight * vTan2Weights = vTan1Weights + maxWeightsPerMask,
\n-
1044 * eTan2Weights = eTan1Weights + maxWeightsPerMask,
\n-
1045 * fTan2Weights = fTan1Weights + maxWeightsPerMask;
\n-
1046
\n-
1047 Mask posMask( vPosWeights, ePosWeights, fPosWeights);
\n-
1048 Mask tan1Mask(vTan1Weights, eTan1Weights, fTan1Weights);
\n-
1049 Mask tan2Mask(vTan2Weights, eTan2Weights, fTan2Weights);
\n-
1050
\n-
1051 // This is a bit obscure -- assigning both parent and child as last level -- but
\n-
1052 // this mask type was intended for another purpose. Consider one for the limit:
\n-
1053 Vtr::internal::VertexInterface vHood(level, level);
\n-
1054
\n-
1055 for (int vert = 0; vert < level.getNumVertices(); ++vert) {
\n-
1056 ConstIndexArray vEdges = level.getVertexEdges(vert);
\n-
1057
\n-
1058 // Incomplete vertices (present in sparse refinement) do not have their full
\n-
1059 // topological neighborhood to determine a proper limit -- just leave the
\n-
1060 // vertex at the refined location and continue to the next:
\n-
1061 if (level.getVertexTag(vert)._incomplete || (vEdges.size() == 0)) {
\n-
1062 dstPos[vert].Clear();
\n-
1063 dstPos[vert].AddWithWeight(src[vert], 1.0);
\n-
1064 if (hasTangents) {
\n-
1065 (*dstTan1Ptr)[vert].Clear();
\n-
1066 (*dstTan2Ptr)[vert].Clear();
\n-
1067 }
\n-
1068 continue;
\n-
1069 }
\n-
1070
\n-
1071 //
\n-
1072 // Limit masks require the subdivision Rule for the vertex in order to deal
\n-
1073 // with infinitely sharp features correctly -- including boundaries and corners.
\n-
1074 // The vertex neighborhood is minimally defined with vertex and edge counts.
\n-
1075 //
\n-
1076 Sdc::Crease::Rule vRule = level.getVertexRule(vert);
\n-
1077
\n-
1078 // This is a bit obscure -- child vertex index will be ignored here
\n-
1079 vHood.SetIndex(vert, vert);
\n-
1080
\n-
1081 if (hasTangents) {
\n-
1082 scheme.ComputeVertexLimitMask(vHood, posMask, tan1Mask, tan2Mask, vRule);
\n-
1083 } else {
\n-
1084 scheme.ComputeVertexLimitMask(vHood, posMask, vRule);
\n-
1085 }
\n-
1086
\n-
1087 //
\n-
1088 // Gather the neighboring vertices of this vertex -- the vertices opposite its
\n-
1089 // incident edges, and the opposite vertices of its incident faces:
\n-
1090 //
\n-
1091 Index * eIndices = indexBuffer;
\n-
1092 Index * fIndices = indexBuffer + vEdges.size();
\n-
1093
\n-
1094 for (int i = 0; i < vEdges.size(); ++i) {
\n-
1095 ConstIndexArray eVerts = level.getEdgeVertices(vEdges[i]);
\n-
1096
\n-
1097 eIndices[i] = (eVerts[0] == vert) ? eVerts[1] : eVerts[0];
\n-
1098 }
\n-
1099 if (posMask.GetNumFaceWeights() || (hasTangents && tan1Mask.GetNumFaceWeights())) {
\n-
1100 ConstIndexArray vFaces = level.getVertexFaces(vert);
\n-
1101 ConstLocalIndexArray vInFace = level.getVertexFaceLocalIndices(vert);
\n-
1102
\n-
1103 for (int i = 0; i < vFaces.size(); ++i) {
\n-
1104 ConstIndexArray fVerts = level.getFaceVertices(vFaces[i]);
\n-
1105
\n-
1106 LocalIndex vOppInFace = (vInFace[i] + 2);
\n-
1107 if (vOppInFace >= fVerts.size()) vOppInFace -= (LocalIndex)fVerts.size();
\n-
1108
\n-
1109 fIndices[i] = level.getFaceVertices(vFaces[i])[vOppInFace];
\n-
1110 }
\n-
1111 }
\n-
1112
\n-
1113 //
\n-
1114 // Combine the weights and indices for position and tangents. As with applying
\n-
1115 // refinement masks to vertex data, in order to improve numerical precision, it's
\n-
1116 // better to apply smaller weights first, so begin with the face-weights followed
\n-
1117 // by the edge-weights and the vertex weight last.
\n-
1118 //
\n-
1119 dstPos[vert].Clear();
\n-
1120 for (int i = 0; i < posMask.GetNumFaceWeights(); ++i) {
\n-
1121 dstPos[vert].AddWithWeight(src[fIndices[i]], fPosWeights[i]);
\n-
1122 }
\n-
1123 for (int i = 0; i < posMask.GetNumEdgeWeights(); ++i) {
\n-
1124 dstPos[vert].AddWithWeight(src[eIndices[i]], ePosWeights[i]);
\n-
1125 }
\n-
1126 dstPos[vert].AddWithWeight(src[vert], vPosWeights[0]);
\n-
1127
\n-
1128 //
\n-
1129 // Apply the tangent masks -- both will have the same number of weights and
\n-
1130 // indices (one tangent may be "padded" to accommodate the other), but these
\n-
1131 // may differ from those of the position:
\n-
1132 //
\n-
1133 if (hasTangents) {
\n-
1134 assert(tan1Mask.GetNumFaceWeights() == tan2Mask.GetNumFaceWeights());
\n-
1135 assert(tan1Mask.GetNumEdgeWeights() == tan2Mask.GetNumEdgeWeights());
\n-
1136
\n-
1137 U1 & dstTan1 = *dstTan1Ptr;
\n-
1138 U2 & dstTan2 = *dstTan2Ptr;
\n-
1139
\n-
1140 dstTan1[vert].Clear();
\n-
1141 dstTan2[vert].Clear();
\n-
1142 for (int i = 0; i < tan1Mask.GetNumFaceWeights(); ++i) {
\n-
1143 dstTan1[vert].AddWithWeight(src[fIndices[i]], fTan1Weights[i]);
\n-
1144 dstTan2[vert].AddWithWeight(src[fIndices[i]], fTan2Weights[i]);
\n-
1145 }
\n-
1146 for (int i = 0; i < tan1Mask.GetNumEdgeWeights(); ++i) {
\n-
1147 dstTan1[vert].AddWithWeight(src[eIndices[i]], eTan1Weights[i]);
\n-
1148 dstTan2[vert].AddWithWeight(src[eIndices[i]], eTan2Weights[i]);
\n-
1149 }
\n-
1150 dstTan1[vert].AddWithWeight(src[vert], vTan1Weights[0]);
\n-
1151 dstTan2[vert].AddWithWeight(src[vert], vTan2Weights[0]);
\n-
1152 }
\n-
1153 }
\n-
1154}
\n-
1155
\n-
1156template <typename REAL>
\n-
1157template <Sdc::SchemeType SCHEME, class T, class U>
\n-
1158inline void
\n-
1159PrimvarRefinerReal<REAL>::limitFVar(T const & src, U & dst, int channel) const {
\n-
1160
\n-
1161 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n-
1162
\n-
1163 Vtr::internal::Level const & level = _refiner.getLevel(_refiner.GetMaxLevel());
\n-
1164 Vtr::internal::FVarLevel const & fvarChannel = level.getFVarLevel(channel);
\n-
1165
\n-
1166 int maxWeightsPerMask = 1 + 2 * level.getMaxValence();
\n-
1167
\n-
1168 Vtr::internal::StackBuffer<Weight,33> weightBuffer(maxWeightsPerMask);
\n-
1169 Vtr::internal::StackBuffer<Index,16> vEdgeBuffer(level.getMaxValence());
\n-
1170
\n-
1171 // This is a bit obscure -- assign both parent and child as last level
\n-
1172 Vtr::internal::VertexInterface vHood(level, level);
\n-
1173
\n-
1174 for (int vert = 0; vert < level.getNumVertices(); ++vert) {
\n-
1175
\n-
1176 ConstIndexArray vEdges = level.getVertexEdges(vert);
\n-
1177 ConstIndexArray vValues = fvarChannel.getVertexValues(vert);
\n-
1178
\n-
1179 // Incomplete vertices (present in sparse refinement) do not have their full
\n-
1180 // topological neighborhood to determine a proper limit -- just leave the
\n-
1181 // values (perhaps more than one per vertex) at the refined location.
\n-
1182 //
\n-
1183 // The same can be done if the face-varying channel is purely linear.
\n-
1184 //
\n-
1185 bool isIncomplete = (level.getVertexTag(vert)._incomplete || (vEdges.size() == 0));
\n-
1186 if (isIncomplete || fvarChannel.isLinear()) {
\n-
1187 for (int i = 0; i < vValues.size(); ++i) {
\n-
1188 Vtr::Index vValue = vValues[i];
\n-
1189
\n-
1190 dst[vValue].Clear();
\n-
1191 dst[vValue].AddWithWeight(src[vValue], 1.0f);
\n-
1192 }
\n-
1193 continue;
\n-
1194 }
\n-
1195
\n-
1196 bool fvarVertMatchesVertex = fvarChannel.valueTopologyMatches(vValues[0]);
\n-
1197 if (fvarVertMatchesVertex) {
\n-
1198
\n-
1199 // Assign the mask weights to the common buffer and compute the mask:
\n-
1200 //
\n-
1201 Weight * vWeights = weightBuffer,
\n-
1202 * eWeights = vWeights + 1,
\n-
1203 * fWeights = eWeights + vEdges.size();
\n-
1204
\n-
1205 Mask vMask(vWeights, eWeights, fWeights);
\n-
1206
\n-
1207 vHood.SetIndex(vert, vert);
\n-
1208
\n-
1209 scheme.ComputeVertexLimitMask(vHood, vMask, level.getVertexRule(vert));
\n-
1210
\n-
1211 //
\n-
1212 // Apply mask to corresponding FVar values for neighboring vertices:
\n-
1213 //
\n-
1214 Vtr::Index vValue = vValues[0];
\n-
1215
\n-
1216 dst[vValue].Clear();
\n-
1217 if (vMask.GetNumFaceWeights() > 0) {
\n-
1218 assert(!vMask.AreFaceWeightsForFaceCenters());
\n-
1219
\n-
1220 ConstIndexArray vFaces = level.getVertexFaces(vert);
\n-
1221 ConstLocalIndexArray vInFace = level.getVertexFaceLocalIndices(vert);
\n-
1222
\n-
1223 for (int i = 0; i < vFaces.size(); ++i) {
\n-
1224 ConstIndexArray faceValues = fvarChannel.getFaceValues(vFaces[i]);
\n-
1225 LocalIndex vOppInFace = vInFace[i] + 2;
\n-
1226 if (vOppInFace >= faceValues.size()) vOppInFace -= faceValues.size();
\n-
1227
\n-
1228 Index vValueOppositeFace = faceValues[vOppInFace];
\n-
1229
\n-
1230 dst[vValue].AddWithWeight(src[vValueOppositeFace], fWeights[i]);
\n-
1231 }
\n-
1232 }
\n-
1233 if (vMask.GetNumEdgeWeights() > 0) {
\n-
1234 Index * vEdgeValues = vEdgeBuffer;
\n-
1235 fvarChannel.getVertexEdgeValues(vert, vEdgeValues);
\n-
1236
\n-
1237 for (int i = 0; i < vEdges.size(); ++i) {
\n-
1238 dst[vValue].AddWithWeight(src[vEdgeValues[i]], eWeights[i]);
\n-
1239 }
\n-
1240 }
\n-
1241 dst[vValue].AddWithWeight(src[vValue], vWeights[0]);
\n-
1242 } else {
\n-
1243 //
\n-
1244 // Sibling FVar values associated with a vertex will be either a corner or a crease:
\n-
1245 //
\n-
1246 for (int i = 0; i < vValues.size(); ++i) {
\n-
1247 Vtr::Index vValue = vValues[i];
\n-
1248
\n-
1249 dst[vValue].Clear();
\n-
1250 if (fvarChannel.getValueTag(vValue).isCorner()) {
\n-
1251 dst[vValue].AddWithWeight(src[vValue], 1.0f);
\n-
1252 } else {
\n-
1253 Index vEndValues[2];
\n-
1254 fvarChannel.getVertexCreaseEndValues(vert, i, vEndValues);
\n-
1255
\n-
1256 dst[vValue].AddWithWeight(src[vEndValues[0]], 1.0f/6.0f);
\n-
1257 dst[vValue].AddWithWeight(src[vEndValues[1]], 1.0f/6.0f);
\n-
1258 dst[vValue].AddWithWeight(src[vValue], 2.0f/3.0f);
\n-
1259 }
\n-
1260 }
\n-
1261 }
\n-
1262 }
\n-
1263}
\n-
1264
\n-
1265class PrimvarRefiner : public PrimvarRefinerReal<float> {
\n-
1266public:
\n-
1267 PrimvarRefiner(TopologyRefiner const & refiner)
\n-
1268 : PrimvarRefinerReal<float>(refiner) { }
\n-
1269};
\n-
1270
\n-
1271} // end namespace Far
\n-
1272
\n-
1273} // end namespace OPENSUBDIV_VERSION
\n-
1274using namespace OPENSUBDIV_VERSION;
\n-
1275} // end namespace OpenSubdiv
\n-
1276
\n-
1277#endif /* OPENSUBDIV3_FAR_PRIMVAR_REFINER_H */
\n+
290
\n+
291private:
\n+
292 friend class TopologyRefiner;
\n+
293
\n+
294 Vtr::internal::Level const * _level;
\n+
295 Vtr::internal::Refinement const * _refToParent;
\n+
296 Vtr::internal::Refinement const * _refToChild;
\n+
297
\n+
298public:
\n+
299 // Not intended for public use, but required by std::vector, etc...
\n+
300 TopologyLevel() { }
\n+
301 ~TopologyLevel() { }
\n+
302};
\n+
303
\n+
304} // end namespace Far
\n+
305
\n+
306} // end namespace OPENSUBDIV_VERSION
\n+
307using namespace OPENSUBDIV_VERSION;
\n+
308} // end namespace OpenSubdiv
\n+
309
\n+
310#endif /* OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H */
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LocalIndex
Vtr::LocalIndex LocalIndex
Definition: types.h:42
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::Error
void Error(ErrorType err, const char *format,...)
Sends an OSD error with a message (internal use only)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstIndexArray
Vtr::ConstIndexArray ConstIndexArray
Definition: types.h:47
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstLocalIndexArray
Vtr::ConstLocalIndexArray ConstLocalIndexArray
Definition: types.h:48
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::FAR_RUNTIME_ERROR
@ FAR_RUNTIME_ERROR
Issue a generic runtime error, but continue execution.
Definition: error.h:40
\n \n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Vtr::IndexIsValid
bool IndexIsValid(Index index)
Definition: types.h:58
\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal
Applies refinement operations to generic primvar data.
Definition: primvarRefiner.h:56
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Limit
void Limit(T const &src, U &dstPos) const
Apply limit weights to a primvar buffer.
Definition: primvarRefiner.h:319
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Limit
void Limit(T const &src, U &dstPos, U1 &dstTan1, U2 &dstTan2) const
Definition: primvarRefiner.h:344
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateFaceUniform
void InterpolateFaceUniform(int level, T const &src, U &dst) const
Refine uniform (per-face) primvar data between levels.
Definition: primvarRefiner.h:394
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateVarying
void InterpolateVarying(int level, T const &src, U &dst) const
Apply only varying interpolation weights to a primvar buffer for a single level of refinement.
Definition: primvarRefiner.h:412
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Interpolate
void Interpolate(int level, T const &src, U &dst) const
Apply vertex interpolation weights to a primvar buffer for a single level of refinement.
Definition: primvarRefiner.h:267
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateFaceVarying
void InterpolateFaceVarying(int level, T const &src, U &dst, int channel=0) const
Apply face-varying interpolation weights to a primvar buffer associated with a particular face-varyin...
Definition: primvarRefiner.h:293
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::LimitFaceVarying
void LimitFaceVarying(T const &src, U &dst, int channel=0) const
Definition: primvarRefiner.h:369
\n-\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel
An interface for accessing data in a specific level of a refined topology hierarchy.
Definition: topologyLevel.h:49
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceParentFace
Index GetFaceParentFace(Index f) const
Return the parent face (in the previous level) of a given face.
Definition: topologyLevel.h:280
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexValenceRegular
bool IsVertexValenceRegular(Index v) const
Return if the valence of the vertex is regular (must be manifold)
Definition: topologyLevel.h:162
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeSharpness
float GetEdgeSharpness(Index e) const
Return the sharpness assigned a given edge.
Definition: topologyLevel.h:173
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesFaceFVarTopologyMatch
bool DoesFaceFVarTopologyMatch(Index f, int channel=0) const
Return if face-varying topology around a face matches.
Definition: topologyLevel.h:252
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexRule
Sdc::Crease::Rule GetVertexRule(Index v) const
Return the subdivision rule assigned a given vertex specific to this level.
Definition: topologyLevel.h:194
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexCorner
bool IsVertexCorner(Index v) const
Return if the vertex is a corner (only one incident face)
Definition: topologyLevel.h:140
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexChildVertex
Index GetVertexChildVertex(Index v) const
Return the child vertex (in the next level) of a given vertex.
Definition: topologyLevel.h:277
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeFaceLocalIndices
ConstLocalIndexArray GetEdgeFaceLocalIndices(Index e) const
Access the local indices of an edge with respect to its incident faces.
Definition: topologyLevel.h:117
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildEdges
ConstIndexArray GetFaceChildEdges(Index f) const
Access the child edges (in the next level) of a given face.
Definition: topologyLevel.h:265
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesEdgeFVarTopologyMatch
bool DoesEdgeFVarTopologyMatch(Index e, int channel=0) const
Return if face-varying topology across the edge only matches.
Definition: topologyLevel.h:247
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::PrintTopology
void PrintTopology(bool children=true) const
Definition: topologyLevel.h:287
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceVertices
ConstIndexArray GetFaceVertices(Index f) const
Access the vertices incident a given face.
Definition: topologyLevel.h:93
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildFaces
ConstIndexArray GetFaceChildFaces(Index f) const
Access the child faces (in the next level) of a given face.
Definition: topologyLevel.h:262
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumEdges
int GetNumEdges() const
Return the number of edges in this level.
Definition: topologyLevel.h:68
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeNonManifold
bool IsEdgeNonManifold(Index e) const
Return if the edge is non-manifold.
Definition: topologyLevel.h:128
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsFaceHole
bool IsFaceHole(Index f) const
Return if a given face has been tagged as a hole.
Definition: topologyLevel.h:191
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexEdges
ConstIndexArray GetVertexEdges(Index v) const
Access the edges incident a given vertex.
Definition: topologyLevel.h:108
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildVertex
Index GetFaceChildVertex(Index f) const
Return the child vertex (in the next level) of a given face.
Definition: topologyLevel.h:271
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesVertexFVarTopologyMatch
bool DoesVertexFVarTopologyMatch(Index v, int channel=0) const
Return if face-varying topology around a vertex matches.
Definition: topologyLevel.h:242
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeChildEdges
ConstIndexArray GetEdgeChildEdges(Index e) const
Access the child edges (in the next level) of a given edge.
Definition: topologyLevel.h:268
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeInfSharp
bool IsEdgeInfSharp(Index e) const
Return if the edge is infinitely-sharp.
Definition: topologyLevel.h:179
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexFaces
ConstIndexArray GetVertexFaces(Index v) const
Access the faces incident a given vertex.
Definition: topologyLevel.h:105
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::FindEdge
Index FindEdge(Index v0, Index v1) const
Identify the edge matching the given vertex pair.
Definition: topologyLevel.h:120
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexEdgeLocalIndices
ConstLocalIndexArray GetVertexEdgeLocalIndices(Index v) const
Access the local indices of a vertex with respect to its incident edges.
Definition: topologyLevel.h:114
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceEdges
ConstIndexArray GetFaceEdges(Index f) const
Access the edges incident a given face.
Definition: topologyLevel.h:96
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFaces
int GetNumFaces() const
Return the number of faces in this level.
Definition: topologyLevel.h:65
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexBoundary
bool IsVertexBoundary(Index v) const
Return if the vertex is on a boundary (at least one incident boundary edge)
Definition: topologyLevel.h:137
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFVarChannels
int GetNumFVarChannels() const
Return the number of face-varying channels (should be same for all levels)
Definition: topologyLevel.h:230
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumVertices
int GetNumVertices() const
Return the number of vertices in this level.
Definition: topologyLevel.h:62
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeFaces
ConstIndexArray GetEdgeFaces(Index e) const
Access the faces incident a given edge.
Definition: topologyLevel.h:102
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceFVarValues
ConstIndexArray GetFaceFVarValues(Index f, int channel=0) const
Access the face-varying values associated with a particular face.
Definition: topologyLevel.h:237
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeVertices
ConstIndexArray GetEdgeVertices(Index e) const
Access the vertices incident a given edge.
Definition: topologyLevel.h:99
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeSemiSharp
bool IsEdgeSemiSharp(Index e) const
Return if the edge is semi-sharp.
Definition: topologyLevel.h:185
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexSharpness
float GetVertexSharpness(Index v) const
Return the sharpness assigned a given vertex.
Definition: topologyLevel.h:176
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeChildVertex
Index GetEdgeChildVertex(Index e) const
Return the child vertex (in the next level) of a given edge.
Definition: topologyLevel.h:274
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexSemiSharp
bool IsVertexSemiSharp(Index v) const
Return if the vertex is semi-sharp.
Definition: topologyLevel.h:188
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFaceVertices
int GetNumFaceVertices() const
Return the total number of face-vertices, i.e. the sum of all vertices for all faces.
Definition: topologyLevel.h:71
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexNonManifold
bool IsVertexNonManifold(Index v) const
Return if the vertex is non-manifold.
Definition: topologyLevel.h:131
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeBoundary
bool IsEdgeBoundary(Index e) const
Return if the edge is a boundary (only one incident face)
Definition: topologyLevel.h:134
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexInfSharp
bool IsVertexInfSharp(Index v) const
Return if the vertex is infinitely-sharp.
Definition: topologyLevel.h:182
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFVarValues
int GetNumFVarValues(int channel=0) const
Return the total number of face-varying values in a particular channel (the upper bound of a face-var...
Definition: topologyLevel.h:234
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexFaceLocalIndices
ConstLocalIndexArray GetVertexFaceLocalIndices(Index v) const
Access the local indices of a vertex with respect to its incident faces.
Definition: topologyLevel.h:111
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner
Stores topology data for a specified set of refinement options.
Definition: topologyRefiner.h:51
\n \n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme
Scheme is a class template which provides all implementation for the subdivision schemes supported by...
Definition: scheme.h:66
\n \n-\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-primvarRefiner.h\n+topologyLevel.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 DreamWorks Animation LLC.\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -29,1398 +29,387 @@\n 17//\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n- 24#ifndef OPENSUBDIV3_FAR_PRIMVAR_REFINER_H\n- 25#define OPENSUBDIV3_FAR_PRIMVAR_REFINER_H\n+ 24#ifndef OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H\n+ 25#define OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H\n 26\n 27#include \"../version.h\"\n 28\n- 29#include \"../sdc/types.h\"\n- 30#include \"../sdc/options.h\"\n- 31#include \"../sdc/bilinearScheme.h\"\n- 32#include \"../sdc/catmarkScheme.h\"\n- 33#include \"../sdc/loopScheme.h\"\n- 34#include \"../vtr/level.h\"\n- 35#include \"../vtr/fvarLevel.h\"\n- 36#include \"../vtr/refinement.h\"\n- 37#include \"../vtr/fvarRefinement.h\"\n- 38#include \"../vtr/stackBuffer.h\"\n- 39#include \"../vtr/componentInterfaces.h\"\n- 40#include \"../far/types.h\"\n- 41#include \"../far/error.h\"\n- 42#include \"../far/topologyLevel.h\"\n- 43#include \"../far/topologyRefiner.h\"\n- 44\n- 45#include \n- 46\n- 47namespace OpenSubdiv {\n- 48namespace OPENSUBDIV_VERSION {\n- 49\n- 50namespace Far {\n- 51\n- 55template \n-56class PrimvarRefinerReal {\n- 57\n- 58public:\n-59 PrimvarRefinerReal(TopologyRefiner const & refiner) : _refiner(refiner) { }\n-60 ~PrimvarRefinerReal() { }\n- 61\n-62 TopologyRefiner const & GetTopologyRefiner() const { return _refiner; }\n+ 29#include \"../vtr/level.h\"\n+ 30#include \"../vtr/refinement.h\"\n+ 31#include \"../far/types.h\"\n+ 32\n+ 33#include \n+ 34\n+ 35namespace OpenSubdiv {\n+ 36namespace OPENSUBDIV_VERSION {\n+ 37\n+ 38namespace Far {\n+ 39\n+49class TopologyLevel {\n+ 50\n+ 51public:\n+ 53\n+ 60\n+62 int GetNumVertices() const { return _level->getNumVertices(); }\n 63\n- 65\n+65 int GetNumFaces() const { return _level->getNumFaces(); }\n+ 66\n+68 int GetNumEdges() const { return _level->getNumEdges(); }\n+ 69\n+71 int GetNumFaceVertices() const { return _level->getNumFaceVerticesTotal(); }\n+ 73\n+ 75\n+ 90 //\n+ 91\n+93 ConstIndexArray GetFaceVertices(Index f) const { return _level-\n+>getFaceVertices(f); }\n+ 94\n+96 ConstIndexArray GetFaceEdges(Index f) const { return _level->getFaceEdges\n+(f); }\n+ 97\n+99 ConstIndexArray GetEdgeVertices(Index e) const { return _level-\n+>getEdgeVertices(e); }\n 100\n-113 template void Interpolate(int level, T const & src, U &\n-dst) const;\n- 114\n-130 template void InterpolateVarying(int level, T const &\n-src, U & dst) const;\n- 131\n-148 template void InterpolateFaceUniform(int level, T const\n-& src, U & dst) const;\n- 149\n-161 template void InterpolateFaceVarying(int level, T const\n-& src, U & dst, int channel = 0) const;\n- 162\n- 163\n-175 template void Limit(T const & src, U & dstPos) const;\n- 176\n- 177 template \n-178 void Limit(T const & src, U & dstPos, U1 & dstTan1, U2 & dstTan2) const;\n- 179\n-180 template void LimitFaceVarying(T const & src, U & dst,\n-int channel = 0) const;\n- 181\n+102 ConstIndexArray GetEdgeFaces(Index e) const { return _level->getEdgeFaces\n+(e); }\n+ 103\n+105 ConstIndexArray GetVertexFaces(Index v) const { return _level-\n+>getVertexFaces(v); }\n+ 106\n+108 ConstIndexArray GetVertexEdges(Index v) const { return _level-\n+>getVertexEdges(v); }\n+ 109\n+111 ConstLocalIndexArray GetVertexFaceLocalIndices(Index v) const { return\n+_level->getVertexFaceLocalIndices(v); }\n+ 112\n+114 ConstLocalIndexArray GetVertexEdgeLocalIndices(Index v) const { return\n+_level->getVertexEdgeLocalIndices(v); }\n+ 115\n+117 ConstLocalIndexArray GetEdgeFaceLocalIndices(Index e) const { return\n+_level->getEdgeFaceLocalIndices(e); }\n+ 118\n+120 Index FindEdge(Index v0, Index v1) const { return _level->findEdge(v0, v1);\n+}\n+ 122\n+ 124\n+ 126\n+128 bool IsEdgeNonManifold(Index e) const { return _level->isEdgeNonManifold\n+(e); }\n+ 129\n+131 bool IsVertexNonManifold(Index v) const { return _level-\n+>isVertexNonManifold(v); }\n+ 132\n+134 bool IsEdgeBoundary(Index e) const { return _level->getEdgeTag\n+(e)._boundary; }\n+ 135\n+137 bool IsVertexBoundary(Index v) const { return _level->getVertexTag\n+(v)._boundary; }\n+ 138\n+140 bool IsVertexCorner(Index v) const { return (_level->getNumVertexFaces(v)\n+== 1); }\n+ 141\n+162 bool IsVertexValenceRegular(Index v) const { return !_level->getVertexTag\n+(v)._xordinary || IsVertexCorner(v); }\n+ 164\n+ 166\n+ 171\n+173 float GetEdgeSharpness(Index e) const { return _level->getEdgeSharpness(e);\n+}\n+ 174\n+176 float GetVertexSharpness(Index v) const { return _level->getVertexSharpness\n+(v); }\n+ 177\n+179 bool IsEdgeInfSharp(Index e) const { return _level->getEdgeTag\n+(e)._infSharp; }\n+ 180\n+182 bool IsVertexInfSharp(Index v) const { return _level->getVertexTag\n+(v)._infSharp; }\n 183\n- 184private:\n- 185 typedef REAL Weight;\n+185 bool IsEdgeSemiSharp(Index e) const { return _level->getEdgeTag\n+(e)._semiSharp; }\n 186\n- 187 // Non-copyable:\n- 188 PrimvarRefinerReal(PrimvarRefinerReal const & src) : _refiner\n-(src._refiner) { }\n- 189 PrimvarRefinerReal & operator=(PrimvarRefinerReal const &) { return *this;\n-}\n- 190\n- 191 template void interpFromFaces\n-(int, T const &, U &) const;\n- 192 template void interpFromEdges\n-(int, T const &, U &) const;\n- 193 template void interpFromVerts\n-(int, T const &, U &) const;\n- 194\n- 195 template void\n-interpFVarFromFaces(int, T const &, U &, int) const;\n- 196 template void\n-interpFVarFromEdges(int, T const &, U &, int) const;\n- 197 template void\n-interpFVarFromVerts(int, T const &, U &, int) const;\n+188 bool IsVertexSemiSharp(Index v) const { return _level->getVertexTag\n+(v)._semiSharp; }\n+ 189\n+191 bool IsFaceHole(Index f) const { return _level->isFaceHole(f); }\n+ 192\n+194 Sdc::Crease::Rule GetVertexRule(Index v) const { return _level-\n+>getVertexRule(v); }\n+ 196\n 198\n- 199 template \n- 200 void limit(T const & src, U & pos, U1 * tan1, U2 * tan2) const;\n- 201\n- 202 template \n- 203 void limitFVar(T const & src, U & dst, int channel) const;\n- 204\n- 205private:\n- 206 TopologyRefiner const & _refiner;\n- 207\n- 208private:\n- 209 //\n- 210 // Local class to fulfill interface for in the Scheme mask\n-queries:\n- 211 //\n- 212 class Mask {\n- 213 public:\n- 214 typedef REAL Weight; // Also part of the expected interface\n- 215\n- 216 public:\n- 217 Mask(Weight* v, Weight* e, Weight* f) :\n- 218 _vertWeights(v), _edgeWeights(e), _faceWeights(f),\n- 219 _vertCount(0), _edgeCount(0), _faceCount(0),\n- 220 _faceWeightsForFaceCenters(false)\n- 221 { }\n- 222\n- 223 ~Mask() { }\n- 224\n- 225 public: // Generic interface expected of :\n- 226 int GetNumVertexWeights() const { return _vertCount; }\n- 227 int GetNumEdgeWeights() const { return _edgeCount; }\n- 228 int GetNumFaceWeights() const { return _faceCount; }\n- 229\n- 230 void SetNumVertexWeights(int count) { _vertCount = count; }\n- 231 void SetNumEdgeWeights( int count) { _edgeCount = count; }\n- 232 void SetNumFaceWeights( int count) { _faceCount = count; }\n- 233\n- 234 Weight const& VertexWeight(int index) const { return _vertWeights[index];\n-}\n- 235 Weight const& EdgeWeight( int index) const { return _edgeWeights[index]; }\n- 236 Weight const& FaceWeight( int index) const { return _faceWeights[index]; }\n- 237\n- 238 Weight& VertexWeight(int index) { return _vertWeights[index]; }\n- 239 Weight& EdgeWeight( int index) { return _edgeWeights[index]; }\n- 240 Weight& FaceWeight( int index) { return _faceWeights[index]; }\n- 241\n- 242 bool AreFaceWeightsForFaceCenters() const { return\n-_faceWeightsForFaceCenters; }\n- 243 void SetFaceWeightsForFaceCenters(bool on) { _faceWeightsForFaceCenters =\n-on; }\n- 244\n- 245 private:\n- 246 Weight* _vertWeights;\n- 247 Weight* _edgeWeights;\n- 248 Weight* _faceWeights;\n- 249\n- 250 int _vertCount;\n- 251 int _edgeCount;\n- 252 int _faceCount;\n- 253\n- 254 bool _faceWeightsForFaceCenters;\n- 255 };\n- 256};\n+ 228\n+230 int GetNumFVarChannels() const { return _level->getNumFVarChannels(); }\n+ 231\n+234 int GetNumFVarValues(int channel = 0) const { return _level-\n+>getNumFVarValues(channel); }\n+ 235\n+237 ConstIndexArray GetFaceFVarValues(Index f, int channel = 0) const {\n+ 238 return _level->getFaceFVarValues(f, channel);\n+ 239 }\n+ 240\n+242 bool DoesVertexFVarTopologyMatch(Index v, int channel = 0) const {\n+ 243 return _level->doesVertexFVarTopologyMatch(v, channel);\n+ 244 }\n+ 245\n+247 bool DoesEdgeFVarTopologyMatch(Index e, int channel = 0) const {\n+ 248 return _level->doesEdgeFVarTopologyMatch(e, channel);\n+ 249 }\n+ 250\n+252 bool DoesFaceFVarTopologyMatch(Index f, int channel = 0) const {\n+ 253 return _level->doesFaceFVarTopologyMatch(f, channel);\n+ 254 }\n+ 255\n 257\n- 258\n- 259//\n- 260// Public entry points to the methods. Queries of the scheme type and its\n- 261// use as a template parameter in subsequent implementation will be\n-factored\n- 262// out of a later release:\n- 263//\n- 264template \n- 265template \n- 266inline void\n-267PrimvarRefinerReal::Interpolate(int level, T const & src, U & dst)\n-const {\n- 268\n- 269 assert(level>0 && level<=(int)_refiner._refinements.size());\n- 270\n- 271 switch (_refiner._subdivType) {\n- 272 case Sdc::SCHEME_CATMARK:\n- 273 interpFromFaces(level, src, dst);\n- 274 interpFromEdges(level, src, dst);\n- 275 interpFromVerts(level, src, dst);\n- 276 break;\n- 277 case Sdc::SCHEME_LOOP:\n- 278 interpFromFaces(level, src, dst);\n- 279 interpFromEdges(level, src, dst);\n- 280 interpFromVerts(level, src, dst);\n- 281 break;\n- 282 case Sdc::SCHEME_BILINEAR:\n- 283 interpFromFaces(level, src, dst);\n- 284 interpFromEdges(level, src, dst);\n- 285 interpFromVerts(level, src, dst);\n- 286 break;\n- 287 }\n- 288}\n+ 259\n+ 260\n+262 ConstIndexArray GetFaceChildFaces(Index f) const { return _refToChild-\n+>getFaceChildFaces(f); }\n+ 263\n+265 ConstIndexArray GetFaceChildEdges(Index f) const { return _refToChild-\n+>getFaceChildEdges(f); }\n+ 266\n+268 ConstIndexArray GetEdgeChildEdges(Index e) const { return _refToChild-\n+>getEdgeChildEdges(e); }\n+ 269\n+271 Index GetFaceChildVertex( Index f) const { return _refToChild-\n+>getFaceChildVertex(f); }\n+ 272\n+274 Index GetEdgeChildVertex( Index e) const { return _refToChild-\n+>getEdgeChildVertex(e); }\n+ 275\n+277 Index GetVertexChildVertex(Index v) const { return _refToChild-\n+>getVertexChildVertex(v); }\n+ 278\n+280 Index GetFaceParentFace(Index f) const { return _refToParent-\n+>getChildFaceParentFace(f); }\n+ 282\n+ 284\n+ 285\n+286 bool ValidateTopology() const { return _level->validateTopology(); }\n+287 void PrintTopology(bool children = true) const { _level->print((children &&\n+_refToChild) ? _refToChild : 0); }\n 289\n- 290template \n- 291template \n- 292inline void\n-293PrimvarRefinerReal::InterpolateFaceVarying(int level, T const & src, U\n-& dst, int channel) const {\n- 294\n- 295 assert(level>0 && level<=(int)_refiner._refinements.size());\n- 296\n- 297 switch (_refiner._subdivType) {\n- 298 case Sdc::SCHEME_CATMARK:\n- 299 interpFVarFromFaces(level, src, dst, channel);\n- 300 interpFVarFromEdges(level, src, dst, channel);\n- 301 interpFVarFromVerts(level, src, dst, channel);\n- 302 break;\n- 303 case Sdc::SCHEME_LOOP:\n- 304 interpFVarFromFaces(level, src, dst, channel);\n- 305 interpFVarFromEdges(level, src, dst, channel);\n- 306 interpFVarFromVerts(level, src, dst, channel);\n- 307 break;\n- 308 case Sdc::SCHEME_BILINEAR:\n- 309 interpFVarFromFaces(level, src, dst, channel);\n- 310 interpFVarFromEdges(level, src, dst, channel);\n- 311 interpFVarFromVerts(level, src, dst, channel);\n- 312 break;\n- 313 }\n- 314}\n- 315\n- 316template \n- 317template \n- 318inline void\n-319PrimvarRefinerReal::Limit(T const & src, U & dst) const {\n- 320\n- 321 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() ==\n-0) {\n- 322 Error(FAR_RUNTIME_ERROR,\n- 323 \"Failure in PrimvarRefiner::Limit() -- \"\n- 324 \"last level of refinement does not include full topology.\");\n- 325 return;\n- 326 }\n- 327\n- 328 switch (_refiner._subdivType) {\n- 329 case Sdc::SCHEME_CATMARK:\n- 330 limit(src, dst, (U*)0, (U*)0);\n- 331 break;\n- 332 case Sdc::SCHEME_LOOP:\n- 333 limit(src, dst, (U*)0, (U*)0);\n- 334 break;\n- 335 case Sdc::SCHEME_BILINEAR:\n- 336 limit(src, dst, (U*)0, (U*)0);\n- 337 break;\n- 338 }\n- 339}\n- 340\n- 341template \n- 342template \n- 343inline void\n-344PrimvarRefinerReal::Limit(T const & src, U & dstPos, U1 & dstTan1, U2\n-& dstTan2) const {\n- 345\n- 346 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() ==\n-0) {\n- 347 Error(FAR_RUNTIME_ERROR,\n- 348 \"Failure in PrimvarRefiner::Limit() -- \"\n- 349 \"last level of refinement does not include full topology.\");\n- 350 return;\n- 351 }\n- 352\n- 353 switch (_refiner._subdivType) {\n- 354 case Sdc::SCHEME_CATMARK:\n- 355 limit(src, dstPos, &dstTan1, &dstTan2);\n- 356 break;\n- 357 case Sdc::SCHEME_LOOP:\n- 358 limit(src, dstPos, &dstTan1, &dstTan2);\n- 359 break;\n- 360 case Sdc::SCHEME_BILINEAR:\n- 361 limit(src, dstPos, &dstTan1, &dstTan2);\n- 362 break;\n- 363 }\n- 364}\n- 365\n- 366template \n- 367template \n- 368inline void\n-369PrimvarRefinerReal::LimitFaceVarying(T const & src, U & dst, int\n-channel) const {\n- 370\n- 371 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() ==\n-0) {\n- 372 Error(FAR_RUNTIME_ERROR,\n- 373 \"Failure in PrimvarRefiner::LimitFaceVarying() -- \"\n- 374 \"last level of refinement does not include full topology.\");\n- 375 return;\n- 376 }\n- 377\n- 378 switch (_refiner._subdivType) {\n- 379 case Sdc::SCHEME_CATMARK:\n- 380 limitFVar(src, dst, channel);\n- 381 break;\n- 382 case Sdc::SCHEME_LOOP:\n- 383 limitFVar(src, dst, channel);\n- 384 break;\n- 385 case Sdc::SCHEME_BILINEAR:\n- 386 limitFVar(src, dst, channel);\n- 387 break;\n- 388 }\n- 389}\n- 390\n- 391template \n- 392template \n- 393inline void\n-394PrimvarRefinerReal::InterpolateFaceUniform(int level, T const & src, U\n-& dst) const {\n- 395\n- 396 assert(level>0 && level<=(int)_refiner._refinements.size());\n- 397\n- 398 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n-(level-1);\n- 399 Vtr::internal::Level const & child = refinement.child();\n- 400\n- 401 for (int cFace = 0; cFace < child.getNumFaces(); ++cFace) {\n- 402\n- 403 Vtr::Index pFace = refinement.getChildFaceParentFace(cFace);\n- 404\n- 405 dst[cFace] = src[pFace];\n- 406 }\n- 407}\n- 408\n- 409template \n- 410template \n- 411inline void\n-412PrimvarRefinerReal::InterpolateVarying(int level, T const & src, U &\n-dst) const {\n- 413\n- 414 assert(level>0 && level<=(int)_refiner._refinements.size());\n- 415\n- 416 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n-(level-1);\n- 417 Vtr::internal::Level const & parent = refinement.parent();\n- 418\n- 419 //\n- 420 // Group values to interpolate based on origin -- note that there may\n- 421 // be none originating from faces:\n- 422 //\n- 423 if (refinement.getNumChildVerticesFromFaces() > 0) {\n- 424\n- 425 for (int face = 0; face < parent.getNumFaces(); ++face) {\n- 426\n- 427 Vtr::Index cVert = refinement.getFaceChildVertex(face);\n- 428 if (Vtr::IndexIsValid(cVert)) {\n- 429\n- 430 // Apply the weights to the parent face's vertices:\n- 431 ConstIndexArray fVerts = parent.getFaceVertices(face);\n- 432\n- 433 Weight fVaryingWeight = 1.0f / (Weight) fVerts.size();\n- 434\n- 435 dst[cVert].Clear();\n- 436 for (int i = 0; i < fVerts.size(); ++i) {\n- 437 dst[cVert].AddWithWeight(src[fVerts[i]], fVaryingWeight);\n- 438 }\n- 439 }\n- 440 }\n- 441 }\n- 442 for (int edge = 0; edge < parent.getNumEdges(); ++edge) {\n- 443\n- 444 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);\n- 445 if (Vtr::IndexIsValid(cVert)) {\n- 446\n- 447 // Apply the weights to the parent edges's vertices\n- 448 ConstIndexArray eVerts = parent.getEdgeVertices(edge);\n- 449\n- 450 dst[cVert].Clear();\n- 451 dst[cVert].AddWithWeight(src[eVerts[0]], 0.5f);\n- 452 dst[cVert].AddWithWeight(src[eVerts[1]], 0.5f);\n- 453 }\n- 454 }\n- 455 for (int vert = 0; vert < parent.getNumVertices(); ++vert) {\n- 456\n- 457 Vtr::Index cVert = refinement.getVertexChildVertex(vert);\n- 458 if (Vtr::IndexIsValid(cVert)) {\n- 459\n- 460 // Essentially copy the parent vertex:\n- 461 dst[cVert].Clear();\n- 462 dst[cVert].AddWithWeight(src[vert], 1.0f);\n- 463 }\n- 464 }\n- 465}\n- 466\n- 467\n- 468//\n- 469// Internal implementation methods -- grouping vertices to be interpolated\n- 470// based on the type of parent component from which they originated:\n- 471//\n- 472template \n- 473template \n- 474inline void\n- 475PrimvarRefinerReal::interpFromFaces(int level, T const & src, U &\n-dst) const {\n- 476\n- 477 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n-(level-1);\n- 478 Vtr::internal::Level const & parent = refinement.parent();\n- 479\n- 480 if (refinement.getNumChildVerticesFromFaces() == 0) return;\n- 481\n- 482 Sdc::Scheme scheme(_refiner._subdivOptions);\n- 483\n- 484 Vtr::internal::StackBuffer fVertWeights(parent.getMaxValence\n-());\n- 485\n- 486 for (int face = 0; face < parent.getNumFaces(); ++face) {\n- 487\n- 488 Vtr::Index cVert = refinement.getFaceChildVertex(face);\n- 489 if (!Vtr::IndexIsValid(cVert))\n- 490 continue;\n- 491\n- 492 // Declare and compute mask weights for this vertex relative to its parent\n-face:\n- 493 ConstIndexArray fVerts = parent.getFaceVertices(face);\n- 494\n- 495 Mask fMask(fVertWeights, 0, 0);\n- 496 Vtr::internal::FaceInterface fHood(fVerts.size());\n- 497\n- 498 scheme.ComputeFaceVertexMask(fHood, fMask);\n- 499\n- 500 // Apply the weights to the parent face's vertices:\n- 501 dst[cVert].Clear();\n- 502\n- 503 for (int i = 0; i < fVerts.size(); ++i) {\n- 504\n- 505 dst[cVert].AddWithWeight(src[fVerts[i]], fVertWeights[i]);\n- 506 }\n- 507 }\n- 508}\n- 509\n- 510template \n- 511template \n- 512inline void\n- 513PrimvarRefinerReal::interpFromEdges(int level, T const & src, U &\n-dst) const {\n- 514\n- 515 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n-(level-1);\n- 516 Vtr::internal::Level const & parent = refinement.parent();\n- 517 Vtr::internal::Level const & child = refinement.child();\n- 518\n- 519 Sdc::Scheme scheme(_refiner._subdivOptions);\n- 520\n- 521 Vtr::internal::EdgeInterface eHood(parent);\n- 522\n- 523 Weight eVertWeights[2];\n- 524 Vtr::internal::StackBuffer eFaceWeights(parent.getMaxEdgeFaces\n-());\n- 525\n- 526 for (int edge = 0; edge < parent.getNumEdges(); ++edge) {\n- 527\n- 528 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);\n- 529 if (!Vtr::IndexIsValid(cVert))\n- 530 continue;\n- 531\n- 532 // Declare and compute mask weights for this vertex relative to its parent\n-edge:\n- 533 ConstIndexArray eVerts = parent.getEdgeVertices(edge),\n- 534 eFaces = parent.getEdgeFaces(edge);\n- 535\n- 536 Mask eMask(eVertWeights, 0, eFaceWeights);\n- 537\n- 538 eHood.SetIndex(edge);\n- 539\n- 540 Sdc::Crease::Rule pRule = (parent.getEdgeSharpness(edge) > 0.0f) ? Sdc::\n-Crease::RULE_CREASE : Sdc::Crease::RULE_SMOOTH;\n- 541 Sdc::Crease::Rule cRule = child.getVertexRule(cVert);\n- 542\n- 543 scheme.ComputeEdgeVertexMask(eHood, eMask, pRule, cRule);\n- 544\n- 545 // Apply the weights to the parent edges's vertices and (if applicable) to\n- 546 // the child vertices of its incident faces:\n- 547 dst[cVert].Clear();\n- 548 dst[cVert].AddWithWeight(src[eVerts[0]], eVertWeights[0]);\n- 549 dst[cVert].AddWithWeight(src[eVerts[1]], eVertWeights[1]);\n- 550\n- 551 if (eMask.GetNumFaceWeights() > 0) {\n- 552\n- 553 for (int i = 0; i < eFaces.size(); ++i) {\n- 554\n- 555 if (eMask.AreFaceWeightsForFaceCenters()) {\n- 556 assert(refinement.getNumChildVerticesFromFaces() > 0);\n- 557 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(eFaces[i]);\n- 558\n- 559 assert(Vtr::IndexIsValid(cVertOfFace));\n- 560 dst[cVert].AddWithWeight(dst[cVertOfFace], eFaceWeights[i]);\n- 561 } else {\n- 562 Vtr::Index pFace = eFaces[i];\n- 563 ConstIndexArray pFaceEdges = parent.getFaceEdges(pFace),\n- 564 pFaceVerts = parent.getFaceVertices(pFace);\n- 565\n- 566 int eInFace = 0;\n- 567 for ( ; pFaceEdges[eInFace] != edge; ++eInFace ) ;\n- 568\n- 569 int vInFace = eInFace + 2;\n- 570 if (vInFace >= pFaceVerts.size()) vInFace -= pFaceVerts.size();\n- 571\n- 572 Vtr::Index pVertNext = pFaceVerts[vInFace];\n- 573 dst[cVert].AddWithWeight(src[pVertNext], eFaceWeights[i]);\n- 574 }\n- 575 }\n- 576 }\n- 577 }\n- 578}\n- 579\n- 580template \n- 581template \n- 582inline void\n- 583PrimvarRefinerReal::interpFromVerts(int level, T const & src, U &\n-dst) const {\n- 584\n- 585 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n-(level-1);\n- 586 Vtr::internal::Level const & parent = refinement.parent();\n- 587 Vtr::internal::Level const & child = refinement.child();\n- 588\n- 589 Sdc::Scheme scheme(_refiner._subdivOptions);\n- 590\n- 591 Vtr::internal::VertexInterface vHood(parent, child);\n- 592\n- 593 Vtr::internal::StackBuffer weightBuffer(2*parent.getMaxValence\n-());\n- 594\n- 595 for (int vert = 0; vert < parent.getNumVertices(); ++vert) {\n- 596\n- 597 Vtr::Index cVert = refinement.getVertexChildVertex(vert);\n- 598 if (!Vtr::IndexIsValid(cVert))\n- 599 continue;\n- 600\n- 601 // Declare and compute mask weights for this vertex relative to its parent\n-edge:\n- 602 ConstIndexArray vEdges = parent.getVertexEdges(vert),\n- 603 vFaces = parent.getVertexFaces(vert);\n- 604\n- 605 Weight vVertWeight,\n- 606 * vEdgeWeights = weightBuffer,\n- 607 * vFaceWeights = vEdgeWeights + vEdges.size();\n- 608\n- 609 Mask vMask(&vVertWeight, vEdgeWeights, vFaceWeights);\n- 610\n- 611 vHood.SetIndex(vert, cVert);\n- 612\n- 613 Sdc::Crease::Rule pRule = parent.getVertexRule(vert);\n- 614 Sdc::Crease::Rule cRule = child.getVertexRule(cVert);\n- 615\n- 616 scheme.ComputeVertexVertexMask(vHood, vMask, pRule, cRule);\n- 617\n- 618 // Apply the weights to the parent vertex, the vertices opposite its\n-incident\n- 619 // edges, and the child vertices of its incident faces:\n- 620 //\n- 621 // In order to improve numerical precision, it's better to apply smaller\n-weights\n- 622 // first, so begin with the face-weights followed by the edge-weights and\n-the\n- 623 // vertex weight last.\n- 624 dst[cVert].Clear();\n- 625\n- 626 if (vMask.GetNumFaceWeights() > 0) {\n- 627 assert(vMask.AreFaceWeightsForFaceCenters());\n- 628\n- 629 for (int i = 0; i < vFaces.size(); ++i) {\n- 630\n- 631 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(vFaces[i]);\n- 632 assert(Vtr::IndexIsValid(cVertOfFace));\n- 633 dst[cVert].AddWithWeight(dst[cVertOfFace], vFaceWeights[i]);\n- 634 }\n- 635 }\n- 636 if (vMask.GetNumEdgeWeights() > 0) {\n- 637\n- 638 for (int i = 0; i < vEdges.size(); ++i) {\n- 639\n- 640 ConstIndexArray eVerts = parent.getEdgeVertices(vEdges[i]);\n- 641 Vtr::Index pVertOppositeEdge = (eVerts[0] == vert) ? eVerts[1] : eVerts\n-[0];\n- 642\n- 643 dst[cVert].AddWithWeight(src[pVertOppositeEdge], vEdgeWeights[i]);\n- 644 }\n- 645 }\n- 646 dst[cVert].AddWithWeight(src[vert], vVertWeight);\n- 647 }\n- 648}\n- 649\n- 650\n- 651//\n- 652// Internal face-varying implementation details:\n- 653//\n- 654template \n- 655template \n- 656inline void\n- 657PrimvarRefinerReal::interpFVarFromFaces(int level, T const & src, U &\n-dst, int channel) const {\n- 658\n- 659 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n-(level-1);\n- 660\n- 661 if (refinement.getNumChildVerticesFromFaces() == 0) return;\n- 662\n- 663 Sdc::Scheme scheme(_refiner._subdivOptions);\n- 664\n- 665 Vtr::internal::Level const & parentLevel = refinement.parent();\n- 666 Vtr::internal::Level const & childLevel = refinement.child();\n- 667\n- 668 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel\n-(channel);\n- 669 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel\n-(channel);\n- 670\n- 671 Vtr::internal::StackBuffer fValueWeights\n-(parentLevel.getMaxValence());\n- 672\n- 673 for (int face = 0; face < parentLevel.getNumFaces(); ++face) {\n- 674\n- 675 Vtr::Index cVert = refinement.getFaceChildVertex(face);\n- 676 if (!Vtr::IndexIsValid(cVert))\n- 677 continue;\n- 678\n- 679 Vtr::Index cVertValue = childFVar.getVertexValueOffset(cVert);\n- 680\n- 681 // The only difference for face-varying here is that we get the values\n-associated\n- 682 // with each face-vertex directly from the FVarLevel, rather than using\n-the parent\n- 683 // face-vertices directly. If any face-vertex has any sibling values, then\n-we may\n- 684 // get the wrong one using the face-vertex index directly.\n- 685\n- 686 // Declare and compute mask weights for this vertex relative to its parent\n-face:\n- 687 ConstIndexArray fValues = parentFVar.getFaceValues(face);\n- 688\n- 689 Mask fMask(fValueWeights, 0, 0);\n- 690 Vtr::internal::FaceInterface fHood(fValues.size());\n- 691\n- 692 scheme.ComputeFaceVertexMask(fHood, fMask);\n- 693\n- 694 // Apply the weights to the parent face's vertices:\n- 695 dst[cVertValue].Clear();\n- 696\n- 697 for (int i = 0; i < fValues.size(); ++i) {\n- 698 dst[cVertValue].AddWithWeight(src[fValues[i]], fValueWeights[i]);\n- 699 }\n- 700 }\n- 701}\n- 702\n- 703template \n- 704template \n- 705inline void\n- 706PrimvarRefinerReal::interpFVarFromEdges(int level, T const & src, U &\n-dst, int channel) const {\n- 707\n- 708 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n-(level-1);\n- 709\n- 710 Sdc::Scheme scheme(_refiner._subdivOptions);\n- 711\n- 712 Vtr::internal::Level const & parentLevel = refinement.parent();\n- 713 Vtr::internal::Level const & childLevel = refinement.child();\n- 714\n- 715 Vtr::internal::FVarRefinement const & refineFVar =\n-refinement.getFVarRefinement(channel);\n- 716 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel\n-(channel);\n- 717 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel\n-(channel);\n- 718\n- 719 //\n- 720 // Allocate and initialize (if linearly interpolated) interpolation\n-weights for\n- 721 // the edge mask:\n- 722 //\n- 723 Weight eVertWeights[2];\n- 724 Vtr::internal::StackBuffer eFaceWeights\n-(parentLevel.getMaxEdgeFaces());\n- 725\n- 726 Mask eMask(eVertWeights, 0, eFaceWeights);\n- 727\n- 728 bool isLinearFVar = parentFVar.isLinear() || (_refiner._subdivType ==\n-Sdc::SCHEME_BILINEAR);\n- 729 if (isLinearFVar) {\n- 730 eMask.SetNumVertexWeights(2);\n- 731 eMask.SetNumEdgeWeights(0);\n- 732 eMask.SetNumFaceWeights(0);\n- 733\n- 734 eVertWeights[0] = 0.5f;\n- 735 eVertWeights[1] = 0.5f;\n- 736 }\n- 737\n- 738 Vtr::internal::EdgeInterface eHood(parentLevel);\n- 739\n- 740 for (int edge = 0; edge < parentLevel.getNumEdges(); ++edge) {\n- 741\n- 742 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);\n- 743 if (!Vtr::IndexIsValid(cVert))\n- 744 continue;\n- 745\n- 746 ConstIndexArray cVertValues = childFVar.getVertexValues(cVert);\n- 747\n- 748 bool fvarEdgeVertMatchesVertex = childFVar.valueTopologyMatches\n-(cVertValues[0]);\n- 749 if (fvarEdgeVertMatchesVertex) {\n- 750 //\n- 751 // If smoothly interpolated, compute new weights for the edge mask:\n- 752 //\n- 753 if (!isLinearFVar) {\n- 754 eHood.SetIndex(edge);\n- 755\n- 756 Sdc::Crease::Rule pRule = (parentLevel.getEdgeSharpness(edge) > 0.0f)\n- 757 ? Sdc::Crease::RULE_CREASE : Sdc::Crease::RULE_SMOOTH;\n- 758 Sdc::Crease::Rule cRule = childLevel.getVertexRule(cVert);\n- 759\n- 760 scheme.ComputeEdgeVertexMask(eHood, eMask, pRule, cRule);\n- 761 }\n- 762\n- 763 // Apply the weights to the parent edge's vertices and (if applicable) to\n- 764 // the child vertices of its incident faces:\n- 765 //\n- 766 // Even though the face-varying topology matches the vertex topology, we\n-need\n- 767 // to be careful here when getting values corresponding to the two end-\n-vertices.\n- 768 // While the edge may be continuous, the vertices at their ends may have\n- 769 // discontinuities elsewhere in their neighborhood (i.e. on the \"other\n-side\"\n- 770 // of the end-vertex) and so have sibling values associated with them. In\n-most\n- 771 // cases the topology for an end-vertex will match and we can use it\n-directly,\n- 772 // but we must still check and retrieve as needed.\n- 773 //\n- 774 // Indices for values corresponding to face-vertices are guaranteed to\n-match,\n- 775 // so we can use the child-vertex indices directly.\n- 776 //\n- 777 // And by \"directly\", we always use getVertexValue(vertexIndex) to\n-reference\n- 778 // values in the \"src\" to account for the possible indirection that may\n-exist at\n- 779 // level 0 -- where there may be fewer values than vertices and an\n-additional\n- 780 // indirection is necessary. We can use a vertex index directly for \"dst\"\n-when\n- 781 // it matches.\n- 782 //\n- 783 Vtr::Index eVertValues[2];\n- 784\n- 785 parentFVar.getEdgeFaceValues(edge, 0, eVertValues);\n- 786\n- 787 Index cVertValue = cVertValues[0];\n- 788\n- 789 dst[cVertValue].Clear();\n- 790 dst[cVertValue].AddWithWeight(src[eVertValues[0]], eVertWeights[0]);\n- 791 dst[cVertValue].AddWithWeight(src[eVertValues[1]], eVertWeights[1]);\n- 792\n- 793 if (eMask.GetNumFaceWeights() > 0) {\n- 794\n- 795 ConstIndexArray eFaces = parentLevel.getEdgeFaces(edge);\n- 796\n- 797 for (int i = 0; i < eFaces.size(); ++i) {\n- 798 if (eMask.AreFaceWeightsForFaceCenters()) {\n- 799\n- 800 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(eFaces[i]);\n- 801 assert(Vtr::IndexIsValid(cVertOfFace));\n- 802\n- 803 Vtr::Index cValueOfFace = childFVar.getVertexValueOffset(cVertOfFace);\n- 804 dst[cVertValue].AddWithWeight(dst[cValueOfFace], eFaceWeights[i]);\n- 805 } else {\n- 806 Vtr::Index pFace = eFaces[i];\n- 807 ConstIndexArray pFaceEdges = parentLevel.getFaceEdges(pFace),\n- 808 pFaceVerts = parentLevel.getFaceVertices(pFace);\n- 809\n- 810 int eInFace = 0;\n- 811 for ( ; pFaceEdges[eInFace] != edge; ++eInFace ) ;\n- 812\n- 813 // Edge \"i\" spans vertices [i,i+1] so we want i+2...\n- 814 int vInFace = eInFace + 2;\n- 815 if (vInFace >= pFaceVerts.size()) vInFace -= pFaceVerts.size();\n- 816\n- 817 Vtr::Index pValueNext = parentFVar.getFaceValues(pFace)[vInFace];\n- 818 dst[cVertValue].AddWithWeight(src[pValueNext], eFaceWeights[i]);\n- 819 }\n- 820 }\n- 821 }\n- 822 } else {\n- 823 //\n- 824 // Mismatched edge-verts should just be linearly interpolated between the\n-pairs of\n- 825 // values for each sibling of the child edge-vertex -- the question is:\n-which face\n- 826 // holds that pair of values for a given sibling?\n- 827 //\n- 828 // In the manifold case, the sibling and edge-face indices will\n-correspond. We\n- 829 // will eventually need to update this to account for > 3 incident faces.\n- 830 //\n- 831 for (int i = 0; i < cVertValues.size(); ++i) {\n- 832 Vtr::Index eVertValues[2];\n- 833 int eFaceIndex = refineFVar.getChildValueParentSource(cVert, i);\n- 834 assert(eFaceIndex == i);\n- 835\n- 836 parentFVar.getEdgeFaceValues(edge, eFaceIndex, eVertValues);\n- 837\n- 838 Index cVertValue = cVertValues[i];\n- 839\n- 840 dst[cVertValue].Clear();\n- 841 dst[cVertValue].AddWithWeight(src[eVertValues[0]], 0.5);\n- 842 dst[cVertValue].AddWithWeight(src[eVertValues[1]], 0.5);\n- 843 }\n- 844 }\n- 845 }\n- 846}\n- 847\n- 848template \n- 849template \n- 850inline void\n- 851PrimvarRefinerReal::interpFVarFromVerts(int level, T const & src, U &\n-dst, int channel) const {\n- 852\n- 853 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n-(level-1);\n- 854\n- 855 Sdc::Scheme scheme(_refiner._subdivOptions);\n- 856\n- 857 Vtr::internal::Level const & parentLevel = refinement.parent();\n- 858 Vtr::internal::Level const & childLevel = refinement.child();\n- 859\n- 860 Vtr::internal::FVarRefinement const & refineFVar =\n-refinement.getFVarRefinement(channel);\n- 861 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel\n-(channel);\n- 862 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel\n-(channel);\n- 863\n- 864 bool isLinearFVar = parentFVar.isLinear() || (_refiner._subdivType ==\n-Sdc::SCHEME_BILINEAR);\n- 865\n- 866 Vtr::internal::StackBuffer weightBuffer\n-(2*parentLevel.getMaxValence());\n- 867\n- 868 Vtr::internal::StackBuffer vEdgeValues\n-(parentLevel.getMaxValence());\n- 869\n- 870 Vtr::internal::VertexInterface vHood(parentLevel, childLevel);\n- 871\n- 872 for (int vert = 0; vert < parentLevel.getNumVertices(); ++vert) {\n- 873\n- 874 Vtr::Index cVert = refinement.getVertexChildVertex(vert);\n- 875 if (!Vtr::IndexIsValid(cVert))\n- 876 continue;\n- 877\n- 878 ConstIndexArray pVertValues = parentFVar.getVertexValues(vert),\n- 879 cVertValues = childFVar.getVertexValues(cVert);\n- 880\n- 881 bool fvarVertVertMatchesVertex = childFVar.valueTopologyMatches\n-(cVertValues[0]);\n- 882 if (isLinearFVar && fvarVertVertMatchesVertex) {\n- 883 dst[cVertValues[0]].Clear();\n- 884 dst[cVertValues[0]].AddWithWeight(src[pVertValues[0]], 1.0f);\n- 885 continue;\n- 886 }\n- 887\n- 888 if (fvarVertVertMatchesVertex) {\n- 889 //\n- 890 // Declare and compute mask weights for this vertex relative to its parent\n-edge:\n- 891 //\n- 892 // (We really need to encapsulate this somewhere else for use here and in\n-the\n- 893 // general case)\n- 894 //\n- 895 ConstIndexArray vEdges = parentLevel.getVertexEdges(vert);\n- 896\n- 897 Weight vVertWeight;\n- 898 Weight * vEdgeWeights = weightBuffer;\n- 899 Weight * vFaceWeights = vEdgeWeights + vEdges.size();\n- 900\n- 901 Mask vMask(&vVertWeight, vEdgeWeights, vFaceWeights);\n- 902\n- 903 vHood.SetIndex(vert, cVert);\n- 904\n- 905 Sdc::Crease::Rule pRule = parentLevel.getVertexRule(vert);\n- 906 Sdc::Crease::Rule cRule = childLevel.getVertexRule(cVert);\n- 907\n- 908 scheme.ComputeVertexVertexMask(vHood, vMask, pRule, cRule);\n- 909\n- 910 // Apply the weights to the parent vertex, the vertices opposite its\n-incident\n- 911 // edges, and the child vertices of its incident faces:\n- 912 //\n- 913 // Even though the face-varying topology matches the vertex topology, we\n-need\n- 914 // to be careful here when getting values corresponding to vertices at the\n- 915 // ends of edges. While the edge may be continuous, the end vertex may\n-have\n- 916 // discontinuities elsewhere in their neighborhood (i.e. on the \"other\n-side\"\n- 917 // of the end-vertex) and so have sibling values associated with them. In\n-most\n- 918 // cases the topology for an end-vertex will match and we can use it\n-directly,\n- 919 // but we must still check and retrieve as needed.\n- 920 //\n- 921 // Indices for values corresponding to face-vertices are guaranteed to\n-match,\n- 922 // so we can use the child-vertex indices directly.\n- 923 //\n- 924 // And by \"directly\", we always use getVertexValue(vertexIndex) to\n-reference\n- 925 // values in the \"src\" to account for the possible indirection that may\n-exist at\n- 926 // level 0 -- where there may be fewer values than vertices and an\n-additional\n- 927 // indirection is necessary. We can use a vertex index directly for \"dst\"\n-when\n- 928 // it matches.\n- 929 //\n- 930 // As with applying the mask to vertex data, in order to improve numerical\n- 931 // precision, it's better to apply smaller weights first, so begin with\n-the\n- 932 // face-weights followed by the edge-weights and the vertex weight last.\n- 933 //\n- 934 Vtr::Index pVertValue = pVertValues[0];\n- 935 Vtr::Index cVertValue = cVertValues[0];\n- 936\n- 937 dst[cVertValue].Clear();\n- 938 if (vMask.GetNumFaceWeights() > 0) {\n- 939 assert(vMask.AreFaceWeightsForFaceCenters());\n- 940\n- 941 ConstIndexArray vFaces = parentLevel.getVertexFaces(vert);\n- 942\n- 943 for (int i = 0; i < vFaces.size(); ++i) {\n- 944\n- 945 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(vFaces[i]);\n- 946 assert(Vtr::IndexIsValid(cVertOfFace));\n- 947\n- 948 Vtr::Index cValueOfFace = childFVar.getVertexValueOffset(cVertOfFace);\n- 949 dst[cVertValue].AddWithWeight(dst[cValueOfFace], vFaceWeights[i]);\n- 950 }\n- 951 }\n- 952 if (vMask.GetNumEdgeWeights() > 0) {\n- 953\n- 954 parentFVar.getVertexEdgeValues(vert, vEdgeValues);\n- 955\n- 956 for (int i = 0; i < vEdges.size(); ++i) {\n- 957 dst[cVertValue].AddWithWeight(src[vEdgeValues[i]], vEdgeWeights[i]);\n- 958 }\n- 959 }\n- 960 dst[cVertValue].AddWithWeight(src[pVertValue], vVertWeight);\n- 961 } else {\n- 962 //\n- 963 // Each FVar value associated with a vertex will be either a corner or a\n-crease,\n- 964 // or potentially in transition from corner to crease:\n- 965 // - if the CHILD is a corner, there can be no transition so we have a\n-corner\n- 966 // - otherwise if the PARENT is a crease, both will be creases (no\n-transition)\n- 967 // - otherwise the parent must be a corner and the child a crease\n-(transition)\n- 968 //\n- 969 Vtr::internal::FVarLevel::ConstValueTagArray pValueTags =\n-parentFVar.getVertexValueTags(vert);\n- 970 Vtr::internal::FVarLevel::ConstValueTagArray cValueTags =\n-childFVar.getVertexValueTags(cVert);\n- 971\n- 972 for (int cSiblingIndex = 0; cSiblingIndex < cVertValues.size();\n-++cSiblingIndex) {\n- 973 int pSiblingIndex = refineFVar.getChildValueParentSource(cVert,\n-cSiblingIndex);\n- 974 assert(pSiblingIndex == cSiblingIndex);\n- 975\n- 976 typedef Vtr::internal::FVarLevel::Sibling SiblingIntType;\n- 977\n- 978 SiblingIntType cSibling = (SiblingIntType) cSiblingIndex;\n- 979 SiblingIntType pSibling = (SiblingIntType) pSiblingIndex;\n- 980\n- 981 Vtr::Index pVertValue = pVertValues[pSibling];\n- 982 Vtr::Index cVertValue = cVertValues[cSibling];\n- 983\n- 984 dst[cVertValue].Clear();\n- 985 if (isLinearFVar || cValueTags[cSibling].isCorner()) {\n- 986 dst[cVertValue].AddWithWeight(src[pVertValue], 1.0f);\n- 987 } else {\n- 988 //\n- 989 // We have either a crease or a transition from corner to crease -- in\n- 990 // either case, we need the end values for the full/fractional crease:\n- 991 //\n- 992 Index pEndValues[2];\n- 993 parentFVar.getVertexCreaseEndValues(vert, pSibling, pEndValues);\n- 994\n- 995 Weight vWeight = 0.75f;\n- 996 Weight eWeight = 0.125f;\n- 997\n- 998 //\n- 999 // If semi-sharp we need to apply fractional weighting -- if made sharp\n-because\n- 1000 // of the other sibling (dependent-sharp) use the fractional weight from\n-that\n- 1001 // other sibling (should only occur when there are 2):\n- 1002 //\n- 1003 if (pValueTags[pSibling].isSemiSharp()) {\n- 1004 Weight wCorner = pValueTags[pSibling].isDepSharp()\n- 1005 ? refineFVar.getFractionalWeight(vert, !pSibling, cVert, !cSibling)\n- 1006 : refineFVar.getFractionalWeight(vert, pSibling, cVert, cSibling);\n- 1007 Weight wCrease = 1.0f - wCorner;\n- 1008\n- 1009 vWeight = wCrease * 0.75f + wCorner;\n- 1010 eWeight = wCrease * 0.125f;\n- 1011 }\n- 1012 dst[cVertValue].AddWithWeight(src[pEndValues[0]], eWeight);\n- 1013 dst[cVertValue].AddWithWeight(src[pEndValues[1]], eWeight);\n- 1014 dst[cVertValue].AddWithWeight(src[pVertValue], vWeight);\n- 1015 }\n- 1016 }\n- 1017 }\n- 1018 }\n- 1019}\n- 1020\n- 1021template \n- 1022template \n- 1023inline void\n- 1024PrimvarRefinerReal::limit(T const & src, U & dstPos, U1 *\n-dstTan1Ptr, U2 * dstTan2Ptr) const {\n- 1025\n- 1026 Sdc::Scheme scheme(_refiner._subdivOptions);\n- 1027\n- 1028 Vtr::internal::Level const & level = _refiner.getLevel\n-(_refiner.GetMaxLevel());\n- 1029\n- 1030 int maxWeightsPerMask = 1 + 2 * level.getMaxValence();\n- 1031 bool hasTangents = (dstTan1Ptr && dstTan2Ptr);\n- 1032 int numMasks = 1 + (hasTangents ? 2 : 0);\n- 1033\n- 1034 Vtr::internal::StackBuffer indexBuffer(maxWeightsPerMask);\n- 1035 Vtr::internal::StackBuffer weightBuffer(numMasks *\n-maxWeightsPerMask);\n- 1036\n- 1037 Weight * vPosWeights = weightBuffer,\n- 1038 * ePosWeights = vPosWeights + 1,\n- 1039 * fPosWeights = ePosWeights + level.getMaxValence();\n- 1040 Weight * vTan1Weights = vPosWeights + maxWeightsPerMask,\n- 1041 * eTan1Weights = ePosWeights + maxWeightsPerMask,\n- 1042 * fTan1Weights = fPosWeights + maxWeightsPerMask;\n- 1043 Weight * vTan2Weights = vTan1Weights + maxWeightsPerMask,\n- 1044 * eTan2Weights = eTan1Weights + maxWeightsPerMask,\n- 1045 * fTan2Weights = fTan1Weights + maxWeightsPerMask;\n- 1046\n- 1047 Mask posMask( vPosWeights, ePosWeights, fPosWeights);\n- 1048 Mask tan1Mask(vTan1Weights, eTan1Weights, fTan1Weights);\n- 1049 Mask tan2Mask(vTan2Weights, eTan2Weights, fTan2Weights);\n- 1050\n- 1051 // This is a bit obscure -- assigning both parent and child as last level\n--- but\n- 1052 // this mask type was intended for another purpose. Consider one for the\n-limit:\n- 1053 Vtr::internal::VertexInterface vHood(level, level);\n- 1054\n- 1055 for (int vert = 0; vert < level.getNumVertices(); ++vert) {\n- 1056 ConstIndexArray vEdges = level.getVertexEdges(vert);\n- 1057\n- 1058 // Incomplete vertices (present in sparse refinement) do not have their\n-full\n- 1059 // topological neighborhood to determine a proper limit -- just leave the\n- 1060 // vertex at the refined location and continue to the next:\n- 1061 if (level.getVertexTag(vert)._incomplete || (vEdges.size() == 0)) {\n- 1062 dstPos[vert].Clear();\n- 1063 dstPos[vert].AddWithWeight(src[vert], 1.0);\n- 1064 if (hasTangents) {\n- 1065 (*dstTan1Ptr)[vert].Clear();\n- 1066 (*dstTan2Ptr)[vert].Clear();\n- 1067 }\n- 1068 continue;\n- 1069 }\n- 1070\n- 1071 //\n- 1072 // Limit masks require the subdivision Rule for the vertex in order to\n-deal\n- 1073 // with infinitely sharp features correctly -- including boundaries and\n-corners.\n- 1074 // The vertex neighborhood is minimally defined with vertex and edge\n-counts.\n- 1075 //\n- 1076 Sdc::Crease::Rule vRule = level.getVertexRule(vert);\n- 1077\n- 1078 // This is a bit obscure -- child vertex index will be ignored here\n- 1079 vHood.SetIndex(vert, vert);\n- 1080\n- 1081 if (hasTangents) {\n- 1082 scheme.ComputeVertexLimitMask(vHood, posMask, tan1Mask, tan2Mask, vRule);\n- 1083 } else {\n- 1084 scheme.ComputeVertexLimitMask(vHood, posMask, vRule);\n- 1085 }\n- 1086\n- 1087 //\n- 1088 // Gather the neighboring vertices of this vertex -- the vertices\n-opposite its\n- 1089 // incident edges, and the opposite vertices of its incident faces:\n- 1090 //\n- 1091 Index * eIndices = indexBuffer;\n- 1092 Index * fIndices = indexBuffer + vEdges.size();\n- 1093\n- 1094 for (int i = 0; i < vEdges.size(); ++i) {\n- 1095 ConstIndexArray eVerts = level.getEdgeVertices(vEdges[i]);\n- 1096\n- 1097 eIndices[i] = (eVerts[0] == vert) ? eVerts[1] : eVerts[0];\n- 1098 }\n- 1099 if (posMask.GetNumFaceWeights() || (hasTangents &&\n-tan1Mask.GetNumFaceWeights())) {\n- 1100 ConstIndexArray vFaces = level.getVertexFaces(vert);\n- 1101 ConstLocalIndexArray vInFace = level.getVertexFaceLocalIndices(vert);\n- 1102\n- 1103 for (int i = 0; i < vFaces.size(); ++i) {\n- 1104 ConstIndexArray fVerts = level.getFaceVertices(vFaces[i]);\n- 1105\n- 1106 LocalIndex vOppInFace = (vInFace[i] + 2);\n- 1107 if (vOppInFace >= fVerts.size()) vOppInFace -= (LocalIndex)fVerts.size();\n- 1108\n- 1109 fIndices[i] = level.getFaceVertices(vFaces[i])[vOppInFace];\n- 1110 }\n- 1111 }\n- 1112\n- 1113 //\n- 1114 // Combine the weights and indices for position and tangents. As with\n-applying\n- 1115 // refinement masks to vertex data, in order to improve numerical\n-precision, it's\n- 1116 // better to apply smaller weights first, so begin with the face-weights\n-followed\n- 1117 // by the edge-weights and the vertex weight last.\n- 1118 //\n- 1119 dstPos[vert].Clear();\n- 1120 for (int i = 0; i < posMask.GetNumFaceWeights(); ++i) {\n- 1121 dstPos[vert].AddWithWeight(src[fIndices[i]], fPosWeights[i]);\n- 1122 }\n- 1123 for (int i = 0; i < posMask.GetNumEdgeWeights(); ++i) {\n- 1124 dstPos[vert].AddWithWeight(src[eIndices[i]], ePosWeights[i]);\n- 1125 }\n- 1126 dstPos[vert].AddWithWeight(src[vert], vPosWeights[0]);\n- 1127\n- 1128 //\n- 1129 // Apply the tangent masks -- both will have the same number of weights\n-and\n- 1130 // indices (one tangent may be \"padded\" to accommodate the other), but\n-these\n- 1131 // may differ from those of the position:\n- 1132 //\n- 1133 if (hasTangents) {\n- 1134 assert(tan1Mask.GetNumFaceWeights() == tan2Mask.GetNumFaceWeights());\n- 1135 assert(tan1Mask.GetNumEdgeWeights() == tan2Mask.GetNumEdgeWeights());\n- 1136\n- 1137 U1 & dstTan1 = *dstTan1Ptr;\n- 1138 U2 & dstTan2 = *dstTan2Ptr;\n- 1139\n- 1140 dstTan1[vert].Clear();\n- 1141 dstTan2[vert].Clear();\n- 1142 for (int i = 0; i < tan1Mask.GetNumFaceWeights(); ++i) {\n- 1143 dstTan1[vert].AddWithWeight(src[fIndices[i]], fTan1Weights[i]);\n- 1144 dstTan2[vert].AddWithWeight(src[fIndices[i]], fTan2Weights[i]);\n- 1145 }\n- 1146 for (int i = 0; i < tan1Mask.GetNumEdgeWeights(); ++i) {\n- 1147 dstTan1[vert].AddWithWeight(src[eIndices[i]], eTan1Weights[i]);\n- 1148 dstTan2[vert].AddWithWeight(src[eIndices[i]], eTan2Weights[i]);\n- 1149 }\n- 1150 dstTan1[vert].AddWithWeight(src[vert], vTan1Weights[0]);\n- 1151 dstTan2[vert].AddWithWeight(src[vert], vTan2Weights[0]);\n- 1152 }\n- 1153 }\n- 1154}\n- 1155\n- 1156template \n- 1157template \n- 1158inline void\n- 1159PrimvarRefinerReal::limitFVar(T const & src, U & dst, int channel)\n-const {\n- 1160\n- 1161 Sdc::Scheme scheme(_refiner._subdivOptions);\n- 1162\n- 1163 Vtr::internal::Level const & level = _refiner.getLevel\n-(_refiner.GetMaxLevel());\n- 1164 Vtr::internal::FVarLevel const & fvarChannel = level.getFVarLevel\n-(channel);\n- 1165\n- 1166 int maxWeightsPerMask = 1 + 2 * level.getMaxValence();\n- 1167\n- 1168 Vtr::internal::StackBuffer weightBuffer(maxWeightsPerMask);\n- 1169 Vtr::internal::StackBuffer vEdgeBuffer(level.getMaxValence());\n- 1170\n- 1171 // This is a bit obscure -- assign both parent and child as last level\n- 1172 Vtr::internal::VertexInterface vHood(level, level);\n- 1173\n- 1174 for (int vert = 0; vert < level.getNumVertices(); ++vert) {\n- 1175\n- 1176 ConstIndexArray vEdges = level.getVertexEdges(vert);\n- 1177 ConstIndexArray vValues = fvarChannel.getVertexValues(vert);\n- 1178\n- 1179 // Incomplete vertices (present in sparse refinement) do not have their\n-full\n- 1180 // topological neighborhood to determine a proper limit -- just leave the\n- 1181 // values (perhaps more than one per vertex) at the refined location.\n- 1182 //\n- 1183 // The same can be done if the face-varying channel is purely linear.\n- 1184 //\n- 1185 bool isIncomplete = (level.getVertexTag(vert)._incomplete || (vEdges.size\n-() == 0));\n- 1186 if (isIncomplete || fvarChannel.isLinear()) {\n- 1187 for (int i = 0; i < vValues.size(); ++i) {\n- 1188 Vtr::Index vValue = vValues[i];\n- 1189\n- 1190 dst[vValue].Clear();\n- 1191 dst[vValue].AddWithWeight(src[vValue], 1.0f);\n- 1192 }\n- 1193 continue;\n- 1194 }\n- 1195\n- 1196 bool fvarVertMatchesVertex = fvarChannel.valueTopologyMatches(vValues\n-[0]);\n- 1197 if (fvarVertMatchesVertex) {\n- 1198\n- 1199 // Assign the mask weights to the common buffer and compute the mask:\n- 1200 //\n- 1201 Weight * vWeights = weightBuffer,\n- 1202 * eWeights = vWeights + 1,\n- 1203 * fWeights = eWeights + vEdges.size();\n- 1204\n- 1205 Mask vMask(vWeights, eWeights, fWeights);\n- 1206\n- 1207 vHood.SetIndex(vert, vert);\n- 1208\n- 1209 scheme.ComputeVertexLimitMask(vHood, vMask, level.getVertexRule(vert));\n- 1210\n- 1211 //\n- 1212 // Apply mask to corresponding FVar values for neighboring vertices:\n- 1213 //\n- 1214 Vtr::Index vValue = vValues[0];\n- 1215\n- 1216 dst[vValue].Clear();\n- 1217 if (vMask.GetNumFaceWeights() > 0) {\n- 1218 assert(!vMask.AreFaceWeightsForFaceCenters());\n- 1219\n- 1220 ConstIndexArray vFaces = level.getVertexFaces(vert);\n- 1221 ConstLocalIndexArray vInFace = level.getVertexFaceLocalIndices(vert);\n- 1222\n- 1223 for (int i = 0; i < vFaces.size(); ++i) {\n- 1224 ConstIndexArray faceValues = fvarChannel.getFaceValues(vFaces[i]);\n- 1225 LocalIndex vOppInFace = vInFace[i] + 2;\n- 1226 if (vOppInFace >= faceValues.size()) vOppInFace -= faceValues.size();\n- 1227\n- 1228 Index vValueOppositeFace = faceValues[vOppInFace];\n- 1229\n- 1230 dst[vValue].AddWithWeight(src[vValueOppositeFace], fWeights[i]);\n- 1231 }\n- 1232 }\n- 1233 if (vMask.GetNumEdgeWeights() > 0) {\n- 1234 Index * vEdgeValues = vEdgeBuffer;\n- 1235 fvarChannel.getVertexEdgeValues(vert, vEdgeValues);\n- 1236\n- 1237 for (int i = 0; i < vEdges.size(); ++i) {\n- 1238 dst[vValue].AddWithWeight(src[vEdgeValues[i]], eWeights[i]);\n- 1239 }\n- 1240 }\n- 1241 dst[vValue].AddWithWeight(src[vValue], vWeights[0]);\n- 1242 } else {\n- 1243 //\n- 1244 // Sibling FVar values associated with a vertex will be either a corner\n-or a crease:\n- 1245 //\n- 1246 for (int i = 0; i < vValues.size(); ++i) {\n- 1247 Vtr::Index vValue = vValues[i];\n- 1248\n- 1249 dst[vValue].Clear();\n- 1250 if (fvarChannel.getValueTag(vValue).isCorner()) {\n- 1251 dst[vValue].AddWithWeight(src[vValue], 1.0f);\n- 1252 } else {\n- 1253 Index vEndValues[2];\n- 1254 fvarChannel.getVertexCreaseEndValues(vert, i, vEndValues);\n- 1255\n- 1256 dst[vValue].AddWithWeight(src[vEndValues[0]], 1.0f/6.0f);\n- 1257 dst[vValue].AddWithWeight(src[vEndValues[1]], 1.0f/6.0f);\n- 1258 dst[vValue].AddWithWeight(src[vValue], 2.0f/3.0f);\n- 1259 }\n- 1260 }\n- 1261 }\n- 1262 }\n- 1263}\n- 1264\n-1265class PrimvarRefiner : public PrimvarRefinerReal {\n- 1266public:\n-1267 PrimvarRefiner(TopologyRefiner const & refiner)\n- 1268 : PrimvarRefinerReal(refiner) { }\n- 1269};\n- 1270\n- 1271} // end namespace Far\n- 1272\n- 1273} // end namespace OPENSUBDIV_VERSION\n- 1274using namespace OPENSUBDIV_VERSION;\n- 1275} // end namespace OpenSubdiv\n- 1276\n- 1277#endif /* OPENSUBDIV3_FAR_PRIMVAR_REFINER_H */\n+ 290\n+ 291private:\n+292 friend class TopologyRefiner;\n+ 293\n+ 294 Vtr::internal::Level const * _level;\n+ 295 Vtr::internal::Refinement const * _refToParent;\n+ 296 Vtr::internal::Refinement const * _refToChild;\n+ 297\n+ 298public:\n+ 299 // Not intended for public use, but required by std::vector, etc...\n+300 TopologyLevel() { }\n+301 ~TopologyLevel() { }\n+ 302};\n+ 303\n+ 304} // end namespace Far\n+ 305\n+ 306} // end namespace OPENSUBDIV_VERSION\n+ 307using namespace OPENSUBDIV_VERSION;\n+ 308} // end namespace OpenSubdiv\n+ 309\n+ 310#endif /* OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H */\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LocalIndex\n-Vtr::LocalIndex LocalIndex\n-Definition: types.h:42\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::Error\n-void Error(ErrorType err, const char *format,...)\n-Sends an OSD error with a message (internal use only)\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstIndexArray\n-Vtr::ConstIndexArray ConstIndexArray\n-Definition: types.h:47\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstLocalIndexArray\n-Vtr::ConstLocalIndexArray ConstLocalIndexArray\n-Definition: types.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::FAR_RUNTIME_ERROR\n-@ FAR_RUNTIME_ERROR\n-Issue a generic runtime error, but continue execution.\n-Definition: error.h:40\n OpenSubdiv::OPENSUBDIV_VERSION::Far::Index\n Vtr::Index Index\n Definition: types.h:41\n-OpenSubdiv::OPENSUBDIV_VERSION::Vtr::Index\n-int Index\n-Definition: types.h:54\n-OpenSubdiv::OPENSUBDIV_VERSION::Vtr::IndexIsValid\n-bool IndexIsValid(Index index)\n-Definition: types.h:58\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SCHEME_BILINEAR\n-@ SCHEME_BILINEAR\n-Definition: types.h:38\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SCHEME_CATMARK\n-@ SCHEME_CATMARK\n-Definition: types.h:39\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SCHEME_LOOP\n-@ SCHEME_LOOP\n-Definition: types.h:40\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal\n-Applies refinement operations to generic primvar data.\n-Definition: primvarRefiner.h:56\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Limit\n-void Limit(T const &src, U &dstPos) const\n-Apply limit weights to a primvar buffer.\n-Definition: primvarRefiner.h:319\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::PrimvarRefinerReal\n-PrimvarRefinerReal(TopologyRefiner const &refiner)\n-Definition: primvarRefiner.h:59\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Limit\n-void Limit(T const &src, U &dstPos, U1 &dstTan1, U2 &dstTan2) const\n-Definition: primvarRefiner.h:344\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateFaceUniform\n-void InterpolateFaceUniform(int level, T const &src, U &dst) const\n-Refine uniform (per-face) primvar data between levels.\n-Definition: primvarRefiner.h:394\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::GetTopologyRefiner\n-TopologyRefiner const & GetTopologyRefiner() const\n-Definition: primvarRefiner.h:62\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateVarying\n-void InterpolateVarying(int level, T const &src, U &dst) const\n-Apply only varying interpolation weights to a primvar buffer for a single level\n-of refinement.\n-Definition: primvarRefiner.h:412\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Interpolate\n-void Interpolate(int level, T const &src, U &dst) const\n-Apply vertex interpolation weights to a primvar buffer for a single level of\n-refinement.\n-Definition: primvarRefiner.h:267\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateFaceVarying\n-void InterpolateFaceVarying(int level, T const &src, U &dst, int channel=0)\n-const\n-Apply face-varying interpolation weights to a primvar buffer associated with a\n-particular face-varyin...\n-Definition: primvarRefiner.h:293\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::~PrimvarRefinerReal\n-~PrimvarRefinerReal()\n-Definition: primvarRefiner.h:60\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::LimitFaceVarying\n-void LimitFaceVarying(T const &src, U &dst, int channel=0) const\n-Definition: primvarRefiner.h:369\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefiner\n-Definition: primvarRefiner.h:1265\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefiner::PrimvarRefiner\n-PrimvarRefiner(TopologyRefiner const &refiner)\n-Definition: primvarRefiner.h:1267\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel\n+An interface for accessing data in a specific level of a refined topology\n+hierarchy.\n+Definition: topologyLevel.h:49\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceParentFace\n+Index GetFaceParentFace(Index f) const\n+Return the parent face (in the previous level) of a given face.\n+Definition: topologyLevel.h:280\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::TopologyLevel\n+TopologyLevel()\n+Definition: topologyLevel.h:300\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexValenceRegular\n+bool IsVertexValenceRegular(Index v) const\n+Return if the valence of the vertex is regular (must be manifold)\n+Definition: topologyLevel.h:162\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeSharpness\n+float GetEdgeSharpness(Index e) const\n+Return the sharpness assigned a given edge.\n+Definition: topologyLevel.h:173\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesFaceFVarTopologyMatch\n+bool DoesFaceFVarTopologyMatch(Index f, int channel=0) const\n+Return if face-varying topology around a face matches.\n+Definition: topologyLevel.h:252\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexRule\n+Sdc::Crease::Rule GetVertexRule(Index v) const\n+Return the subdivision rule assigned a given vertex specific to this level.\n+Definition: topologyLevel.h:194\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexCorner\n+bool IsVertexCorner(Index v) const\n+Return if the vertex is a corner (only one incident face)\n+Definition: topologyLevel.h:140\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexChildVertex\n+Index GetVertexChildVertex(Index v) const\n+Return the child vertex (in the next level) of a given vertex.\n+Definition: topologyLevel.h:277\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeFaceLocalIndices\n+ConstLocalIndexArray GetEdgeFaceLocalIndices(Index e) const\n+Access the local indices of an edge with respect to its incident faces.\n+Definition: topologyLevel.h:117\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildEdges\n+ConstIndexArray GetFaceChildEdges(Index f) const\n+Access the child edges (in the next level) of a given face.\n+Definition: topologyLevel.h:265\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesEdgeFVarTopologyMatch\n+bool DoesEdgeFVarTopologyMatch(Index e, int channel=0) const\n+Return if face-varying topology across the edge only matches.\n+Definition: topologyLevel.h:247\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::PrintTopology\n+void PrintTopology(bool children=true) const\n+Definition: topologyLevel.h:287\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceVertices\n+ConstIndexArray GetFaceVertices(Index f) const\n+Access the vertices incident a given face.\n+Definition: topologyLevel.h:93\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildFaces\n+ConstIndexArray GetFaceChildFaces(Index f) const\n+Access the child faces (in the next level) of a given face.\n+Definition: topologyLevel.h:262\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumEdges\n+int GetNumEdges() const\n+Return the number of edges in this level.\n+Definition: topologyLevel.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::ValidateTopology\n+bool ValidateTopology() const\n+Definition: topologyLevel.h:286\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeNonManifold\n+bool IsEdgeNonManifold(Index e) const\n+Return if the edge is non-manifold.\n+Definition: topologyLevel.h:128\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsFaceHole\n+bool IsFaceHole(Index f) const\n+Return if a given face has been tagged as a hole.\n+Definition: topologyLevel.h:191\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexEdges\n+ConstIndexArray GetVertexEdges(Index v) const\n+Access the edges incident a given vertex.\n+Definition: topologyLevel.h:108\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildVertex\n+Index GetFaceChildVertex(Index f) const\n+Return the child vertex (in the next level) of a given face.\n+Definition: topologyLevel.h:271\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesVertexFVarTopologyMatch\n+bool DoesVertexFVarTopologyMatch(Index v, int channel=0) const\n+Return if face-varying topology around a vertex matches.\n+Definition: topologyLevel.h:242\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeChildEdges\n+ConstIndexArray GetEdgeChildEdges(Index e) const\n+Access the child edges (in the next level) of a given edge.\n+Definition: topologyLevel.h:268\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeInfSharp\n+bool IsEdgeInfSharp(Index e) const\n+Return if the edge is infinitely-sharp.\n+Definition: topologyLevel.h:179\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexFaces\n+ConstIndexArray GetVertexFaces(Index v) const\n+Access the faces incident a given vertex.\n+Definition: topologyLevel.h:105\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::FindEdge\n+Index FindEdge(Index v0, Index v1) const\n+Identify the edge matching the given vertex pair.\n+Definition: topologyLevel.h:120\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexEdgeLocalIndices\n+ConstLocalIndexArray GetVertexEdgeLocalIndices(Index v) const\n+Access the local indices of a vertex with respect to its incident edges.\n+Definition: topologyLevel.h:114\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceEdges\n+ConstIndexArray GetFaceEdges(Index f) const\n+Access the edges incident a given face.\n+Definition: topologyLevel.h:96\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFaces\n+int GetNumFaces() const\n+Return the number of faces in this level.\n+Definition: topologyLevel.h:65\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexBoundary\n+bool IsVertexBoundary(Index v) const\n+Return if the vertex is on a boundary (at least one incident boundary edge)\n+Definition: topologyLevel.h:137\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFVarChannels\n+int GetNumFVarChannels() const\n+Return the number of face-varying channels (should be same for all levels)\n+Definition: topologyLevel.h:230\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumVertices\n+int GetNumVertices() const\n+Return the number of vertices in this level.\n+Definition: topologyLevel.h:62\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeFaces\n+ConstIndexArray GetEdgeFaces(Index e) const\n+Access the faces incident a given edge.\n+Definition: topologyLevel.h:102\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceFVarValues\n+ConstIndexArray GetFaceFVarValues(Index f, int channel=0) const\n+Access the face-varying values associated with a particular face.\n+Definition: topologyLevel.h:237\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeVertices\n+ConstIndexArray GetEdgeVertices(Index e) const\n+Access the vertices incident a given edge.\n+Definition: topologyLevel.h:99\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeSemiSharp\n+bool IsEdgeSemiSharp(Index e) const\n+Return if the edge is semi-sharp.\n+Definition: topologyLevel.h:185\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::~TopologyLevel\n+~TopologyLevel()\n+Definition: topologyLevel.h:301\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexSharpness\n+float GetVertexSharpness(Index v) const\n+Return the sharpness assigned a given vertex.\n+Definition: topologyLevel.h:176\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeChildVertex\n+Index GetEdgeChildVertex(Index e) const\n+Return the child vertex (in the next level) of a given edge.\n+Definition: topologyLevel.h:274\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexSemiSharp\n+bool IsVertexSemiSharp(Index v) const\n+Return if the vertex is semi-sharp.\n+Definition: topologyLevel.h:188\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFaceVertices\n+int GetNumFaceVertices() const\n+Return the total number of face-vertices, i.e. the sum of all vertices for all\n+faces.\n+Definition: topologyLevel.h:71\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexNonManifold\n+bool IsVertexNonManifold(Index v) const\n+Return if the vertex is non-manifold.\n+Definition: topologyLevel.h:131\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeBoundary\n+bool IsEdgeBoundary(Index e) const\n+Return if the edge is a boundary (only one incident face)\n+Definition: topologyLevel.h:134\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexInfSharp\n+bool IsVertexInfSharp(Index v) const\n+Return if the vertex is infinitely-sharp.\n+Definition: topologyLevel.h:182\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFVarValues\n+int GetNumFVarValues(int channel=0) const\n+Return the total number of face-varying values in a particular channel (the\n+upper bound of a face-var...\n+Definition: topologyLevel.h:234\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexFaceLocalIndices\n+ConstLocalIndexArray GetVertexFaceLocalIndices(Index v) const\n+Access the local indices of a vertex with respect to its incident faces.\n+Definition: topologyLevel.h:111\n OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner\n Stores topology data for a specified set of refinement options.\n Definition: topologyRefiner.h:51\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::Rule\n Rule\n Definition: crease.h:82\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::RULE_SMOOTH\n-@ RULE_SMOOTH\n-Definition: crease.h:84\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::RULE_CREASE\n-@ RULE_CREASE\n-Definition: crease.h:86\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme\n-Scheme is a class template which provides all implementation for the\n-subdivision schemes supported by...\n-Definition: scheme.h:66\n OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstArray\n Definition: array.h:53\n-OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstArray::size\n-size_type size() const\n-Definition: array.h:72\n * opensubdiv\n * far\n- * primvarRefiner.h\n+ * topologyLevel.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00722.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00722.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/topologyLevel.h File Reference\n+OpenSubdiv: opensubdiv/far/ptexIndices.h File Reference\n \n \n \n \n \n \n \n@@ -83,29 +83,28 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
topologyLevel.h File Reference
\n+
ptexIndices.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../vtr/level.h"
\n-#include "../vtr/refinement.h"
\n+#include "../far/topologyRefiner.h"
\n #include "../far/types.h"
\n #include <vector>
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n+\n+\n \n

\n Classes

class  TopologyLevel
 An interface for accessing data in a specific level of a refined topology hierarchy. More...
class  PtexIndices
 Object used to compute and query ptex face indices. More...
 
\n \n \n \n \n@@ -114,13 +113,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,30 +5,28 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-topologyLevel.h File Reference\n+ptexIndices.h File Reference\n #include \"../version.h\"\n-#include \"../vtr/level.h\"\n-#include \"../vtr/refinement.h\"\n+#include \"../far/topologyRefiner.h\"\n #include \"../far/types.h\"\n #include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0TopologyLevel\n-\u00a0 An interface for accessing data in a specific level of a refined\n- topology hierarchy. More...\n+class \u00a0PtexIndices\n+\u00a0 Object used to compute and query ptex face indices. More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n * opensubdiv\n * far\n- * topologyLevel.h\n+ * ptexIndices.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00722.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00722.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00722 = [\n- [\"TopologyLevel\", \"a01105.html\", \"a01105\"]\n+ [\"PtexIndices\", \"a01033.html\", \"a01033\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00722_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00722_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/topologyLevel.h Source File\n+OpenSubdiv: opensubdiv/far/ptexIndices.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
topologyLevel.h
\n+
ptexIndices.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2015 DreamWorks Animation LLC.
\n+
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -106,218 +106,79 @@\n
17//
\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n-
24#ifndef OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H
\n-
25#define OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H
\n+
24#ifndef OPENSUBDIV3_FAR_PTEX_INDICES_H
\n+
25#define OPENSUBDIV3_FAR_PTEX_INDICES_H
\n
26
\n
27#include "../version.h"
\n
28
\n-
29#include "../vtr/level.h"
\n-
30#include "../vtr/refinement.h"
\n-
31#include "../far/types.h"
\n-
32
\n-
33#include <vector>
\n-
34
\n-
35namespace OpenSubdiv {
\n-
36namespace OPENSUBDIV_VERSION {
\n-
37
\n-
38namespace Far {
\n-
39
\n-
49class TopologyLevel {
\n-
50
\n-
51public:
\n-
53
\n+
29#include "../far/topologyRefiner.h"
\n+
30#include "../far/types.h"
\n+
31
\n+
32#include <vector>
\n+
33
\n+
34namespace OpenSubdiv {
\n+
35namespace OPENSUBDIV_VERSION {
\n+
36
\n+
37namespace Far {
\n+
38
\n+
46class PtexIndices {
\n+
47
\n+
48public:
\n+
49
\n+
51 PtexIndices(TopologyRefiner const &refiner);
\n+
52
\n+
54 ~PtexIndices();
\n+
55
\n+
57
\n
60
\n-
62 int GetNumVertices() const { return _level->getNumVertices(); }
\n-
63
\n-
65 int GetNumFaces() const { return _level->getNumFaces(); }
\n-
66
\n-
68 int GetNumEdges() const { return _level->getNumEdges(); }
\n-
69
\n-
71 int GetNumFaceVertices() const { return _level->getNumFaceVerticesTotal(); }
\n-
73
\n-
75
\n-
90 //
\n+
63 int GetNumFaces() const;
\n+
64
\n+
67 int GetFaceId(Index f) const;
\n+
68
\n+
83 void GetAdjacency(
\n+
84 TopologyRefiner const &refiner,
\n+
85 int face, int quadrant,
\n+
86 int adjFaces[4], int adjEdges[4]) const;
\n+
87
\n+
89
\n+
90private:
\n
91
\n-
93 ConstIndexArray GetFaceVertices(Index f) const { return _level->getFaceVertices(f); }
\n-
94
\n-
96 ConstIndexArray GetFaceEdges(Index f) const { return _level->getFaceEdges(f); }
\n-
97
\n-
99 ConstIndexArray GetEdgeVertices(Index e) const { return _level->getEdgeVertices(e); }
\n-
100
\n-
102 ConstIndexArray GetEdgeFaces(Index e) const { return _level->getEdgeFaces(e); }
\n-
103
\n-
105 ConstIndexArray GetVertexFaces(Index v) const { return _level->getVertexFaces(v); }
\n-
106
\n-
108 ConstIndexArray GetVertexEdges(Index v) const { return _level->getVertexEdges(v); }
\n-
109
\n-
111 ConstLocalIndexArray GetVertexFaceLocalIndices(Index v) const { return _level->getVertexFaceLocalIndices(v); }
\n-
112
\n-
114 ConstLocalIndexArray GetVertexEdgeLocalIndices(Index v) const { return _level->getVertexEdgeLocalIndices(v); }
\n-
115
\n-
117 ConstLocalIndexArray GetEdgeFaceLocalIndices(Index e) const { return _level->getEdgeFaceLocalIndices(e); }
\n-
118
\n-
120 Index FindEdge(Index v0, Index v1) const { return _level->findEdge(v0, v1); }
\n-
122
\n-
124
\n-
126
\n-
128 bool IsEdgeNonManifold(Index e) const { return _level->isEdgeNonManifold(e); }
\n-
129
\n-
131 bool IsVertexNonManifold(Index v) const { return _level->isVertexNonManifold(v); }
\n-
132
\n-
134 bool IsEdgeBoundary(Index e) const { return _level->getEdgeTag(e)._boundary; }
\n-
135
\n-
137 bool IsVertexBoundary(Index v) const { return _level->getVertexTag(v)._boundary; }
\n-
138
\n-
140 bool IsVertexCorner(Index v) const { return (_level->getNumVertexFaces(v) == 1); }
\n-
141
\n-
162 bool IsVertexValenceRegular(Index v) const { return !_level->getVertexTag(v)._xordinary || IsVertexCorner(v); }
\n-
164
\n-
166
\n-
171
\n-
173 float GetEdgeSharpness(Index e) const { return _level->getEdgeSharpness(e); }
\n-
174
\n-
176 float GetVertexSharpness(Index v) const { return _level->getVertexSharpness(v); }
\n-
177
\n-
179 bool IsEdgeInfSharp(Index e) const { return _level->getEdgeTag(e)._infSharp; }
\n-
180
\n-
182 bool IsVertexInfSharp(Index v) const { return _level->getVertexTag(v)._infSharp; }
\n-
183
\n-
185 bool IsEdgeSemiSharp(Index e) const { return _level->getEdgeTag(e)._semiSharp; }
\n-
186
\n-
188 bool IsVertexSemiSharp(Index v) const { return _level->getVertexTag(v)._semiSharp; }
\n-
189
\n-
191 bool IsFaceHole(Index f) const { return _level->isFaceHole(f); }
\n-
192
\n-
194 Sdc::Crease::Rule GetVertexRule(Index v) const { return _level->getVertexRule(v); }
\n-
196
\n-
198
\n-
228
\n-
230 int GetNumFVarChannels() const { return _level->getNumFVarChannels(); }
\n-
231
\n-
234 int GetNumFVarValues(int channel = 0) const { return _level->getNumFVarValues(channel); }
\n-
235
\n-
237 ConstIndexArray GetFaceFVarValues(Index f, int channel = 0) const {
\n-
238 return _level->getFaceFVarValues(f, channel);
\n-
239 }
\n-
240
\n-
242 bool DoesVertexFVarTopologyMatch(Index v, int channel = 0) const {
\n-
243 return _level->doesVertexFVarTopologyMatch(v, channel);
\n-
244 }
\n-
245
\n-
247 bool DoesEdgeFVarTopologyMatch(Index e, int channel = 0) const {
\n-
248 return _level->doesEdgeFVarTopologyMatch(e, channel);
\n-
249 }
\n-
250
\n-
252 bool DoesFaceFVarTopologyMatch(Index f, int channel = 0) const {
\n-
253 return _level->doesFaceFVarTopologyMatch(f, channel);
\n-
254 }
\n-
255
\n-
257
\n-
259
\n-
260
\n-
262 ConstIndexArray GetFaceChildFaces(Index f) const { return _refToChild->getFaceChildFaces(f); }
\n-
263
\n-
265 ConstIndexArray GetFaceChildEdges(Index f) const { return _refToChild->getFaceChildEdges(f); }
\n-
266
\n-
268 ConstIndexArray GetEdgeChildEdges(Index e) const { return _refToChild->getEdgeChildEdges(e); }
\n-
269
\n-
271 Index GetFaceChildVertex( Index f) const { return _refToChild->getFaceChildVertex(f); }
\n-
272
\n-
274 Index GetEdgeChildVertex( Index e) const { return _refToChild->getEdgeChildVertex(e); }
\n-
275
\n-
277 Index GetVertexChildVertex(Index v) const { return _refToChild->getVertexChildVertex(v); }
\n-
278
\n-
280 Index GetFaceParentFace(Index f) const { return _refToParent->getChildFaceParentFace(f); }
\n-
282
\n-
284
\n-
285
\n-
286 bool ValidateTopology() const { return _level->validateTopology(); }
\n-
287 void PrintTopology(bool children = true) const { _level->print((children && _refToChild) ? _refToChild : 0); }
\n-
289
\n-
290
\n-
291private:
\n-
292 friend class TopologyRefiner;
\n-
293
\n-
294 Vtr::internal::Level const * _level;
\n-
295 Vtr::internal::Refinement const * _refToParent;
\n-
296 Vtr::internal::Refinement const * _refToChild;
\n-
297
\n-
298public:
\n-
299 // Not intended for public use, but required by std::vector, etc...
\n-
300 TopologyLevel() { }
\n-
301 ~TopologyLevel() { }
\n-
302};
\n-
303
\n-
304} // end namespace Far
\n-
305
\n-
306} // end namespace OPENSUBDIV_VERSION
\n-
307using namespace OPENSUBDIV_VERSION;
\n-
308} // end namespace OpenSubdiv
\n-
309
\n-
310#endif /* OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H */
\n+
92 void initializePtexIndices(TopologyRefiner const &refiner);
\n+
93
\n+
94private:
\n+
95
\n+
96 std::vector<Index> _ptexIndices;
\n+
97};
\n+
98
\n+
99
\n+
100} // end namespace Far
\n+
101
\n+
102} // end namespace OPENSUBDIV_VERSION
\n+
103using namespace OPENSUBDIV_VERSION;
\n+
104} // end namespace OpenSubdiv
\n+
105
\n+
106#endif /* OPENSUBDIV3_FAR_PTEX_INDICES_H */
\n \n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel
An interface for accessing data in a specific level of a refined topology hierarchy.
Definition: topologyLevel.h:49
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceParentFace
Index GetFaceParentFace(Index f) const
Return the parent face (in the previous level) of a given face.
Definition: topologyLevel.h:280
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexValenceRegular
bool IsVertexValenceRegular(Index v) const
Return if the valence of the vertex is regular (must be manifold)
Definition: topologyLevel.h:162
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeSharpness
float GetEdgeSharpness(Index e) const
Return the sharpness assigned a given edge.
Definition: topologyLevel.h:173
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesFaceFVarTopologyMatch
bool DoesFaceFVarTopologyMatch(Index f, int channel=0) const
Return if face-varying topology around a face matches.
Definition: topologyLevel.h:252
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexRule
Sdc::Crease::Rule GetVertexRule(Index v) const
Return the subdivision rule assigned a given vertex specific to this level.
Definition: topologyLevel.h:194
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexCorner
bool IsVertexCorner(Index v) const
Return if the vertex is a corner (only one incident face)
Definition: topologyLevel.h:140
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexChildVertex
Index GetVertexChildVertex(Index v) const
Return the child vertex (in the next level) of a given vertex.
Definition: topologyLevel.h:277
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeFaceLocalIndices
ConstLocalIndexArray GetEdgeFaceLocalIndices(Index e) const
Access the local indices of an edge with respect to its incident faces.
Definition: topologyLevel.h:117
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildEdges
ConstIndexArray GetFaceChildEdges(Index f) const
Access the child edges (in the next level) of a given face.
Definition: topologyLevel.h:265
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesEdgeFVarTopologyMatch
bool DoesEdgeFVarTopologyMatch(Index e, int channel=0) const
Return if face-varying topology across the edge only matches.
Definition: topologyLevel.h:247
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::PrintTopology
void PrintTopology(bool children=true) const
Definition: topologyLevel.h:287
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceVertices
ConstIndexArray GetFaceVertices(Index f) const
Access the vertices incident a given face.
Definition: topologyLevel.h:93
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildFaces
ConstIndexArray GetFaceChildFaces(Index f) const
Access the child faces (in the next level) of a given face.
Definition: topologyLevel.h:262
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumEdges
int GetNumEdges() const
Return the number of edges in this level.
Definition: topologyLevel.h:68
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeNonManifold
bool IsEdgeNonManifold(Index e) const
Return if the edge is non-manifold.
Definition: topologyLevel.h:128
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsFaceHole
bool IsFaceHole(Index f) const
Return if a given face has been tagged as a hole.
Definition: topologyLevel.h:191
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexEdges
ConstIndexArray GetVertexEdges(Index v) const
Access the edges incident a given vertex.
Definition: topologyLevel.h:108
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildVertex
Index GetFaceChildVertex(Index f) const
Return the child vertex (in the next level) of a given face.
Definition: topologyLevel.h:271
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesVertexFVarTopologyMatch
bool DoesVertexFVarTopologyMatch(Index v, int channel=0) const
Return if face-varying topology around a vertex matches.
Definition: topologyLevel.h:242
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeChildEdges
ConstIndexArray GetEdgeChildEdges(Index e) const
Access the child edges (in the next level) of a given edge.
Definition: topologyLevel.h:268
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeInfSharp
bool IsEdgeInfSharp(Index e) const
Return if the edge is infinitely-sharp.
Definition: topologyLevel.h:179
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexFaces
ConstIndexArray GetVertexFaces(Index v) const
Access the faces incident a given vertex.
Definition: topologyLevel.h:105
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::FindEdge
Index FindEdge(Index v0, Index v1) const
Identify the edge matching the given vertex pair.
Definition: topologyLevel.h:120
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexEdgeLocalIndices
ConstLocalIndexArray GetVertexEdgeLocalIndices(Index v) const
Access the local indices of a vertex with respect to its incident edges.
Definition: topologyLevel.h:114
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceEdges
ConstIndexArray GetFaceEdges(Index f) const
Access the edges incident a given face.
Definition: topologyLevel.h:96
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFaces
int GetNumFaces() const
Return the number of faces in this level.
Definition: topologyLevel.h:65
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexBoundary
bool IsVertexBoundary(Index v) const
Return if the vertex is on a boundary (at least one incident boundary edge)
Definition: topologyLevel.h:137
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFVarChannels
int GetNumFVarChannels() const
Return the number of face-varying channels (should be same for all levels)
Definition: topologyLevel.h:230
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumVertices
int GetNumVertices() const
Return the number of vertices in this level.
Definition: topologyLevel.h:62
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeFaces
ConstIndexArray GetEdgeFaces(Index e) const
Access the faces incident a given edge.
Definition: topologyLevel.h:102
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceFVarValues
ConstIndexArray GetFaceFVarValues(Index f, int channel=0) const
Access the face-varying values associated with a particular face.
Definition: topologyLevel.h:237
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeVertices
ConstIndexArray GetEdgeVertices(Index e) const
Access the vertices incident a given edge.
Definition: topologyLevel.h:99
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeSemiSharp
bool IsEdgeSemiSharp(Index e) const
Return if the edge is semi-sharp.
Definition: topologyLevel.h:185
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexSharpness
float GetVertexSharpness(Index v) const
Return the sharpness assigned a given vertex.
Definition: topologyLevel.h:176
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeChildVertex
Index GetEdgeChildVertex(Index e) const
Return the child vertex (in the next level) of a given edge.
Definition: topologyLevel.h:274
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexSemiSharp
bool IsVertexSemiSharp(Index v) const
Return if the vertex is semi-sharp.
Definition: topologyLevel.h:188
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFaceVertices
int GetNumFaceVertices() const
Return the total number of face-vertices, i.e. the sum of all vertices for all faces.
Definition: topologyLevel.h:71
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexNonManifold
bool IsVertexNonManifold(Index v) const
Return if the vertex is non-manifold.
Definition: topologyLevel.h:131
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeBoundary
bool IsEdgeBoundary(Index e) const
Return if the edge is a boundary (only one incident face)
Definition: topologyLevel.h:134
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexInfSharp
bool IsVertexInfSharp(Index v) const
Return if the vertex is infinitely-sharp.
Definition: topologyLevel.h:182
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFVarValues
int GetNumFVarValues(int channel=0) const
Return the total number of face-varying values in a particular channel (the upper bound of a face-var...
Definition: topologyLevel.h:234
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexFaceLocalIndices
ConstLocalIndexArray GetVertexFaceLocalIndices(Index v) const
Access the local indices of a vertex with respect to its incident faces.
Definition: topologyLevel.h:111
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices
Object used to compute and query ptex face indices.
Definition: ptexIndices.h:46
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetFaceId
int GetFaceId(Index f) const
Returns the ptex face index given a coarse face 'f' or -1.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::PtexIndices
PtexIndices(TopologyRefiner const &refiner)
Constructor.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetAdjacency
void GetAdjacency(TopologyRefiner const &refiner, int face, int quadrant, int adjFaces[4], int adjEdges[4]) const
Returns ptex face adjacency information for a given coarse face.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetNumFaces
int GetNumFaces() const
Returns the number of ptex faces in the mesh.
\n+\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner
Stores topology data for a specified set of refinement options.
Definition: topologyRefiner.h:51
\n-\n-\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-topologyLevel.h\n+ptexIndices.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2015 DreamWorks Animation LLC.\n+ 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -29,387 +29,90 @@\n 17//\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n- 24#ifndef OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H\n- 25#define OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H\n+ 24#ifndef OPENSUBDIV3_FAR_PTEX_INDICES_H\n+ 25#define OPENSUBDIV3_FAR_PTEX_INDICES_H\n 26\n 27#include \"../version.h\"\n 28\n- 29#include \"../vtr/level.h\"\n- 30#include \"../vtr/refinement.h\"\n- 31#include \"../far/types.h\"\n- 32\n- 33#include \n- 34\n- 35namespace OpenSubdiv {\n- 36namespace OPENSUBDIV_VERSION {\n- 37\n- 38namespace Far {\n- 39\n-49class TopologyLevel {\n- 50\n- 51public:\n- 53\n+ 29#include \"../far/topologyRefiner.h\"\n+ 30#include \"../far/types.h\"\n+ 31\n+ 32#include \n+ 33\n+ 34namespace OpenSubdiv {\n+ 35namespace OPENSUBDIV_VERSION {\n+ 36\n+ 37namespace Far {\n+ 38\n+46class PtexIndices {\n+ 47\n+ 48public:\n+ 49\n+51 PtexIndices(TopologyRefiner const &refiner);\n+ 52\n+54 ~PtexIndices();\n+ 55\n+ 57\n 60\n-62 int GetNumVertices() const { return _level->getNumVertices(); }\n- 63\n-65 int GetNumFaces() const { return _level->getNumFaces(); }\n- 66\n-68 int GetNumEdges() const { return _level->getNumEdges(); }\n- 69\n-71 int GetNumFaceVertices() const { return _level->getNumFaceVerticesTotal(); }\n- 73\n- 75\n- 90 //\n+63 int GetNumFaces() const;\n+ 64\n+67 int GetFaceId(Index f) const;\n+ 68\n+83 void GetAdjacency(\n+ 84 TopologyRefiner const &refiner,\n+ 85 int face, int quadrant,\n+ 86 int adjFaces[4], int adjEdges[4]) const;\n+ 87\n+ 89\n+ 90private:\n 91\n-93 ConstIndexArray GetFaceVertices(Index f) const { return _level-\n->getFaceVertices(f); }\n- 94\n-96 ConstIndexArray GetFaceEdges(Index f) const { return _level->getFaceEdges\n-(f); }\n- 97\n-99 ConstIndexArray GetEdgeVertices(Index e) const { return _level-\n->getEdgeVertices(e); }\n- 100\n-102 ConstIndexArray GetEdgeFaces(Index e) const { return _level->getEdgeFaces\n-(e); }\n- 103\n-105 ConstIndexArray GetVertexFaces(Index v) const { return _level-\n->getVertexFaces(v); }\n- 106\n-108 ConstIndexArray GetVertexEdges(Index v) const { return _level-\n->getVertexEdges(v); }\n- 109\n-111 ConstLocalIndexArray GetVertexFaceLocalIndices(Index v) const { return\n-_level->getVertexFaceLocalIndices(v); }\n- 112\n-114 ConstLocalIndexArray GetVertexEdgeLocalIndices(Index v) const { return\n-_level->getVertexEdgeLocalIndices(v); }\n- 115\n-117 ConstLocalIndexArray GetEdgeFaceLocalIndices(Index e) const { return\n-_level->getEdgeFaceLocalIndices(e); }\n- 118\n-120 Index FindEdge(Index v0, Index v1) const { return _level->findEdge(v0, v1);\n-}\n- 122\n- 124\n- 126\n-128 bool IsEdgeNonManifold(Index e) const { return _level->isEdgeNonManifold\n-(e); }\n- 129\n-131 bool IsVertexNonManifold(Index v) const { return _level-\n->isVertexNonManifold(v); }\n- 132\n-134 bool IsEdgeBoundary(Index e) const { return _level->getEdgeTag\n-(e)._boundary; }\n- 135\n-137 bool IsVertexBoundary(Index v) const { return _level->getVertexTag\n-(v)._boundary; }\n- 138\n-140 bool IsVertexCorner(Index v) const { return (_level->getNumVertexFaces(v)\n-== 1); }\n- 141\n-162 bool IsVertexValenceRegular(Index v) const { return !_level->getVertexTag\n-(v)._xordinary || IsVertexCorner(v); }\n- 164\n- 166\n- 171\n-173 float GetEdgeSharpness(Index e) const { return _level->getEdgeSharpness(e);\n-}\n- 174\n-176 float GetVertexSharpness(Index v) const { return _level->getVertexSharpness\n-(v); }\n- 177\n-179 bool IsEdgeInfSharp(Index e) const { return _level->getEdgeTag\n-(e)._infSharp; }\n- 180\n-182 bool IsVertexInfSharp(Index v) const { return _level->getVertexTag\n-(v)._infSharp; }\n- 183\n-185 bool IsEdgeSemiSharp(Index e) const { return _level->getEdgeTag\n-(e)._semiSharp; }\n- 186\n-188 bool IsVertexSemiSharp(Index v) const { return _level->getVertexTag\n-(v)._semiSharp; }\n- 189\n-191 bool IsFaceHole(Index f) const { return _level->isFaceHole(f); }\n- 192\n-194 Sdc::Crease::Rule GetVertexRule(Index v) const { return _level-\n->getVertexRule(v); }\n- 196\n- 198\n- 228\n-230 int GetNumFVarChannels() const { return _level->getNumFVarChannels(); }\n- 231\n-234 int GetNumFVarValues(int channel = 0) const { return _level-\n->getNumFVarValues(channel); }\n- 235\n-237 ConstIndexArray GetFaceFVarValues(Index f, int channel = 0) const {\n- 238 return _level->getFaceFVarValues(f, channel);\n- 239 }\n- 240\n-242 bool DoesVertexFVarTopologyMatch(Index v, int channel = 0) const {\n- 243 return _level->doesVertexFVarTopologyMatch(v, channel);\n- 244 }\n- 245\n-247 bool DoesEdgeFVarTopologyMatch(Index e, int channel = 0) const {\n- 248 return _level->doesEdgeFVarTopologyMatch(e, channel);\n- 249 }\n- 250\n-252 bool DoesFaceFVarTopologyMatch(Index f, int channel = 0) const {\n- 253 return _level->doesFaceFVarTopologyMatch(f, channel);\n- 254 }\n- 255\n- 257\n- 259\n- 260\n-262 ConstIndexArray GetFaceChildFaces(Index f) const { return _refToChild-\n->getFaceChildFaces(f); }\n- 263\n-265 ConstIndexArray GetFaceChildEdges(Index f) const { return _refToChild-\n->getFaceChildEdges(f); }\n- 266\n-268 ConstIndexArray GetEdgeChildEdges(Index e) const { return _refToChild-\n->getEdgeChildEdges(e); }\n- 269\n-271 Index GetFaceChildVertex( Index f) const { return _refToChild-\n->getFaceChildVertex(f); }\n- 272\n-274 Index GetEdgeChildVertex( Index e) const { return _refToChild-\n->getEdgeChildVertex(e); }\n- 275\n-277 Index GetVertexChildVertex(Index v) const { return _refToChild-\n->getVertexChildVertex(v); }\n- 278\n-280 Index GetFaceParentFace(Index f) const { return _refToParent-\n->getChildFaceParentFace(f); }\n- 282\n- 284\n- 285\n-286 bool ValidateTopology() const { return _level->validateTopology(); }\n-287 void PrintTopology(bool children = true) const { _level->print((children &&\n-_refToChild) ? _refToChild : 0); }\n- 289\n- 290\n- 291private:\n-292 friend class TopologyRefiner;\n- 293\n- 294 Vtr::internal::Level const * _level;\n- 295 Vtr::internal::Refinement const * _refToParent;\n- 296 Vtr::internal::Refinement const * _refToChild;\n- 297\n- 298public:\n- 299 // Not intended for public use, but required by std::vector, etc...\n-300 TopologyLevel() { }\n-301 ~TopologyLevel() { }\n- 302};\n- 303\n- 304} // end namespace Far\n- 305\n- 306} // end namespace OPENSUBDIV_VERSION\n- 307using namespace OPENSUBDIV_VERSION;\n- 308} // end namespace OpenSubdiv\n- 309\n- 310#endif /* OPENSUBDIV3_FAR_TOPOLOGY_LEVEL_H */\n+ 92 void initializePtexIndices(TopologyRefiner const &refiner);\n+ 93\n+ 94private:\n+ 95\n+ 96 std::vector _ptexIndices;\n+ 97};\n+ 98\n+ 99\n+ 100} // end namespace Far\n+ 101\n+ 102} // end namespace OPENSUBDIV_VERSION\n+ 103using namespace OPENSUBDIV_VERSION;\n+ 104} // end namespace OpenSubdiv\n+ 105\n+ 106#endif /* OPENSUBDIV3_FAR_PTEX_INDICES_H */\n OpenSubdiv\n Definition: limits.h:32\n OpenSubdiv::OPENSUBDIV_VERSION::Far::Index\n Vtr::Index Index\n Definition: types.h:41\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel\n-An interface for accessing data in a specific level of a refined topology\n-hierarchy.\n-Definition: topologyLevel.h:49\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceParentFace\n-Index GetFaceParentFace(Index f) const\n-Return the parent face (in the previous level) of a given face.\n-Definition: topologyLevel.h:280\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::TopologyLevel\n-TopologyLevel()\n-Definition: topologyLevel.h:300\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexValenceRegular\n-bool IsVertexValenceRegular(Index v) const\n-Return if the valence of the vertex is regular (must be manifold)\n-Definition: topologyLevel.h:162\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeSharpness\n-float GetEdgeSharpness(Index e) const\n-Return the sharpness assigned a given edge.\n-Definition: topologyLevel.h:173\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesFaceFVarTopologyMatch\n-bool DoesFaceFVarTopologyMatch(Index f, int channel=0) const\n-Return if face-varying topology around a face matches.\n-Definition: topologyLevel.h:252\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexRule\n-Sdc::Crease::Rule GetVertexRule(Index v) const\n-Return the subdivision rule assigned a given vertex specific to this level.\n-Definition: topologyLevel.h:194\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexCorner\n-bool IsVertexCorner(Index v) const\n-Return if the vertex is a corner (only one incident face)\n-Definition: topologyLevel.h:140\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexChildVertex\n-Index GetVertexChildVertex(Index v) const\n-Return the child vertex (in the next level) of a given vertex.\n-Definition: topologyLevel.h:277\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeFaceLocalIndices\n-ConstLocalIndexArray GetEdgeFaceLocalIndices(Index e) const\n-Access the local indices of an edge with respect to its incident faces.\n-Definition: topologyLevel.h:117\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildEdges\n-ConstIndexArray GetFaceChildEdges(Index f) const\n-Access the child edges (in the next level) of a given face.\n-Definition: topologyLevel.h:265\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesEdgeFVarTopologyMatch\n-bool DoesEdgeFVarTopologyMatch(Index e, int channel=0) const\n-Return if face-varying topology across the edge only matches.\n-Definition: topologyLevel.h:247\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::PrintTopology\n-void PrintTopology(bool children=true) const\n-Definition: topologyLevel.h:287\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceVertices\n-ConstIndexArray GetFaceVertices(Index f) const\n-Access the vertices incident a given face.\n-Definition: topologyLevel.h:93\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildFaces\n-ConstIndexArray GetFaceChildFaces(Index f) const\n-Access the child faces (in the next level) of a given face.\n-Definition: topologyLevel.h:262\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumEdges\n-int GetNumEdges() const\n-Return the number of edges in this level.\n-Definition: topologyLevel.h:68\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::ValidateTopology\n-bool ValidateTopology() const\n-Definition: topologyLevel.h:286\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeNonManifold\n-bool IsEdgeNonManifold(Index e) const\n-Return if the edge is non-manifold.\n-Definition: topologyLevel.h:128\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsFaceHole\n-bool IsFaceHole(Index f) const\n-Return if a given face has been tagged as a hole.\n-Definition: topologyLevel.h:191\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexEdges\n-ConstIndexArray GetVertexEdges(Index v) const\n-Access the edges incident a given vertex.\n-Definition: topologyLevel.h:108\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceChildVertex\n-Index GetFaceChildVertex(Index f) const\n-Return the child vertex (in the next level) of a given face.\n-Definition: topologyLevel.h:271\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::DoesVertexFVarTopologyMatch\n-bool DoesVertexFVarTopologyMatch(Index v, int channel=0) const\n-Return if face-varying topology around a vertex matches.\n-Definition: topologyLevel.h:242\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeChildEdges\n-ConstIndexArray GetEdgeChildEdges(Index e) const\n-Access the child edges (in the next level) of a given edge.\n-Definition: topologyLevel.h:268\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeInfSharp\n-bool IsEdgeInfSharp(Index e) const\n-Return if the edge is infinitely-sharp.\n-Definition: topologyLevel.h:179\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexFaces\n-ConstIndexArray GetVertexFaces(Index v) const\n-Access the faces incident a given vertex.\n-Definition: topologyLevel.h:105\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::FindEdge\n-Index FindEdge(Index v0, Index v1) const\n-Identify the edge matching the given vertex pair.\n-Definition: topologyLevel.h:120\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexEdgeLocalIndices\n-ConstLocalIndexArray GetVertexEdgeLocalIndices(Index v) const\n-Access the local indices of a vertex with respect to its incident edges.\n-Definition: topologyLevel.h:114\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceEdges\n-ConstIndexArray GetFaceEdges(Index f) const\n-Access the edges incident a given face.\n-Definition: topologyLevel.h:96\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFaces\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices\n+Object used to compute and query ptex face indices.\n+Definition: ptexIndices.h:46\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetFaceId\n+int GetFaceId(Index f) const\n+Returns the ptex face index given a coarse face 'f' or -1.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::PtexIndices\n+PtexIndices(TopologyRefiner const &refiner)\n+Constructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetAdjacency\n+void GetAdjacency(TopologyRefiner const &refiner, int face, int quadrant, int\n+adjFaces[4], int adjEdges[4]) const\n+Returns ptex face adjacency information for a given coarse face.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::GetNumFaces\n int GetNumFaces() const\n-Return the number of faces in this level.\n-Definition: topologyLevel.h:65\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexBoundary\n-bool IsVertexBoundary(Index v) const\n-Return if the vertex is on a boundary (at least one incident boundary edge)\n-Definition: topologyLevel.h:137\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFVarChannels\n-int GetNumFVarChannels() const\n-Return the number of face-varying channels (should be same for all levels)\n-Definition: topologyLevel.h:230\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumVertices\n-int GetNumVertices() const\n-Return the number of vertices in this level.\n-Definition: topologyLevel.h:62\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeFaces\n-ConstIndexArray GetEdgeFaces(Index e) const\n-Access the faces incident a given edge.\n-Definition: topologyLevel.h:102\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetFaceFVarValues\n-ConstIndexArray GetFaceFVarValues(Index f, int channel=0) const\n-Access the face-varying values associated with a particular face.\n-Definition: topologyLevel.h:237\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeVertices\n-ConstIndexArray GetEdgeVertices(Index e) const\n-Access the vertices incident a given edge.\n-Definition: topologyLevel.h:99\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeSemiSharp\n-bool IsEdgeSemiSharp(Index e) const\n-Return if the edge is semi-sharp.\n-Definition: topologyLevel.h:185\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::~TopologyLevel\n-~TopologyLevel()\n-Definition: topologyLevel.h:301\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexSharpness\n-float GetVertexSharpness(Index v) const\n-Return the sharpness assigned a given vertex.\n-Definition: topologyLevel.h:176\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetEdgeChildVertex\n-Index GetEdgeChildVertex(Index e) const\n-Return the child vertex (in the next level) of a given edge.\n-Definition: topologyLevel.h:274\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexSemiSharp\n-bool IsVertexSemiSharp(Index v) const\n-Return if the vertex is semi-sharp.\n-Definition: topologyLevel.h:188\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFaceVertices\n-int GetNumFaceVertices() const\n-Return the total number of face-vertices, i.e. the sum of all vertices for all\n-faces.\n-Definition: topologyLevel.h:71\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexNonManifold\n-bool IsVertexNonManifold(Index v) const\n-Return if the vertex is non-manifold.\n-Definition: topologyLevel.h:131\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsEdgeBoundary\n-bool IsEdgeBoundary(Index e) const\n-Return if the edge is a boundary (only one incident face)\n-Definition: topologyLevel.h:134\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::IsVertexInfSharp\n-bool IsVertexInfSharp(Index v) const\n-Return if the vertex is infinitely-sharp.\n-Definition: topologyLevel.h:182\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumFVarValues\n-int GetNumFVarValues(int channel=0) const\n-Return the total number of face-varying values in a particular channel (the\n-upper bound of a face-var...\n-Definition: topologyLevel.h:234\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetVertexFaceLocalIndices\n-ConstLocalIndexArray GetVertexFaceLocalIndices(Index v) const\n-Access the local indices of a vertex with respect to its incident faces.\n-Definition: topologyLevel.h:111\n+Returns the number of ptex faces in the mesh.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::~PtexIndices\n+~PtexIndices()\n+Destructor.\n OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner\n Stores topology data for a specified set of refinement options.\n Definition: topologyRefiner.h:51\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::Rule\n-Rule\n-Definition: crease.h:82\n-OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstArray\n-Definition: array.h:53\n * opensubdiv\n * far\n- * topologyLevel.h\n+ * ptexIndices.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00725.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00725.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/stencilTable.h File Reference\n+OpenSubdiv: opensubdiv/far/primvarRefiner.h File Reference\n \n \n \n \n \n \n \n@@ -83,51 +83,43 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
stencilTable.h File Reference
\n+
primvarRefiner.h File Reference
\n
\n
\n
#include "../version.h"
\n+#include "../sdc/types.h"
\n+#include "../sdc/options.h"
\n+#include "../sdc/bilinearScheme.h"
\n+#include "../sdc/catmarkScheme.h"
\n+#include "../sdc/loopScheme.h"
\n+#include "../vtr/level.h"
\n+#include "../vtr/fvarLevel.h"
\n+#include "../vtr/refinement.h"
\n+#include "../vtr/fvarRefinement.h"
\n+#include "../vtr/stackBuffer.h"
\n+#include "../vtr/componentInterfaces.h"
\n #include "../far/types.h"
\n+#include "../far/error.h"
\n+#include "../far/topologyLevel.h"
\n+#include "../far/topologyRefiner.h"
\n #include <cassert>
\n-#include <cstring>
\n-#include <vector>
\n-#include <iostream>
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n+\n+\n \n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n+\n \n

\n Classes

class  StencilReal< REAL >
 Vertex stencil descriptor. More...
class  PrimvarRefinerReal< REAL >
 Applies refinement operations to generic primvar data. More...
 
class  Stencil
 Vertex stencil class wrapping the template for compatibility. More...
 
class  StencilTableReal< REAL >
 Table of subdivision stencils. More...
 
class  StencilTable
 Stencil table class wrapping the template for compatibility. More...
 
class  LimitStencilReal< REAL >
 Limit point stencil descriptor. More...
 
class  LimitStencil
 Limit point stencil class wrapping the template for compatibility. More...
 
class  LimitStencilTableReal< REAL >
 Table of limit subdivision stencils. More...
 
class  LimitStencilTable
 Limit stencil table class wrapping the template for compatibility. More...
class  PrimvarRefiner
 
\n \n \n \n \n@@ -136,13 +128,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,53 +5,43 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-stencilTable.h File Reference\n+primvarRefiner.h File Reference\n #include \"../version.h\"\n+#include \"../sdc/types.h\"\n+#include \"../sdc/options.h\"\n+#include \"../sdc/bilinearScheme.h\"\n+#include \"../sdc/catmarkScheme.h\"\n+#include \"../sdc/loopScheme.h\"\n+#include \"../vtr/level.h\"\n+#include \"../vtr/fvarLevel.h\"\n+#include \"../vtr/refinement.h\"\n+#include \"../vtr/fvarRefinement.h\"\n+#include \"../vtr/stackBuffer.h\"\n+#include \"../vtr/componentInterfaces.h\"\n #include \"../far/types.h\"\n+#include \"../far/error.h\"\n+#include \"../far/topologyLevel.h\"\n+#include \"../far/topologyRefiner.h\"\n #include \n-#include \n-#include \n-#include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0StencilReal<_REAL_>\n-\u00a0 Vertex stencil descriptor. More...\n+class \u00a0PrimvarRefinerReal<_REAL_>\n+\u00a0 Applies refinement operations to generic primvar data. More...\n \u00a0\n-class \u00a0Stencil\n-\u00a0 Vertex stencil class wrapping the template for compatibility. More...\n-\u00a0\n-class \u00a0StencilTableReal<_REAL_>\n-\u00a0 Table of subdivision stencils. More...\n-\u00a0\n-class \u00a0StencilTable\n-\u00a0 Stencil table class wrapping the template for compatibility. More...\n-\u00a0\n-class \u00a0LimitStencilReal<_REAL_>\n-\u00a0 Limit point stencil descriptor. More...\n-\u00a0\n-class \u00a0LimitStencil\n-\u00a0 Limit point stencil class wrapping the template for compatibility.\n- More...\n-\u00a0\n-class \u00a0LimitStencilTableReal<_REAL_>\n-\u00a0 Table of limit subdivision stencils. More...\n-\u00a0\n-class \u00a0LimitStencilTable\n-\u00a0 Limit stencil table class wrapping the template for compatibility.\n- More...\n+class \u00a0PrimvarRefiner\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n * opensubdiv\n * far\n- * stencilTable.h\n+ * primvarRefiner.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00725.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00725.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,10 +1,4 @@\n var a00725 = [\n- [\"StencilReal< REAL >\", \"a01045.html\", \"a01045\"],\n- [\"Stencil\", \"a01049.html\", \"a01049\"],\n- [\"StencilTableReal< REAL >\", \"a01053.html\", \"a01053\"],\n- [\"StencilTable\", \"a01057.html\", \"a01057\"],\n- [\"LimitStencilReal< REAL >\", \"a01061.html\", \"a01061\"],\n- [\"LimitStencil\", \"a01065.html\", \"a01065\"],\n- [\"LimitStencilTableReal< REAL >\", \"a01069.html\", \"a01069\"],\n- [\"LimitStencilTable\", \"a01073.html\", \"a01073\"]\n+ [\"PrimvarRefinerReal< REAL >\", \"a01021.html\", \"a01021\"],\n+ [\"PrimvarRefiner\", \"a01029.html\", \"a01029\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00725_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00725_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/stencilTable.h Source File\n+OpenSubdiv: opensubdiv/far/primvarRefiner.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
stencilTable.h
\n+
primvarRefiner.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2013 Pixar
\n+
2// Copyright 2015 DreamWorks Animation LLC.
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -106,747 +106,1200 @@\n
17//
\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n-
24
\n-
25#ifndef OPENSUBDIV3_FAR_STENCILTABLE_H
\n-
26#define OPENSUBDIV3_FAR_STENCILTABLE_H
\n-
27
\n-
28#include "../version.h"
\n-
29
\n-
30#include "../far/types.h"
\n-
31
\n-
32#include <cassert>
\n-
33#include <cstring>
\n-
34#include <vector>
\n-
35#include <iostream>
\n-
36
\n-
37namespace OpenSubdiv {
\n-
38namespace OPENSUBDIV_VERSION {
\n-
39
\n-
40namespace Far {
\n-
41
\n-
42// Forward declarations for friends:
\n-
43class PatchTableBuilder;
\n+
24#ifndef OPENSUBDIV3_FAR_PRIMVAR_REFINER_H
\n+
25#define OPENSUBDIV3_FAR_PRIMVAR_REFINER_H
\n+
26
\n+
27#include "../version.h"
\n+
28
\n+
29#include "../sdc/types.h"
\n+
30#include "../sdc/options.h"
\n+
31#include "../sdc/bilinearScheme.h"
\n+
32#include "../sdc/catmarkScheme.h"
\n+
33#include "../sdc/loopScheme.h"
\n+
34#include "../vtr/level.h"
\n+
35#include "../vtr/fvarLevel.h"
\n+
36#include "../vtr/refinement.h"
\n+
37#include "../vtr/fvarRefinement.h"
\n+
38#include "../vtr/stackBuffer.h"
\n+
39#include "../vtr/componentInterfaces.h"
\n+
40#include "../far/types.h"
\n+
41#include "../far/error.h"
\n+
42#include "../far/topologyLevel.h"
\n+
43#include "../far/topologyRefiner.h"
\n
44
\n-
45template <typename REAL> class StencilTableFactoryReal;
\n-
46template <typename REAL> class LimitStencilTableFactoryReal;
\n-
47
\n-
52template <typename REAL>
\n-
53class StencilReal {
\n-
54public:
\n-
55
\n-
57 StencilReal() {}
\n-
58
\n-
67 StencilReal(int * size, Index * indices, REAL * weights)
\n-
68 : _size(size), _indices(indices), _weights(weights) { }
\n-
69
\n-
71 StencilReal(StencilReal const & other) {
\n-
72 _size = other._size;
\n-
73 _indices = other._indices;
\n-
74 _weights = other._weights;
\n-
75 }
\n-
76
\n-
78 int GetSize() const {
\n-
79 return *_size;
\n-
80 }
\n-
81
\n-
83 int * GetSizePtr() const {
\n-
84 return _size;
\n-
85 }
\n-
86
\n-
88 Index const * GetVertexIndices() const {
\n-
89 return _indices;
\n-
90 }
\n-
91
\n-
93 REAL const * GetWeights() const {
\n-
94 return _weights;
\n-
95 }
\n-
96
\n-
98 void Next() {
\n-
99 int stride = *_size;
\n-
100 ++_size;
\n-
101 _indices += stride;
\n-
102 _weights += stride;
\n-
103 }
\n-
104
\n-
105protected:
\n-
106 friend class StencilTableFactoryReal<REAL>;
\n-
107 friend class LimitStencilTableFactoryReal<REAL>;
\n-
108
\n-
109 int * _size;
\n-
110 Index * _indices;
\n-
111 REAL * _weights;
\n-
112};
\n-
113
\n-
116class Stencil : public StencilReal<float> {
\n-
117protected:
\n-
118 typedef StencilReal<float> BaseStencil;
\n-
119
\n-
120public:
\n-
121 Stencil() : BaseStencil() { }
\n-
122 Stencil(BaseStencil const & other) : BaseStencil(other) { }
\n-
123 Stencil(int * size, Index * indices, float * weights)
\n-
124 : BaseStencil(size, indices, weights) { }
\n-
125};
\n-
126
\n-
127
\n-
140template <typename REAL>
\n-
141class StencilTableReal {
\n-
142protected:
\n-
143 StencilTableReal(int numControlVerts,
\n-
144 std::vector<int> const& offsets,
\n-
145 std::vector<int> const& sizes,
\n-
146 std::vector<int> const& sources,
\n-
147 std::vector<REAL> const& weights,
\n-
148 bool includeCoarseVerts,
\n-
149 size_t firstOffset);
\n-
150
\n-
151public:
\n-
152
\n-
153 virtual ~StencilTableReal() {};
\n-
154
\n-
156 int GetNumStencils() const {
\n-
157 return (int)_sizes.size();
\n-
158 }
\n-
159
\n-
161 int GetNumControlVertices() const {
\n-
162 return _numControlVertices;
\n-
163 }
\n-
164
\n-
166 StencilReal<REAL> GetStencil(Index i) const;
\n-
167
\n-
169 std::vector<int> const & GetSizes() const {
\n-
170 return _sizes;
\n-
171 }
\n-
172
\n-
174 std::vector<Index> const & GetOffsets() const {
\n-
175 return _offsets;
\n-
176 }
\n-
177
\n-
179 std::vector<Index> const & GetControlIndices() const {
\n-
180 return _indices;
\n-
181 }
\n-
182
\n-
184 std::vector<REAL> const & GetWeights() const {
\n-
185 return _weights;
\n-
186 }
\n-
187
\n-
189 StencilReal<REAL> operator[] (Index index) const;
\n+
45#include <cassert>
\n+
46
\n+
47namespace OpenSubdiv {
\n+
48namespace OPENSUBDIV_VERSION {
\n+
49
\n+
50namespace Far {
\n+
51
\n+
55template <typename REAL>
\n+
56class PrimvarRefinerReal {
\n+
57
\n+
58public:
\n+
59 PrimvarRefinerReal(TopologyRefiner const & refiner) : _refiner(refiner) { }
\n+
60 ~PrimvarRefinerReal() { }
\n+
61
\n+
62 TopologyRefiner const & GetTopologyRefiner() const { return _refiner; }
\n+
63
\n+
65
\n+
100
\n+
113 template <class T, class U> void Interpolate(int level, T const & src, U & dst) const;
\n+
114
\n+
130 template <class T, class U> void InterpolateVarying(int level, T const & src, U & dst) const;
\n+
131
\n+
148 template <class T, class U> void InterpolateFaceUniform(int level, T const & src, U & dst) const;
\n+
149
\n+
161 template <class T, class U> void InterpolateFaceVarying(int level, T const & src, U & dst, int channel = 0) const;
\n+
162
\n+
163
\n+
175 template <class T, class U> void Limit(T const & src, U & dstPos) const;
\n+
176
\n+
177 template <class T, class U, class U1, class U2>
\n+
178 void Limit(T const & src, U & dstPos, U1 & dstTan1, U2 & dstTan2) const;
\n+
179
\n+
180 template <class T, class U> void LimitFaceVarying(T const & src, U & dst, int channel = 0) const;
\n+
181
\n+
183
\n+
184private:
\n+
185 typedef REAL Weight;
\n+
186
\n+
187 // Non-copyable:
\n+
188 PrimvarRefinerReal(PrimvarRefinerReal const & src) : _refiner(src._refiner) { }
\n+
189 PrimvarRefinerReal & operator=(PrimvarRefinerReal const &) { return *this; }
\n
190
\n-
204 template <class T, class U>
\n-
205 void UpdateValues(T const &srcValues, U &dstValues, Index start=-1, Index end=-1) const {
\n-
206 this->update(srcValues, dstValues, _weights, start, end);
\n-
207 }
\n-
208
\n-
209 template <class T1, class T2, class U>
\n-
210 void UpdateValues(T1 const &srcBase, int numBase, T2 const &srcRef,
\n-
211 U &dstValues, Index start=-1, Index end=-1) const {
\n-
212 this->update(srcBase, numBase, srcRef, dstValues, _weights, start, end);
\n-
213 }
\n-
214
\n-
215 // Pointer interface for backward compatibility
\n-
216 template <class T, class U>
\n-
217 void UpdateValues(T const *src, U *dst, Index start=-1, Index end=-1) const {
\n-
218 this->update(src, dst, _weights, start, end);
\n-
219 }
\n-
220 template <class T1, class T2, class U>
\n-
221 void UpdateValues(T1 const *srcBase, int numBase, T2 const *srcRef,
\n-
222 U *dst, Index start=-1, Index end=-1) const {
\n-
223 this->update(srcBase, numBase, srcRef, dst, _weights, start, end);
\n-
224 }
\n-
225
\n-
227 void Clear();
\n-
228
\n-
229protected:
\n-
230
\n-
231 // Update values by applying cached stencil weights to new control values
\n-
232 template <class T, class U>
\n-
233 void update( T const &srcValues, U &dstValues,
\n-
234 std::vector<REAL> const & valueWeights, Index start, Index end) const;
\n-
235 template <class T1, class T2, class U>
\n-
236 void update( T1 const &srcBase, int numBase, T2 const &srcRef, U &dstValues,
\n-
237 std::vector<REAL> const & valueWeights, Index start, Index end) const;
\n-
238
\n-
239 // Populate the offsets table from the stencil sizes in _sizes (factory helper)
\n-
240 void generateOffsets();
\n+
191 template <Sdc::SchemeType SCHEME, class T, class U> void interpFromFaces(int, T const &, U &) const;
\n+
192 template <Sdc::SchemeType SCHEME, class T, class U> void interpFromEdges(int, T const &, U &) const;
\n+
193 template <Sdc::SchemeType SCHEME, class T, class U> void interpFromVerts(int, T const &, U &) const;
\n+
194
\n+
195 template <Sdc::SchemeType SCHEME, class T, class U> void interpFVarFromFaces(int, T const &, U &, int) const;
\n+
196 template <Sdc::SchemeType SCHEME, class T, class U> void interpFVarFromEdges(int, T const &, U &, int) const;
\n+
197 template <Sdc::SchemeType SCHEME, class T, class U> void interpFVarFromVerts(int, T const &, U &, int) const;
\n+
198
\n+
199 template <Sdc::SchemeType SCHEME, class T, class U, class U1, class U2>
\n+
200 void limit(T const & src, U & pos, U1 * tan1, U2 * tan2) const;
\n+
201
\n+
202 template <Sdc::SchemeType SCHEME, class T, class U>
\n+
203 void limitFVar(T const & src, U & dst, int channel) const;
\n+
204
\n+
205private:
\n+
206 TopologyRefiner const & _refiner;
\n+
207
\n+
208private:
\n+
209 //
\n+
210 // Local class to fulfill interface for <typename MASK> in the Scheme mask queries:
\n+
211 //
\n+
212 class Mask {
\n+
213 public:
\n+
214 typedef REAL Weight; // Also part of the expected interface
\n+
215
\n+
216 public:
\n+
217 Mask(Weight* v, Weight* e, Weight* f) :
\n+
218 _vertWeights(v), _edgeWeights(e), _faceWeights(f),
\n+
219 _vertCount(0), _edgeCount(0), _faceCount(0),
\n+
220 _faceWeightsForFaceCenters(false)
\n+
221 { }
\n+
222
\n+
223 ~Mask() { }
\n+
224
\n+
225 public: // Generic interface expected of <typename MASK>:
\n+
226 int GetNumVertexWeights() const { return _vertCount; }
\n+
227 int GetNumEdgeWeights() const { return _edgeCount; }
\n+
228 int GetNumFaceWeights() const { return _faceCount; }
\n+
229
\n+
230 void SetNumVertexWeights(int count) { _vertCount = count; }
\n+
231 void SetNumEdgeWeights( int count) { _edgeCount = count; }
\n+
232 void SetNumFaceWeights( int count) { _faceCount = count; }
\n+
233
\n+
234 Weight const& VertexWeight(int index) const { return _vertWeights[index]; }
\n+
235 Weight const& EdgeWeight( int index) const { return _edgeWeights[index]; }
\n+
236 Weight const& FaceWeight( int index) const { return _faceWeights[index]; }
\n+
237
\n+
238 Weight& VertexWeight(int index) { return _vertWeights[index]; }
\n+
239 Weight& EdgeWeight( int index) { return _edgeWeights[index]; }
\n+
240 Weight& FaceWeight( int index) { return _faceWeights[index]; }
\n
241
\n-
242 // Resize the table arrays (factory helper)
\n-
243 void resize(int nstencils, int nelems);
\n+
242 bool AreFaceWeightsForFaceCenters() const { return _faceWeightsForFaceCenters; }
\n+
243 void SetFaceWeightsForFaceCenters(bool on) { _faceWeightsForFaceCenters = on; }
\n
244
\n-
245 // Reserves the table arrays (factory helper)
\n-
246 void reserve(int nstencils, int nelems);
\n-
247
\n-
248 // Reallocates the table arrays to remove excess capacity (factory helper)
\n-
249 void shrinkToFit();
\n-
250
\n-
251 // Performs any final operations on internal tables (factory helper)
\n-
252 void finalize();
\n+
245 private:
\n+
246 Weight* _vertWeights;
\n+
247 Weight* _edgeWeights;
\n+
248 Weight* _faceWeights;
\n+
249
\n+
250 int _vertCount;
\n+
251 int _edgeCount;
\n+
252 int _faceCount;
\n
253
\n-
254protected:
\n-
255 StencilTableReal() : _numControlVertices(0) {}
\n-
256 StencilTableReal(int numControlVerts)
\n-
257 : _numControlVertices(numControlVerts)
\n-
258 { }
\n-
259
\n-
260 friend class StencilTableFactoryReal<REAL>;
\n-
261 friend class Far::PatchTableBuilder;
\n-
262
\n-
263 int _numControlVertices; // number of control vertices
\n-
264
\n-
265 std::vector<int> _sizes; // number of coefficients for each stencil
\n-
266 std::vector<Index> _offsets, // offset to the start of each stencil
\n-
267 _indices; // indices of contributing coarse vertices
\n-
268 std::vector<REAL> _weights; // stencil weight coefficients
\n-
269};
\n+
254 bool _faceWeightsForFaceCenters;
\n+
255 };
\n+
256};
\n+
257
\n+
258
\n+
259//
\n+
260// Public entry points to the methods. Queries of the scheme type and its
\n+
261// use as a template parameter in subsequent implementation will be factored
\n+
262// out of a later release:
\n+
263//
\n+
264template <typename REAL>
\n+
265template <class T, class U>
\n+
266inline void
\n+
267PrimvarRefinerReal<REAL>::Interpolate(int level, T const & src, U & dst) const {
\n+
268
\n+
269 assert(level>0 && level<=(int)_refiner._refinements.size());
\n
270
\n-
273class StencilTable : public StencilTableReal<float> {
\n-
274protected:
\n-
275 typedef StencilTableReal<float> BaseTable;
\n-
276
\n-
277public:
\n-
278 Stencil GetStencil(Index index) const {
\n-
279 return Stencil(BaseTable::GetStencil(index));
\n-
280 }
\n-
281 Stencil operator[] (Index index) const {
\n-
282 return Stencil(BaseTable::GetStencil(index));
\n-
283 }
\n-
284
\n-
285protected:
\n-
286 StencilTable() : BaseTable() { }
\n-
287 StencilTable(int numControlVerts) : BaseTable(numControlVerts) { }
\n-
288 StencilTable(int numControlVerts,
\n-
289 std::vector<int> const& offsets,
\n-
290 std::vector<int> const& sizes,
\n-
291 std::vector<int> const& sources,
\n-
292 std::vector<float> const& weights,
\n-
293 bool includeCoarseVerts,
\n-
294 size_t firstOffset)
\n-
295 : BaseTable(numControlVerts, offsets,
\n-
296 sizes, sources, weights, includeCoarseVerts, firstOffset) { }
\n-
297};
\n-
298
\n-
299
\n-
302template <typename REAL>
\n-
303class LimitStencilReal : public StencilReal<REAL> {
\n-
304public:
\n-
305
\n-
324 LimitStencilReal( int* size,
\n-
325 Index * indices,
\n-
326 REAL * weights,
\n-
327 REAL * duWeights=0,
\n-
328 REAL * dvWeights=0,
\n-
329 REAL * duuWeights=0,
\n-
330 REAL * duvWeights=0,
\n-
331 REAL * dvvWeights=0)
\n-
332 : StencilReal<REAL>(size, indices, weights),
\n-
333 _duWeights(duWeights),
\n-
334 _dvWeights(dvWeights),
\n-
335 _duuWeights(duuWeights),
\n-
336 _duvWeights(duvWeights),
\n-
337 _dvvWeights(dvvWeights) {
\n+
271 switch (_refiner._subdivType) {
\n+
272 case Sdc::SCHEME_CATMARK:
\n+
273 interpFromFaces<Sdc::SCHEME_CATMARK>(level, src, dst);
\n+
274 interpFromEdges<Sdc::SCHEME_CATMARK>(level, src, dst);
\n+
275 interpFromVerts<Sdc::SCHEME_CATMARK>(level, src, dst);
\n+
276 break;
\n+
277 case Sdc::SCHEME_LOOP:
\n+
278 interpFromFaces<Sdc::SCHEME_LOOP>(level, src, dst);
\n+
279 interpFromEdges<Sdc::SCHEME_LOOP>(level, src, dst);
\n+
280 interpFromVerts<Sdc::SCHEME_LOOP>(level, src, dst);
\n+
281 break;
\n+
282 case Sdc::SCHEME_BILINEAR:
\n+
283 interpFromFaces<Sdc::SCHEME_BILINEAR>(level, src, dst);
\n+
284 interpFromEdges<Sdc::SCHEME_BILINEAR>(level, src, dst);
\n+
285 interpFromVerts<Sdc::SCHEME_BILINEAR>(level, src, dst);
\n+
286 break;
\n+
287 }
\n+
288}
\n+
289
\n+
290template <typename REAL>
\n+
291template <class T, class U>
\n+
292inline void
\n+
293PrimvarRefinerReal<REAL>::InterpolateFaceVarying(int level, T const & src, U & dst, int channel) const {
\n+
294
\n+
295 assert(level>0 && level<=(int)_refiner._refinements.size());
\n+
296
\n+
297 switch (_refiner._subdivType) {
\n+
298 case Sdc::SCHEME_CATMARK:
\n+
299 interpFVarFromFaces<Sdc::SCHEME_CATMARK>(level, src, dst, channel);
\n+
300 interpFVarFromEdges<Sdc::SCHEME_CATMARK>(level, src, dst, channel);
\n+
301 interpFVarFromVerts<Sdc::SCHEME_CATMARK>(level, src, dst, channel);
\n+
302 break;
\n+
303 case Sdc::SCHEME_LOOP:
\n+
304 interpFVarFromFaces<Sdc::SCHEME_LOOP>(level, src, dst, channel);
\n+
305 interpFVarFromEdges<Sdc::SCHEME_LOOP>(level, src, dst, channel);
\n+
306 interpFVarFromVerts<Sdc::SCHEME_LOOP>(level, src, dst, channel);
\n+
307 break;
\n+
308 case Sdc::SCHEME_BILINEAR:
\n+
309 interpFVarFromFaces<Sdc::SCHEME_BILINEAR>(level, src, dst, channel);
\n+
310 interpFVarFromEdges<Sdc::SCHEME_BILINEAR>(level, src, dst, channel);
\n+
311 interpFVarFromVerts<Sdc::SCHEME_BILINEAR>(level, src, dst, channel);
\n+
312 break;
\n+
313 }
\n+
314}
\n+
315
\n+
316template <typename REAL>
\n+
317template <class T, class U>
\n+
318inline void
\n+
319PrimvarRefinerReal<REAL>::Limit(T const & src, U & dst) const {
\n+
320
\n+
321 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() == 0) {
\n+
322 Error(FAR_RUNTIME_ERROR,
\n+
323 "Failure in PrimvarRefiner::Limit() -- "
\n+
324 "last level of refinement does not include full topology.");
\n+
325 return;
\n+
326 }
\n+
327
\n+
328 switch (_refiner._subdivType) {
\n+
329 case Sdc::SCHEME_CATMARK:
\n+
330 limit<Sdc::SCHEME_CATMARK>(src, dst, (U*)0, (U*)0);
\n+
331 break;
\n+
332 case Sdc::SCHEME_LOOP:
\n+
333 limit<Sdc::SCHEME_LOOP>(src, dst, (U*)0, (U*)0);
\n+
334 break;
\n+
335 case Sdc::SCHEME_BILINEAR:
\n+
336 limit<Sdc::SCHEME_BILINEAR>(src, dst, (U*)0, (U*)0);
\n+
337 break;
\n
338 }
\n-
339
\n-
341 REAL const * GetDuWeights() const {
\n-
342 return _duWeights;
\n-
343 }
\n-
344
\n-
346 REAL const * GetDvWeights() const {
\n-
347 return _dvWeights;
\n-
348 }
\n-
349
\n-
351 REAL const * GetDuuWeights() const {
\n-
352 return _duuWeights;
\n-
353 }
\n-
354
\n-
356 REAL const * GetDuvWeights() const {
\n-
357 return _duvWeights;
\n-
358 }
\n-
359
\n-
361 REAL const * GetDvvWeights() const {
\n-
362 return _dvvWeights;
\n+
339}
\n+
340
\n+
341template <typename REAL>
\n+
342template <class T, class U, class U1, class U2>
\n+
343inline void
\n+
344PrimvarRefinerReal<REAL>::Limit(T const & src, U & dstPos, U1 & dstTan1, U2 & dstTan2) const {
\n+
345
\n+
346 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() == 0) {
\n+
347 Error(FAR_RUNTIME_ERROR,
\n+
348 "Failure in PrimvarRefiner::Limit() -- "
\n+
349 "last level of refinement does not include full topology.");
\n+
350 return;
\n+
351 }
\n+
352
\n+
353 switch (_refiner._subdivType) {
\n+
354 case Sdc::SCHEME_CATMARK:
\n+
355 limit<Sdc::SCHEME_CATMARK>(src, dstPos, &dstTan1, &dstTan2);
\n+
356 break;
\n+
357 case Sdc::SCHEME_LOOP:
\n+
358 limit<Sdc::SCHEME_LOOP>(src, dstPos, &dstTan1, &dstTan2);
\n+
359 break;
\n+
360 case Sdc::SCHEME_BILINEAR:
\n+
361 limit<Sdc::SCHEME_BILINEAR>(src, dstPos, &dstTan1, &dstTan2);
\n+
362 break;
\n
363 }
\n-
364
\n-
366 void Next() {
\n-
367 int stride = *this->_size;
\n-
368 ++this->_size;
\n-
369 this->_indices += stride;
\n-
370 this->_weights += stride;
\n-
371 if (_duWeights) _duWeights += stride;
\n-
372 if (_dvWeights) _dvWeights += stride;
\n-
373 if (_duuWeights) _duuWeights += stride;
\n-
374 if (_duvWeights) _duvWeights += stride;
\n-
375 if (_dvvWeights) _dvvWeights += stride;
\n+
364}
\n+
365
\n+
366template <typename REAL>
\n+
367template <class T, class U>
\n+
368inline void
\n+
369PrimvarRefinerReal<REAL>::LimitFaceVarying(T const & src, U & dst, int channel) const {
\n+
370
\n+
371 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() == 0) {
\n+
372 Error(FAR_RUNTIME_ERROR,
\n+
373 "Failure in PrimvarRefiner::LimitFaceVarying() -- "
\n+
374 "last level of refinement does not include full topology.");
\n+
375 return;
\n
376 }
\n
377
\n-
378private:
\n-
379
\n-
380 friend class StencilTableFactoryReal<REAL>;
\n-
381 friend class LimitStencilTableFactoryReal<REAL>;
\n-
382
\n-
383 REAL * _duWeights, // pointer to stencil u derivative limit weights
\n-
384 * _dvWeights, // pointer to stencil v derivative limit weights
\n-
385 * _duuWeights, // pointer to stencil uu derivative limit weights
\n-
386 * _duvWeights, // pointer to stencil uv derivative limit weights
\n-
387 * _dvvWeights; // pointer to stencil vv derivative limit weights
\n-
388};
\n-
389
\n-
392class LimitStencil : public LimitStencilReal<float> {
\n-
393protected:
\n-
394 typedef LimitStencilReal<float> BaseStencil;
\n+
378 switch (_refiner._subdivType) {
\n+
379 case Sdc::SCHEME_CATMARK:
\n+
380 limitFVar<Sdc::SCHEME_CATMARK>(src, dst, channel);
\n+
381 break;
\n+
382 case Sdc::SCHEME_LOOP:
\n+
383 limitFVar<Sdc::SCHEME_LOOP>(src, dst, channel);
\n+
384 break;
\n+
385 case Sdc::SCHEME_BILINEAR:
\n+
386 limitFVar<Sdc::SCHEME_BILINEAR>(src, dst, channel);
\n+
387 break;
\n+
388 }
\n+
389}
\n+
390
\n+
391template <typename REAL>
\n+
392template <class T, class U>
\n+
393inline void
\n+
394PrimvarRefinerReal<REAL>::InterpolateFaceUniform(int level, T const & src, U & dst) const {
\n
395
\n-
396public:
\n-
397 LimitStencil(BaseStencil const & other) : BaseStencil(other) { }
\n-
398 LimitStencil(int* size, Index * indices, float * weights,
\n-
399 float * duWeights=0, float * dvWeights=0,
\n-
400 float * duuWeights=0, float * duvWeights=0, float * dvvWeights=0)
\n-
401 : BaseStencil(size, indices, weights,
\n-
402 duWeights, dvWeights, duuWeights, duvWeights, dvvWeights) { }
\n-
403};
\n+
396 assert(level>0 && level<=(int)_refiner._refinements.size());
\n+
397
\n+
398 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n+
399 Vtr::internal::Level const & child = refinement.child();
\n+
400
\n+
401 for (int cFace = 0; cFace < child.getNumFaces(); ++cFace) {
\n+
402
\n+
403 Vtr::Index pFace = refinement.getChildFaceParentFace(cFace);
\n
404
\n-
405
\n-
408template <typename REAL>
\n-
409class LimitStencilTableReal : public StencilTableReal<REAL> {
\n-
410protected:
\n-
411 LimitStencilTableReal(
\n-
412 int numControlVerts,
\n-
413 std::vector<int> const& offsets,
\n-
414 std::vector<int> const& sizes,
\n-
415 std::vector<int> const& sources,
\n-
416 std::vector<REAL> const& weights,
\n-
417 std::vector<REAL> const& duWeights,
\n-
418 std::vector<REAL> const& dvWeights,
\n-
419 std::vector<REAL> const& duuWeights,
\n-
420 std::vector<REAL> const& duvWeights,
\n-
421 std::vector<REAL> const& dvvWeights,
\n-
422 bool includeCoarseVerts,
\n-
423 size_t firstOffset);
\n+
405 dst[cFace] = src[pFace];
\n+
406 }
\n+
407}
\n+
408
\n+
409template <typename REAL>
\n+
410template <class T, class U>
\n+
411inline void
\n+
412PrimvarRefinerReal<REAL>::InterpolateVarying(int level, T const & src, U & dst) const {
\n+
413
\n+
414 assert(level>0 && level<=(int)_refiner._refinements.size());
\n+
415
\n+
416 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n+
417 Vtr::internal::Level const & parent = refinement.parent();
\n+
418
\n+
419 //
\n+
420 // Group values to interpolate based on origin -- note that there may
\n+
421 // be none originating from faces:
\n+
422 //
\n+
423 if (refinement.getNumChildVerticesFromFaces() > 0) {
\n
424
\n-
425public:
\n+
425 for (int face = 0; face < parent.getNumFaces(); ++face) {
\n
426
\n-
428 LimitStencilReal<REAL> GetLimitStencil(Index i) const;
\n+
427 Vtr::Index cVert = refinement.getFaceChildVertex(face);
\n+
428 if (Vtr::IndexIsValid(cVert)) {
\n
429
\n-
431 LimitStencilReal<REAL> operator[] (Index index) const;
\n+
430 // Apply the weights to the parent face's vertices:
\n+
431 ConstIndexArray fVerts = parent.getFaceVertices(face);
\n
432
\n-
434 std::vector<REAL> const & GetDuWeights() const {
\n-
435 return _duWeights;
\n-
436 }
\n-
437
\n-
439 std::vector<REAL> const & GetDvWeights() const {
\n-
440 return _dvWeights;
\n+
433 Weight fVaryingWeight = 1.0f / (Weight) fVerts.size();
\n+
434
\n+
435 dst[cVert].Clear();
\n+
436 for (int i = 0; i < fVerts.size(); ++i) {
\n+
437 dst[cVert].AddWithWeight(src[fVerts[i]], fVaryingWeight);
\n+
438 }
\n+
439 }
\n+
440 }
\n
441 }
\n-
442
\n-
444 std::vector<REAL> const & GetDuuWeights() const {
\n-
445 return _duuWeights;
\n-
446 }
\n-
447
\n-
449 std::vector<REAL> const & GetDuvWeights() const {
\n-
450 return _duvWeights;
\n-
451 }
\n-
452
\n-
454 std::vector<REAL> const & GetDvvWeights() const {
\n-
455 return _dvvWeights;
\n-
456 }
\n-
457
\n-
475 template <class T, class U>
\n-
476 void UpdateDerivs(T const & srcValues, U & uderivs, U & vderivs,
\n-
477 int start=-1, int end=-1) const {
\n-
478
\n-
479 this->update(srcValues, uderivs, _duWeights, start, end);
\n-
480 this->update(srcValues, vderivs, _dvWeights, start, end);
\n-
481 }
\n-
482
\n-
483 template <class T1, class T2, class U>
\n-
484 void UpdateDerivs(T1 const & srcBase, int numBase, T2 const & srcRef,
\n-
485 U & uderivs, U & vderivs, int start=-1, int end=-1) const {
\n-
486
\n-
487 this->update(srcBase, numBase, srcRef, uderivs, _duWeights, start, end);
\n-
488 this->update(srcBase, numBase, srcRef, vderivs, _dvWeights, start, end);
\n-
489 }
\n-
490
\n-
491 // Pointer interface for backward compatibility
\n-
492 template <class T, class U>
\n-
493 void UpdateDerivs(T const *src, U *uderivs, U *vderivs,
\n-
494 int start=-1, int end=-1) const {
\n-
495
\n-
496 this->update(src, uderivs, _duWeights, start, end);
\n-
497 this->update(src, vderivs, _dvWeights, start, end);
\n-
498 }
\n-
499 template <class T1, class T2, class U>
\n-
500 void UpdateDerivs(T1 const *srcBase, int numBase, T2 const *srcRef,
\n-
501 U *uderivs, U *vderivs, int start=-1, int end=-1) const {
\n+
442 for (int edge = 0; edge < parent.getNumEdges(); ++edge) {
\n+
443
\n+
444 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);
\n+
445 if (Vtr::IndexIsValid(cVert)) {
\n+
446
\n+
447 // Apply the weights to the parent edges's vertices
\n+
448 ConstIndexArray eVerts = parent.getEdgeVertices(edge);
\n+
449
\n+
450 dst[cVert].Clear();
\n+
451 dst[cVert].AddWithWeight(src[eVerts[0]], 0.5f);
\n+
452 dst[cVert].AddWithWeight(src[eVerts[1]], 0.5f);
\n+
453 }
\n+
454 }
\n+
455 for (int vert = 0; vert < parent.getNumVertices(); ++vert) {
\n+
456
\n+
457 Vtr::Index cVert = refinement.getVertexChildVertex(vert);
\n+
458 if (Vtr::IndexIsValid(cVert)) {
\n+
459
\n+
460 // Essentially copy the parent vertex:
\n+
461 dst[cVert].Clear();
\n+
462 dst[cVert].AddWithWeight(src[vert], 1.0f);
\n+
463 }
\n+
464 }
\n+
465}
\n+
466
\n+
467
\n+
468//
\n+
469// Internal implementation methods -- grouping vertices to be interpolated
\n+
470// based on the type of parent component from which they originated:
\n+
471//
\n+
472template <typename REAL>
\n+
473template <Sdc::SchemeType SCHEME, class T, class U>
\n+
474inline void
\n+
475PrimvarRefinerReal<REAL>::interpFromFaces(int level, T const & src, U & dst) const {
\n+
476
\n+
477 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n+
478 Vtr::internal::Level const & parent = refinement.parent();
\n+
479
\n+
480 if (refinement.getNumChildVerticesFromFaces() == 0) return;
\n+
481
\n+
482 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n+
483
\n+
484 Vtr::internal::StackBuffer<Weight,16> fVertWeights(parent.getMaxValence());
\n+
485
\n+
486 for (int face = 0; face < parent.getNumFaces(); ++face) {
\n+
487
\n+
488 Vtr::Index cVert = refinement.getFaceChildVertex(face);
\n+
489 if (!Vtr::IndexIsValid(cVert))
\n+
490 continue;
\n+
491
\n+
492 // Declare and compute mask weights for this vertex relative to its parent face:
\n+
493 ConstIndexArray fVerts = parent.getFaceVertices(face);
\n+
494
\n+
495 Mask fMask(fVertWeights, 0, 0);
\n+
496 Vtr::internal::FaceInterface fHood(fVerts.size());
\n+
497
\n+
498 scheme.ComputeFaceVertexMask(fHood, fMask);
\n+
499
\n+
500 // Apply the weights to the parent face's vertices:
\n+
501 dst[cVert].Clear();
\n
502
\n-
503 this->update(srcBase, numBase, srcRef, uderivs, _duWeights, start, end);
\n-
504 this->update(srcBase, numBase, srcRef, vderivs, _dvWeights, start, end);
\n-
505 }
\n-
506
\n-
527 template <class T, class U>
\n-
528 void Update2ndDerivs(T const & srcValues,
\n-
529 U & uuderivs, U & uvderivs, U & vvderivs,
\n-
530 int start=-1, int end=-1) const {
\n+
503 for (int i = 0; i < fVerts.size(); ++i) {
\n+
504
\n+
505 dst[cVert].AddWithWeight(src[fVerts[i]], fVertWeights[i]);
\n+
506 }
\n+
507 }
\n+
508}
\n+
509
\n+
510template <typename REAL>
\n+
511template <Sdc::SchemeType SCHEME, class T, class U>
\n+
512inline void
\n+
513PrimvarRefinerReal<REAL>::interpFromEdges(int level, T const & src, U & dst) const {
\n+
514
\n+
515 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n+
516 Vtr::internal::Level const & parent = refinement.parent();
\n+
517 Vtr::internal::Level const & child = refinement.child();
\n+
518
\n+
519 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n+
520
\n+
521 Vtr::internal::EdgeInterface eHood(parent);
\n+
522
\n+
523 Weight eVertWeights[2];
\n+
524 Vtr::internal::StackBuffer<Weight,8> eFaceWeights(parent.getMaxEdgeFaces());
\n+
525
\n+
526 for (int edge = 0; edge < parent.getNumEdges(); ++edge) {
\n+
527
\n+
528 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);
\n+
529 if (!Vtr::IndexIsValid(cVert))
\n+
530 continue;
\n
531
\n-
532 this->update(srcValues, uuderivs, _duuWeights, start, end);
\n-
533 this->update(srcValues, uvderivs, _duvWeights, start, end);
\n-
534 this->update(srcValues, vvderivs, _dvvWeights, start, end);
\n-
535 }
\n-
536
\n-
537 template <class T1, class T2, class U>
\n-
538 void Update2ndDerivs(T1 const & srcBase, int numBase, T2 const & srcRef,
\n-
539 U & uuderivs, U & uvderivs, U & vvderivs, int start=-1, int end=-1) const {
\n-
540
\n-
541 this->update(srcBase, numBase, srcRef, uuderivs, _duuWeights, start, end);
\n-
542 this->update(srcBase, numBase, srcRef, uvderivs, _duvWeights, start, end);
\n-
543 this->update(srcBase, numBase, srcRef, vvderivs, _dvvWeights, start, end);
\n-
544 }
\n-
545
\n-
546 // Pointer interface for backward compatibility
\n-
547 template <class T, class U>
\n-
548 void Update2ndDerivs(T const *src, T *uuderivs, U *uvderivs, U *vvderivs,
\n-
549 int start=-1, int end=-1) const {
\n+
532 // Declare and compute mask weights for this vertex relative to its parent edge:
\n+
533 ConstIndexArray eVerts = parent.getEdgeVertices(edge),
\n+
534 eFaces = parent.getEdgeFaces(edge);
\n+
535
\n+
536 Mask eMask(eVertWeights, 0, eFaceWeights);
\n+
537
\n+
538 eHood.SetIndex(edge);
\n+
539
\n+
540 Sdc::Crease::Rule pRule = (parent.getEdgeSharpness(edge) > 0.0f) ? Sdc::Crease::RULE_CREASE : Sdc::Crease::RULE_SMOOTH;
\n+
541 Sdc::Crease::Rule cRule = child.getVertexRule(cVert);
\n+
542
\n+
543 scheme.ComputeEdgeVertexMask(eHood, eMask, pRule, cRule);
\n+
544
\n+
545 // Apply the weights to the parent edges's vertices and (if applicable) to
\n+
546 // the child vertices of its incident faces:
\n+
547 dst[cVert].Clear();
\n+
548 dst[cVert].AddWithWeight(src[eVerts[0]], eVertWeights[0]);
\n+
549 dst[cVert].AddWithWeight(src[eVerts[1]], eVertWeights[1]);
\n
550
\n-
551 this->update(src, uuderivs, _duuWeights, start, end);
\n-
552 this->update(src, uvderivs, _duvWeights, start, end);
\n-
553 this->update(src, vvderivs, _dvvWeights, start, end);
\n-
554 }
\n-
555 template <class T1, class T2, class U>
\n-
556 void Update2ndDerivs(T1 const *srcBase, int numBase, T2 const *srcRef,
\n-
557 U *uuderivs, U *uvderivs, U *vvderivs, int start=-1, int end=-1) const {
\n+
551 if (eMask.GetNumFaceWeights() > 0) {
\n+
552
\n+
553 for (int i = 0; i < eFaces.size(); ++i) {
\n+
554
\n+
555 if (eMask.AreFaceWeightsForFaceCenters()) {
\n+
556 assert(refinement.getNumChildVerticesFromFaces() > 0);
\n+
557 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(eFaces[i]);
\n
558
\n-
559 this->update(srcBase, numBase, srcRef, uuderivs, _duuWeights, start, end);
\n-
560 this->update(srcBase, numBase, srcRef, uvderivs, _duvWeights, start, end);
\n-
561 this->update(srcBase, numBase, srcRef, vvderivs, _dvvWeights, start, end);
\n-
562 }
\n-
563
\n-
565 void Clear();
\n-
566
\n-
567private:
\n-
568 friend class LimitStencilTableFactoryReal<REAL>;
\n-
569
\n-
570 // Resize the table arrays (factory helper)
\n-
571 void resize(int nstencils, int nelems);
\n-
572
\n-
573private:
\n-
574 std::vector<REAL> _duWeights, // u derivative limit stencil weights
\n-
575 _dvWeights, // v derivative limit stencil weights
\n-
576 _duuWeights, // uu derivative limit stencil weights
\n-
577 _duvWeights, // uv derivative limit stencil weights
\n-
578 _dvvWeights; // vv derivative limit stencil weights
\n-
579};
\n-
580
\n-
583class LimitStencilTable : public LimitStencilTableReal<float> {
\n-
584protected:
\n-
585 typedef LimitStencilTableReal<float> BaseTable;
\n-
586
\n-
587public:
\n-
588 LimitStencil GetLimitStencil(Index index) const {
\n-
589 return LimitStencil(BaseTable::GetLimitStencil(index));
\n-
590 }
\n-
591 LimitStencil operator[] (Index index) const {
\n-
592 return LimitStencil(BaseTable::GetLimitStencil(index));
\n-
593 }
\n+
559 assert(Vtr::IndexIsValid(cVertOfFace));
\n+
560 dst[cVert].AddWithWeight(dst[cVertOfFace], eFaceWeights[i]);
\n+
561 } else {
\n+
562 Vtr::Index pFace = eFaces[i];
\n+
563 ConstIndexArray pFaceEdges = parent.getFaceEdges(pFace),
\n+
564 pFaceVerts = parent.getFaceVertices(pFace);
\n+
565
\n+
566 int eInFace = 0;
\n+
567 for ( ; pFaceEdges[eInFace] != edge; ++eInFace ) ;
\n+
568
\n+
569 int vInFace = eInFace + 2;
\n+
570 if (vInFace >= pFaceVerts.size()) vInFace -= pFaceVerts.size();
\n+
571
\n+
572 Vtr::Index pVertNext = pFaceVerts[vInFace];
\n+
573 dst[cVert].AddWithWeight(src[pVertNext], eFaceWeights[i]);
\n+
574 }
\n+
575 }
\n+
576 }
\n+
577 }
\n+
578}
\n+
579
\n+
580template <typename REAL>
\n+
581template <Sdc::SchemeType SCHEME, class T, class U>
\n+
582inline void
\n+
583PrimvarRefinerReal<REAL>::interpFromVerts(int level, T const & src, U & dst) const {
\n+
584
\n+
585 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n+
586 Vtr::internal::Level const & parent = refinement.parent();
\n+
587 Vtr::internal::Level const & child = refinement.child();
\n+
588
\n+
589 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n+
590
\n+
591 Vtr::internal::VertexInterface vHood(parent, child);
\n+
592
\n+
593 Vtr::internal::StackBuffer<Weight,32> weightBuffer(2*parent.getMaxValence());
\n
594
\n-
595protected:
\n-
596 LimitStencilTable(int numControlVerts,
\n-
597 std::vector<int> const& offsets,
\n-
598 std::vector<int> const& sizes,
\n-
599 std::vector<int> const& sources,
\n-
600 std::vector<float> const& weights,
\n-
601 std::vector<float> const& duWeights,
\n-
602 std::vector<float> const& dvWeights,
\n-
603 std::vector<float> const& duuWeights,
\n-
604 std::vector<float> const& duvWeights,
\n-
605 std::vector<float> const& dvvWeights,
\n-
606 bool includeCoarseVerts,
\n-
607 size_t firstOffset)
\n-
608 : BaseTable(numControlVerts,
\n-
609 offsets, sizes, sources, weights,
\n-
610 duWeights, dvWeights, duuWeights, duvWeights, dvvWeights,
\n-
611 includeCoarseVerts, firstOffset) { }
\n-
612};
\n-
613
\n-
614
\n-
615// Update values by applying cached stencil weights to new control values
\n-
616template <typename REAL>
\n-
617template <class T1, class T2, class U> void
\n-
618StencilTableReal<REAL>::update(T1 const &srcBase, int numBase,
\n-
619 T2 const &srcRef, U &dstValues,
\n-
620 std::vector<REAL> const &valueWeights, Index start, Index end) const {
\n-
621
\n-
622 int const * sizes = &_sizes.at(0);
\n-
623 Index const * indices = &_indices.at(0);
\n-
624 REAL const * weights = &valueWeights.at(0);
\n+
595 for (int vert = 0; vert < parent.getNumVertices(); ++vert) {
\n+
596
\n+
597 Vtr::Index cVert = refinement.getVertexChildVertex(vert);
\n+
598 if (!Vtr::IndexIsValid(cVert))
\n+
599 continue;
\n+
600
\n+
601 // Declare and compute mask weights for this vertex relative to its parent edge:
\n+
602 ConstIndexArray vEdges = parent.getVertexEdges(vert),
\n+
603 vFaces = parent.getVertexFaces(vert);
\n+
604
\n+
605 Weight vVertWeight,
\n+
606 * vEdgeWeights = weightBuffer,
\n+
607 * vFaceWeights = vEdgeWeights + vEdges.size();
\n+
608
\n+
609 Mask vMask(&vVertWeight, vEdgeWeights, vFaceWeights);
\n+
610
\n+
611 vHood.SetIndex(vert, cVert);
\n+
612
\n+
613 Sdc::Crease::Rule pRule = parent.getVertexRule(vert);
\n+
614 Sdc::Crease::Rule cRule = child.getVertexRule(cVert);
\n+
615
\n+
616 scheme.ComputeVertexVertexMask(vHood, vMask, pRule, cRule);
\n+
617
\n+
618 // Apply the weights to the parent vertex, the vertices opposite its incident
\n+
619 // edges, and the child vertices of its incident faces:
\n+
620 //
\n+
621 // In order to improve numerical precision, it's better to apply smaller weights
\n+
622 // first, so begin with the face-weights followed by the edge-weights and the
\n+
623 // vertex weight last.
\n+
624 dst[cVert].Clear();
\n
625
\n-
626 if (start > 0) {
\n-
627 assert(start < (Index)_offsets.size());
\n-
628 sizes += start;
\n-
629 indices += _offsets[start];
\n-
630 weights += _offsets[start];
\n-
631 } else {
\n-
632 start = 0;
\n-
633 }
\n-
634
\n-
635 int nstencils = ((end < start) ? GetNumStencils() : end) - start;
\n-
636
\n-
637 for (int i = 0; i < nstencils; ++i, ++sizes) {
\n-
638 dstValues[start + i].Clear();
\n-
639 for (int j = 0; j < *sizes; ++j, ++indices, ++weights) {
\n-
640 if (*indices < numBase) {
\n-
641 dstValues[start + i].AddWithWeight(srcBase[*indices], *weights);
\n-
642 } else {
\n-
643 dstValues[start + i].AddWithWeight(srcRef[*indices - numBase], *weights);
\n+
626 if (vMask.GetNumFaceWeights() > 0) {
\n+
627 assert(vMask.AreFaceWeightsForFaceCenters());
\n+
628
\n+
629 for (int i = 0; i < vFaces.size(); ++i) {
\n+
630
\n+
631 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(vFaces[i]);
\n+
632 assert(Vtr::IndexIsValid(cVertOfFace));
\n+
633 dst[cVert].AddWithWeight(dst[cVertOfFace], vFaceWeights[i]);
\n+
634 }
\n+
635 }
\n+
636 if (vMask.GetNumEdgeWeights() > 0) {
\n+
637
\n+
638 for (int i = 0; i < vEdges.size(); ++i) {
\n+
639
\n+
640 ConstIndexArray eVerts = parent.getEdgeVertices(vEdges[i]);
\n+
641 Vtr::Index pVertOppositeEdge = (eVerts[0] == vert) ? eVerts[1] : eVerts[0];
\n+
642
\n+
643 dst[cVert].AddWithWeight(src[pVertOppositeEdge], vEdgeWeights[i]);
\n
644 }
\n
645 }
\n-
646 }
\n-
647}
\n-
648template <typename REAL>
\n-
649template <class T, class U> void
\n-
650StencilTableReal<REAL>::update(T const &srcValues, U &dstValues,
\n-
651 std::vector<REAL> const &valueWeights, Index start, Index end) const {
\n-
652
\n-
653 int const * sizes = &_sizes.at(0);
\n-
654 Index const * indices = &_indices.at(0);
\n-
655 REAL const * weights = &valueWeights.at(0);
\n-
656
\n-
657 if (start > 0) {
\n-
658 assert(start < (Index)_offsets.size());
\n-
659 sizes += start;
\n-
660 indices += _offsets[start];
\n-
661 weights += _offsets[start];
\n-
662 } else {
\n-
663 start = 0;
\n-
664 }
\n-
665
\n-
666 int nstencils = ((end < start) ? GetNumStencils() : end) - start;
\n+
646 dst[cVert].AddWithWeight(src[vert], vVertWeight);
\n+
647 }
\n+
648}
\n+
649
\n+
650
\n+
651//
\n+
652// Internal face-varying implementation details:
\n+
653//
\n+
654template <typename REAL>
\n+
655template <Sdc::SchemeType SCHEME, class T, class U>
\n+
656inline void
\n+
657PrimvarRefinerReal<REAL>::interpFVarFromFaces(int level, T const & src, U & dst, int channel) const {
\n+
658
\n+
659 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n+
660
\n+
661 if (refinement.getNumChildVerticesFromFaces() == 0) return;
\n+
662
\n+
663 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n+
664
\n+
665 Vtr::internal::Level const & parentLevel = refinement.parent();
\n+
666 Vtr::internal::Level const & childLevel = refinement.child();
\n
667
\n-
668 for (int i = 0; i < nstencils; ++i, ++sizes) {
\n-
669 dstValues[start + i].Clear();
\n-
670 for (int j = 0; j < *sizes; ++j, ++indices, ++weights) {
\n-
671 dstValues[start + i].AddWithWeight(srcValues[*indices], *weights);
\n-
672 }
\n-
673 }
\n-
674}
\n-
675
\n-
676template <typename REAL>
\n-
677inline void
\n-
678StencilTableReal<REAL>::generateOffsets() {
\n-
679 Index offset=0;
\n-
680 int noffsets = (int)_sizes.size();
\n-
681 _offsets.resize(noffsets);
\n-
682 for (int i=0; i<(int)_sizes.size(); ++i ) {
\n-
683 _offsets[i]=offset;
\n-
684 offset+=_sizes[i];
\n-
685 }
\n-
686}
\n-
687
\n-
688template <typename REAL>
\n-
689inline void
\n-
690StencilTableReal<REAL>::resize(int nstencils, int nelems) {
\n-
691 _sizes.resize(nstencils);
\n-
692 _indices.resize(nelems);
\n-
693 _weights.resize(nelems);
\n-
694}
\n-
695
\n-
696template <typename REAL>
\n-
697inline void
\n-
698StencilTableReal<REAL>::reserve(int nstencils, int nelems) {
\n-
699 _sizes.reserve(nstencils);
\n-
700 _indices.reserve(nelems);
\n-
701 _weights.reserve(nelems);
\n-
702}
\n-
703
\n-
704template <typename REAL>
\n+
668 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel(channel);
\n+
669 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel(channel);
\n+
670
\n+
671 Vtr::internal::StackBuffer<Weight,16> fValueWeights(parentLevel.getMaxValence());
\n+
672
\n+
673 for (int face = 0; face < parentLevel.getNumFaces(); ++face) {
\n+
674
\n+
675 Vtr::Index cVert = refinement.getFaceChildVertex(face);
\n+
676 if (!Vtr::IndexIsValid(cVert))
\n+
677 continue;
\n+
678
\n+
679 Vtr::Index cVertValue = childFVar.getVertexValueOffset(cVert);
\n+
680
\n+
681 // The only difference for face-varying here is that we get the values associated
\n+
682 // with each face-vertex directly from the FVarLevel, rather than using the parent
\n+
683 // face-vertices directly. If any face-vertex has any sibling values, then we may
\n+
684 // get the wrong one using the face-vertex index directly.
\n+
685
\n+
686 // Declare and compute mask weights for this vertex relative to its parent face:
\n+
687 ConstIndexArray fValues = parentFVar.getFaceValues(face);
\n+
688
\n+
689 Mask fMask(fValueWeights, 0, 0);
\n+
690 Vtr::internal::FaceInterface fHood(fValues.size());
\n+
691
\n+
692 scheme.ComputeFaceVertexMask(fHood, fMask);
\n+
693
\n+
694 // Apply the weights to the parent face's vertices:
\n+
695 dst[cVertValue].Clear();
\n+
696
\n+
697 for (int i = 0; i < fValues.size(); ++i) {
\n+
698 dst[cVertValue].AddWithWeight(src[fValues[i]], fValueWeights[i]);
\n+
699 }
\n+
700 }
\n+
701}
\n+
702
\n+
703template <typename REAL>
\n+
704template <Sdc::SchemeType SCHEME, class T, class U>
\n
705inline void
\n-
706StencilTableReal<REAL>::shrinkToFit() {
\n-
707 std::vector<int>(_sizes).swap(_sizes);
\n-
708 std::vector<Index>(_indices).swap(_indices);
\n-
709 std::vector<REAL>(_weights).swap(_weights);
\n-
710}
\n+
706PrimvarRefinerReal<REAL>::interpFVarFromEdges(int level, T const & src, U & dst, int channel) const {
\n+
707
\n+
708 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n+
709
\n+
710 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n
711
\n-
712template <typename REAL>
\n-
713inline void
\n-
714StencilTableReal<REAL>::finalize() {
\n-
715 shrinkToFit();
\n-
716 generateOffsets();
\n-
717}
\n+
712 Vtr::internal::Level const & parentLevel = refinement.parent();
\n+
713 Vtr::internal::Level const & childLevel = refinement.child();
\n+
714
\n+
715 Vtr::internal::FVarRefinement const & refineFVar = refinement.getFVarRefinement(channel);
\n+
716 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel(channel);
\n+
717 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel(channel);
\n
718
\n-
719// Returns a Stencil at index i in the table
\n-
720template <typename REAL>
\n-
721inline StencilReal<REAL>
\n-
722StencilTableReal<REAL>::GetStencil(Index i) const {
\n-
723 assert((! _offsets.empty()) && i<(int)_offsets.size());
\n-
724
\n-
725 Index ofs = _offsets[i];
\n-
726
\n-
727 return StencilReal<REAL>(const_cast<int*>(&_sizes[i]),
\n-
728 const_cast<Index*>(&_indices[ofs]),
\n-
729 const_cast<REAL*>(&_weights[ofs]));
\n-
730}
\n-
731
\n-
732template <typename REAL>
\n-
733inline StencilReal<REAL>
\n-
734StencilTableReal<REAL>::operator[] (Index index) const {
\n-
735 return GetStencil(index);
\n-
736}
\n+
719 //
\n+
720 // Allocate and initialize (if linearly interpolated) interpolation weights for
\n+
721 // the edge mask:
\n+
722 //
\n+
723 Weight eVertWeights[2];
\n+
724 Vtr::internal::StackBuffer<Weight,8> eFaceWeights(parentLevel.getMaxEdgeFaces());
\n+
725
\n+
726 Mask eMask(eVertWeights, 0, eFaceWeights);
\n+
727
\n+
728 bool isLinearFVar = parentFVar.isLinear() || (_refiner._subdivType == Sdc::SCHEME_BILINEAR);
\n+
729 if (isLinearFVar) {
\n+
730 eMask.SetNumVertexWeights(2);
\n+
731 eMask.SetNumEdgeWeights(0);
\n+
732 eMask.SetNumFaceWeights(0);
\n+
733
\n+
734 eVertWeights[0] = 0.5f;
\n+
735 eVertWeights[1] = 0.5f;
\n+
736 }
\n
737
\n-
738template <typename REAL>
\n-
739inline void
\n-
740LimitStencilTableReal<REAL>::resize(int nstencils, int nelems) {
\n-
741 StencilTableReal<REAL>::resize(nstencils, nelems);
\n-
742 _duWeights.resize(nelems);
\n-
743 _dvWeights.resize(nelems);
\n-
744}
\n+
738 Vtr::internal::EdgeInterface eHood(parentLevel);
\n+
739
\n+
740 for (int edge = 0; edge < parentLevel.getNumEdges(); ++edge) {
\n+
741
\n+
742 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);
\n+
743 if (!Vtr::IndexIsValid(cVert))
\n+
744 continue;
\n
745
\n-
746// Returns a LimitStencil at index i in the table
\n-
747template <typename REAL>
\n-
748inline LimitStencilReal<REAL>
\n-
749LimitStencilTableReal<REAL>::GetLimitStencil(Index i) const {
\n-
750 assert((! this->GetOffsets().empty()) && i<(int)this->GetOffsets().size());
\n-
751
\n-
752 Index ofs = this->GetOffsets()[i];
\n-
753
\n-
754 if (!_duWeights.empty() && !_dvWeights.empty() &&
\n-
755 !_duuWeights.empty() && !_duvWeights.empty() && !_dvvWeights.empty()) {
\n-
756 return LimitStencilReal<REAL>(
\n-
757 const_cast<int *>(&this->GetSizes()[i]),
\n-
758 const_cast<Index *>(&this->GetControlIndices()[ofs]),
\n-
759 const_cast<REAL *>(&this->GetWeights()[ofs]),
\n-
760 const_cast<REAL *>(&GetDuWeights()[ofs]),
\n-
761 const_cast<REAL *>(&GetDvWeights()[ofs]),
\n-
762 const_cast<REAL *>(&GetDuuWeights()[ofs]),
\n-
763 const_cast<REAL *>(&GetDuvWeights()[ofs]),
\n-
764 const_cast<REAL *>(&GetDvvWeights()[ofs]) );
\n-
765 } else if (!_duWeights.empty() && !_dvWeights.empty()) {
\n-
766 return LimitStencilReal<REAL>(
\n-
767 const_cast<int *>(&this->GetSizes()[i]),
\n-
768 const_cast<Index *>(&this->GetControlIndices()[ofs]),
\n-
769 const_cast<REAL *>(&this->GetWeights()[ofs]),
\n-
770 const_cast<REAL *>(&GetDuWeights()[ofs]),
\n-
771 const_cast<REAL *>(&GetDvWeights()[ofs]) );
\n-
772 } else {
\n-
773 return LimitStencilReal<REAL>(
\n-
774 const_cast<int *>(&this->GetSizes()[i]),
\n-
775 const_cast<Index *>(&this->GetControlIndices()[ofs]),
\n-
776 const_cast<REAL *>(&this->GetWeights()[ofs]) );
\n-
777 }
\n-
778}
\n-
779
\n-
780template <typename REAL>
\n-
781inline LimitStencilReal<REAL>
\n-
782LimitStencilTableReal<REAL>::operator[] (Index index) const {
\n-
783 return GetLimitStencil(index);
\n-
784}
\n-
785
\n-
786} // end namespace Far
\n-
787
\n-
788} // end namespace OPENSUBDIV_VERSION
\n-
789using namespace OPENSUBDIV_VERSION;
\n-
790
\n-
791} // end namespace OpenSubdiv
\n+
746 ConstIndexArray cVertValues = childFVar.getVertexValues(cVert);
\n+
747
\n+
748 bool fvarEdgeVertMatchesVertex = childFVar.valueTopologyMatches(cVertValues[0]);
\n+
749 if (fvarEdgeVertMatchesVertex) {
\n+
750 //
\n+
751 // If smoothly interpolated, compute new weights for the edge mask:
\n+
752 //
\n+
753 if (!isLinearFVar) {
\n+
754 eHood.SetIndex(edge);
\n+
755
\n+
756 Sdc::Crease::Rule pRule = (parentLevel.getEdgeSharpness(edge) > 0.0f)
\n+
757 ? Sdc::Crease::RULE_CREASE : Sdc::Crease::RULE_SMOOTH;
\n+
758 Sdc::Crease::Rule cRule = childLevel.getVertexRule(cVert);
\n+
759
\n+
760 scheme.ComputeEdgeVertexMask(eHood, eMask, pRule, cRule);
\n+
761 }
\n+
762
\n+
763 // Apply the weights to the parent edge's vertices and (if applicable) to
\n+
764 // the child vertices of its incident faces:
\n+
765 //
\n+
766 // Even though the face-varying topology matches the vertex topology, we need
\n+
767 // to be careful here when getting values corresponding to the two end-vertices.
\n+
768 // While the edge may be continuous, the vertices at their ends may have
\n+
769 // discontinuities elsewhere in their neighborhood (i.e. on the "other side"
\n+
770 // of the end-vertex) and so have sibling values associated with them. In most
\n+
771 // cases the topology for an end-vertex will match and we can use it directly,
\n+
772 // but we must still check and retrieve as needed.
\n+
773 //
\n+
774 // Indices for values corresponding to face-vertices are guaranteed to match,
\n+
775 // so we can use the child-vertex indices directly.
\n+
776 //
\n+
777 // And by "directly", we always use getVertexValue(vertexIndex) to reference
\n+
778 // values in the "src" to account for the possible indirection that may exist at
\n+
779 // level 0 -- where there may be fewer values than vertices and an additional
\n+
780 // indirection is necessary. We can use a vertex index directly for "dst" when
\n+
781 // it matches.
\n+
782 //
\n+
783 Vtr::Index eVertValues[2];
\n+
784
\n+
785 parentFVar.getEdgeFaceValues(edge, 0, eVertValues);
\n+
786
\n+
787 Index cVertValue = cVertValues[0];
\n+
788
\n+
789 dst[cVertValue].Clear();
\n+
790 dst[cVertValue].AddWithWeight(src[eVertValues[0]], eVertWeights[0]);
\n+
791 dst[cVertValue].AddWithWeight(src[eVertValues[1]], eVertWeights[1]);
\n
792
\n-
793#endif // OPENSUBDIV3_FAR_STENCILTABLE_H
\n+
793 if (eMask.GetNumFaceWeights() > 0) {
\n+
794
\n+
795 ConstIndexArray eFaces = parentLevel.getEdgeFaces(edge);
\n+
796
\n+
797 for (int i = 0; i < eFaces.size(); ++i) {
\n+
798 if (eMask.AreFaceWeightsForFaceCenters()) {
\n+
799
\n+
800 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(eFaces[i]);
\n+
801 assert(Vtr::IndexIsValid(cVertOfFace));
\n+
802
\n+
803 Vtr::Index cValueOfFace = childFVar.getVertexValueOffset(cVertOfFace);
\n+
804 dst[cVertValue].AddWithWeight(dst[cValueOfFace], eFaceWeights[i]);
\n+
805 } else {
\n+
806 Vtr::Index pFace = eFaces[i];
\n+
807 ConstIndexArray pFaceEdges = parentLevel.getFaceEdges(pFace),
\n+
808 pFaceVerts = parentLevel.getFaceVertices(pFace);
\n+
809
\n+
810 int eInFace = 0;
\n+
811 for ( ; pFaceEdges[eInFace] != edge; ++eInFace ) ;
\n+
812
\n+
813 // Edge "i" spans vertices [i,i+1] so we want i+2...
\n+
814 int vInFace = eInFace + 2;
\n+
815 if (vInFace >= pFaceVerts.size()) vInFace -= pFaceVerts.size();
\n+
816
\n+
817 Vtr::Index pValueNext = parentFVar.getFaceValues(pFace)[vInFace];
\n+
818 dst[cVertValue].AddWithWeight(src[pValueNext], eFaceWeights[i]);
\n+
819 }
\n+
820 }
\n+
821 }
\n+
822 } else {
\n+
823 //
\n+
824 // Mismatched edge-verts should just be linearly interpolated between the pairs of
\n+
825 // values for each sibling of the child edge-vertex -- the question is: which face
\n+
826 // holds that pair of values for a given sibling?
\n+
827 //
\n+
828 // In the manifold case, the sibling and edge-face indices will correspond. We
\n+
829 // will eventually need to update this to account for > 3 incident faces.
\n+
830 //
\n+
831 for (int i = 0; i < cVertValues.size(); ++i) {
\n+
832 Vtr::Index eVertValues[2];
\n+
833 int eFaceIndex = refineFVar.getChildValueParentSource(cVert, i);
\n+
834 assert(eFaceIndex == i);
\n+
835
\n+
836 parentFVar.getEdgeFaceValues(edge, eFaceIndex, eVertValues);
\n+
837
\n+
838 Index cVertValue = cVertValues[i];
\n+
839
\n+
840 dst[cVertValue].Clear();
\n+
841 dst[cVertValue].AddWithWeight(src[eVertValues[0]], 0.5);
\n+
842 dst[cVertValue].AddWithWeight(src[eVertValues[1]], 0.5);
\n+
843 }
\n+
844 }
\n+
845 }
\n+
846}
\n+
847
\n+
848template <typename REAL>
\n+
849template <Sdc::SchemeType SCHEME, class T, class U>
\n+
850inline void
\n+
851PrimvarRefinerReal<REAL>::interpFVarFromVerts(int level, T const & src, U & dst, int channel) const {
\n+
852
\n+
853 Vtr::internal::Refinement const & refinement = _refiner.getRefinement(level-1);
\n+
854
\n+
855 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n+
856
\n+
857 Vtr::internal::Level const & parentLevel = refinement.parent();
\n+
858 Vtr::internal::Level const & childLevel = refinement.child();
\n+
859
\n+
860 Vtr::internal::FVarRefinement const & refineFVar = refinement.getFVarRefinement(channel);
\n+
861 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel(channel);
\n+
862 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel(channel);
\n+
863
\n+
864 bool isLinearFVar = parentFVar.isLinear() || (_refiner._subdivType == Sdc::SCHEME_BILINEAR);
\n+
865
\n+
866 Vtr::internal::StackBuffer<Weight,32> weightBuffer(2*parentLevel.getMaxValence());
\n+
867
\n+
868 Vtr::internal::StackBuffer<Vtr::Index,16> vEdgeValues(parentLevel.getMaxValence());
\n+
869
\n+
870 Vtr::internal::VertexInterface vHood(parentLevel, childLevel);
\n+
871
\n+
872 for (int vert = 0; vert < parentLevel.getNumVertices(); ++vert) {
\n+
873
\n+
874 Vtr::Index cVert = refinement.getVertexChildVertex(vert);
\n+
875 if (!Vtr::IndexIsValid(cVert))
\n+
876 continue;
\n+
877
\n+
878 ConstIndexArray pVertValues = parentFVar.getVertexValues(vert),
\n+
879 cVertValues = childFVar.getVertexValues(cVert);
\n+
880
\n+
881 bool fvarVertVertMatchesVertex = childFVar.valueTopologyMatches(cVertValues[0]);
\n+
882 if (isLinearFVar && fvarVertVertMatchesVertex) {
\n+
883 dst[cVertValues[0]].Clear();
\n+
884 dst[cVertValues[0]].AddWithWeight(src[pVertValues[0]], 1.0f);
\n+
885 continue;
\n+
886 }
\n+
887
\n+
888 if (fvarVertVertMatchesVertex) {
\n+
889 //
\n+
890 // Declare and compute mask weights for this vertex relative to its parent edge:
\n+
891 //
\n+
892 // (We really need to encapsulate this somewhere else for use here and in the
\n+
893 // general case)
\n+
894 //
\n+
895 ConstIndexArray vEdges = parentLevel.getVertexEdges(vert);
\n+
896
\n+
897 Weight vVertWeight;
\n+
898 Weight * vEdgeWeights = weightBuffer;
\n+
899 Weight * vFaceWeights = vEdgeWeights + vEdges.size();
\n+
900
\n+
901 Mask vMask(&vVertWeight, vEdgeWeights, vFaceWeights);
\n+
902
\n+
903 vHood.SetIndex(vert, cVert);
\n+
904
\n+
905 Sdc::Crease::Rule pRule = parentLevel.getVertexRule(vert);
\n+
906 Sdc::Crease::Rule cRule = childLevel.getVertexRule(cVert);
\n+
907
\n+
908 scheme.ComputeVertexVertexMask(vHood, vMask, pRule, cRule);
\n+
909
\n+
910 // Apply the weights to the parent vertex, the vertices opposite its incident
\n+
911 // edges, and the child vertices of its incident faces:
\n+
912 //
\n+
913 // Even though the face-varying topology matches the vertex topology, we need
\n+
914 // to be careful here when getting values corresponding to vertices at the
\n+
915 // ends of edges. While the edge may be continuous, the end vertex may have
\n+
916 // discontinuities elsewhere in their neighborhood (i.e. on the "other side"
\n+
917 // of the end-vertex) and so have sibling values associated with them. In most
\n+
918 // cases the topology for an end-vertex will match and we can use it directly,
\n+
919 // but we must still check and retrieve as needed.
\n+
920 //
\n+
921 // Indices for values corresponding to face-vertices are guaranteed to match,
\n+
922 // so we can use the child-vertex indices directly.
\n+
923 //
\n+
924 // And by "directly", we always use getVertexValue(vertexIndex) to reference
\n+
925 // values in the "src" to account for the possible indirection that may exist at
\n+
926 // level 0 -- where there may be fewer values than vertices and an additional
\n+
927 // indirection is necessary. We can use a vertex index directly for "dst" when
\n+
928 // it matches.
\n+
929 //
\n+
930 // As with applying the mask to vertex data, in order to improve numerical
\n+
931 // precision, it's better to apply smaller weights first, so begin with the
\n+
932 // face-weights followed by the edge-weights and the vertex weight last.
\n+
933 //
\n+
934 Vtr::Index pVertValue = pVertValues[0];
\n+
935 Vtr::Index cVertValue = cVertValues[0];
\n+
936
\n+
937 dst[cVertValue].Clear();
\n+
938 if (vMask.GetNumFaceWeights() > 0) {
\n+
939 assert(vMask.AreFaceWeightsForFaceCenters());
\n+
940
\n+
941 ConstIndexArray vFaces = parentLevel.getVertexFaces(vert);
\n+
942
\n+
943 for (int i = 0; i < vFaces.size(); ++i) {
\n+
944
\n+
945 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(vFaces[i]);
\n+
946 assert(Vtr::IndexIsValid(cVertOfFace));
\n+
947
\n+
948 Vtr::Index cValueOfFace = childFVar.getVertexValueOffset(cVertOfFace);
\n+
949 dst[cVertValue].AddWithWeight(dst[cValueOfFace], vFaceWeights[i]);
\n+
950 }
\n+
951 }
\n+
952 if (vMask.GetNumEdgeWeights() > 0) {
\n+
953
\n+
954 parentFVar.getVertexEdgeValues(vert, vEdgeValues);
\n+
955
\n+
956 for (int i = 0; i < vEdges.size(); ++i) {
\n+
957 dst[cVertValue].AddWithWeight(src[vEdgeValues[i]], vEdgeWeights[i]);
\n+
958 }
\n+
959 }
\n+
960 dst[cVertValue].AddWithWeight(src[pVertValue], vVertWeight);
\n+
961 } else {
\n+
962 //
\n+
963 // Each FVar value associated with a vertex will be either a corner or a crease,
\n+
964 // or potentially in transition from corner to crease:
\n+
965 // - if the CHILD is a corner, there can be no transition so we have a corner
\n+
966 // - otherwise if the PARENT is a crease, both will be creases (no transition)
\n+
967 // - otherwise the parent must be a corner and the child a crease (transition)
\n+
968 //
\n+
969 Vtr::internal::FVarLevel::ConstValueTagArray pValueTags = parentFVar.getVertexValueTags(vert);
\n+
970 Vtr::internal::FVarLevel::ConstValueTagArray cValueTags = childFVar.getVertexValueTags(cVert);
\n+
971
\n+
972 for (int cSiblingIndex = 0; cSiblingIndex < cVertValues.size(); ++cSiblingIndex) {
\n+
973 int pSiblingIndex = refineFVar.getChildValueParentSource(cVert, cSiblingIndex);
\n+
974 assert(pSiblingIndex == cSiblingIndex);
\n+
975
\n+
976 typedef Vtr::internal::FVarLevel::Sibling SiblingIntType;
\n+
977
\n+
978 SiblingIntType cSibling = (SiblingIntType) cSiblingIndex;
\n+
979 SiblingIntType pSibling = (SiblingIntType) pSiblingIndex;
\n+
980
\n+
981 Vtr::Index pVertValue = pVertValues[pSibling];
\n+
982 Vtr::Index cVertValue = cVertValues[cSibling];
\n+
983
\n+
984 dst[cVertValue].Clear();
\n+
985 if (isLinearFVar || cValueTags[cSibling].isCorner()) {
\n+
986 dst[cVertValue].AddWithWeight(src[pVertValue], 1.0f);
\n+
987 } else {
\n+
988 //
\n+
989 // We have either a crease or a transition from corner to crease -- in
\n+
990 // either case, we need the end values for the full/fractional crease:
\n+
991 //
\n+
992 Index pEndValues[2];
\n+
993 parentFVar.getVertexCreaseEndValues(vert, pSibling, pEndValues);
\n+
994
\n+
995 Weight vWeight = 0.75f;
\n+
996 Weight eWeight = 0.125f;
\n+
997
\n+
998 //
\n+
999 // If semi-sharp we need to apply fractional weighting -- if made sharp because
\n+
1000 // of the other sibling (dependent-sharp) use the fractional weight from that
\n+
1001 // other sibling (should only occur when there are 2):
\n+
1002 //
\n+
1003 if (pValueTags[pSibling].isSemiSharp()) {
\n+
1004 Weight wCorner = pValueTags[pSibling].isDepSharp()
\n+
1005 ? refineFVar.getFractionalWeight(vert, !pSibling, cVert, !cSibling)
\n+
1006 : refineFVar.getFractionalWeight(vert, pSibling, cVert, cSibling);
\n+
1007 Weight wCrease = 1.0f - wCorner;
\n+
1008
\n+
1009 vWeight = wCrease * 0.75f + wCorner;
\n+
1010 eWeight = wCrease * 0.125f;
\n+
1011 }
\n+
1012 dst[cVertValue].AddWithWeight(src[pEndValues[0]], eWeight);
\n+
1013 dst[cVertValue].AddWithWeight(src[pEndValues[1]], eWeight);
\n+
1014 dst[cVertValue].AddWithWeight(src[pVertValue], vWeight);
\n+
1015 }
\n+
1016 }
\n+
1017 }
\n+
1018 }
\n+
1019}
\n+
1020
\n+
1021template <typename REAL>
\n+
1022template <Sdc::SchemeType SCHEME, class T, class U, class U1, class U2>
\n+
1023inline void
\n+
1024PrimvarRefinerReal<REAL>::limit(T const & src, U & dstPos, U1 * dstTan1Ptr, U2 * dstTan2Ptr) const {
\n+
1025
\n+
1026 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n+
1027
\n+
1028 Vtr::internal::Level const & level = _refiner.getLevel(_refiner.GetMaxLevel());
\n+
1029
\n+
1030 int maxWeightsPerMask = 1 + 2 * level.getMaxValence();
\n+
1031 bool hasTangents = (dstTan1Ptr && dstTan2Ptr);
\n+
1032 int numMasks = 1 + (hasTangents ? 2 : 0);
\n+
1033
\n+
1034 Vtr::internal::StackBuffer<Index,33> indexBuffer(maxWeightsPerMask);
\n+
1035 Vtr::internal::StackBuffer<Weight,99> weightBuffer(numMasks * maxWeightsPerMask);
\n+
1036
\n+
1037 Weight * vPosWeights = weightBuffer,
\n+
1038 * ePosWeights = vPosWeights + 1,
\n+
1039 * fPosWeights = ePosWeights + level.getMaxValence();
\n+
1040 Weight * vTan1Weights = vPosWeights + maxWeightsPerMask,
\n+
1041 * eTan1Weights = ePosWeights + maxWeightsPerMask,
\n+
1042 * fTan1Weights = fPosWeights + maxWeightsPerMask;
\n+
1043 Weight * vTan2Weights = vTan1Weights + maxWeightsPerMask,
\n+
1044 * eTan2Weights = eTan1Weights + maxWeightsPerMask,
\n+
1045 * fTan2Weights = fTan1Weights + maxWeightsPerMask;
\n+
1046
\n+
1047 Mask posMask( vPosWeights, ePosWeights, fPosWeights);
\n+
1048 Mask tan1Mask(vTan1Weights, eTan1Weights, fTan1Weights);
\n+
1049 Mask tan2Mask(vTan2Weights, eTan2Weights, fTan2Weights);
\n+
1050
\n+
1051 // This is a bit obscure -- assigning both parent and child as last level -- but
\n+
1052 // this mask type was intended for another purpose. Consider one for the limit:
\n+
1053 Vtr::internal::VertexInterface vHood(level, level);
\n+
1054
\n+
1055 for (int vert = 0; vert < level.getNumVertices(); ++vert) {
\n+
1056 ConstIndexArray vEdges = level.getVertexEdges(vert);
\n+
1057
\n+
1058 // Incomplete vertices (present in sparse refinement) do not have their full
\n+
1059 // topological neighborhood to determine a proper limit -- just leave the
\n+
1060 // vertex at the refined location and continue to the next:
\n+
1061 if (level.getVertexTag(vert)._incomplete || (vEdges.size() == 0)) {
\n+
1062 dstPos[vert].Clear();
\n+
1063 dstPos[vert].AddWithWeight(src[vert], 1.0);
\n+
1064 if (hasTangents) {
\n+
1065 (*dstTan1Ptr)[vert].Clear();
\n+
1066 (*dstTan2Ptr)[vert].Clear();
\n+
1067 }
\n+
1068 continue;
\n+
1069 }
\n+
1070
\n+
1071 //
\n+
1072 // Limit masks require the subdivision Rule for the vertex in order to deal
\n+
1073 // with infinitely sharp features correctly -- including boundaries and corners.
\n+
1074 // The vertex neighborhood is minimally defined with vertex and edge counts.
\n+
1075 //
\n+
1076 Sdc::Crease::Rule vRule = level.getVertexRule(vert);
\n+
1077
\n+
1078 // This is a bit obscure -- child vertex index will be ignored here
\n+
1079 vHood.SetIndex(vert, vert);
\n+
1080
\n+
1081 if (hasTangents) {
\n+
1082 scheme.ComputeVertexLimitMask(vHood, posMask, tan1Mask, tan2Mask, vRule);
\n+
1083 } else {
\n+
1084 scheme.ComputeVertexLimitMask(vHood, posMask, vRule);
\n+
1085 }
\n+
1086
\n+
1087 //
\n+
1088 // Gather the neighboring vertices of this vertex -- the vertices opposite its
\n+
1089 // incident edges, and the opposite vertices of its incident faces:
\n+
1090 //
\n+
1091 Index * eIndices = indexBuffer;
\n+
1092 Index * fIndices = indexBuffer + vEdges.size();
\n+
1093
\n+
1094 for (int i = 0; i < vEdges.size(); ++i) {
\n+
1095 ConstIndexArray eVerts = level.getEdgeVertices(vEdges[i]);
\n+
1096
\n+
1097 eIndices[i] = (eVerts[0] == vert) ? eVerts[1] : eVerts[0];
\n+
1098 }
\n+
1099 if (posMask.GetNumFaceWeights() || (hasTangents && tan1Mask.GetNumFaceWeights())) {
\n+
1100 ConstIndexArray vFaces = level.getVertexFaces(vert);
\n+
1101 ConstLocalIndexArray vInFace = level.getVertexFaceLocalIndices(vert);
\n+
1102
\n+
1103 for (int i = 0; i < vFaces.size(); ++i) {
\n+
1104 ConstIndexArray fVerts = level.getFaceVertices(vFaces[i]);
\n+
1105
\n+
1106 LocalIndex vOppInFace = (vInFace[i] + 2);
\n+
1107 if (vOppInFace >= fVerts.size()) vOppInFace -= (LocalIndex)fVerts.size();
\n+
1108
\n+
1109 fIndices[i] = level.getFaceVertices(vFaces[i])[vOppInFace];
\n+
1110 }
\n+
1111 }
\n+
1112
\n+
1113 //
\n+
1114 // Combine the weights and indices for position and tangents. As with applying
\n+
1115 // refinement masks to vertex data, in order to improve numerical precision, it's
\n+
1116 // better to apply smaller weights first, so begin with the face-weights followed
\n+
1117 // by the edge-weights and the vertex weight last.
\n+
1118 //
\n+
1119 dstPos[vert].Clear();
\n+
1120 for (int i = 0; i < posMask.GetNumFaceWeights(); ++i) {
\n+
1121 dstPos[vert].AddWithWeight(src[fIndices[i]], fPosWeights[i]);
\n+
1122 }
\n+
1123 for (int i = 0; i < posMask.GetNumEdgeWeights(); ++i) {
\n+
1124 dstPos[vert].AddWithWeight(src[eIndices[i]], ePosWeights[i]);
\n+
1125 }
\n+
1126 dstPos[vert].AddWithWeight(src[vert], vPosWeights[0]);
\n+
1127
\n+
1128 //
\n+
1129 // Apply the tangent masks -- both will have the same number of weights and
\n+
1130 // indices (one tangent may be "padded" to accommodate the other), but these
\n+
1131 // may differ from those of the position:
\n+
1132 //
\n+
1133 if (hasTangents) {
\n+
1134 assert(tan1Mask.GetNumFaceWeights() == tan2Mask.GetNumFaceWeights());
\n+
1135 assert(tan1Mask.GetNumEdgeWeights() == tan2Mask.GetNumEdgeWeights());
\n+
1136
\n+
1137 U1 & dstTan1 = *dstTan1Ptr;
\n+
1138 U2 & dstTan2 = *dstTan2Ptr;
\n+
1139
\n+
1140 dstTan1[vert].Clear();
\n+
1141 dstTan2[vert].Clear();
\n+
1142 for (int i = 0; i < tan1Mask.GetNumFaceWeights(); ++i) {
\n+
1143 dstTan1[vert].AddWithWeight(src[fIndices[i]], fTan1Weights[i]);
\n+
1144 dstTan2[vert].AddWithWeight(src[fIndices[i]], fTan2Weights[i]);
\n+
1145 }
\n+
1146 for (int i = 0; i < tan1Mask.GetNumEdgeWeights(); ++i) {
\n+
1147 dstTan1[vert].AddWithWeight(src[eIndices[i]], eTan1Weights[i]);
\n+
1148 dstTan2[vert].AddWithWeight(src[eIndices[i]], eTan2Weights[i]);
\n+
1149 }
\n+
1150 dstTan1[vert].AddWithWeight(src[vert], vTan1Weights[0]);
\n+
1151 dstTan2[vert].AddWithWeight(src[vert], vTan2Weights[0]);
\n+
1152 }
\n+
1153 }
\n+
1154}
\n+
1155
\n+
1156template <typename REAL>
\n+
1157template <Sdc::SchemeType SCHEME, class T, class U>
\n+
1158inline void
\n+
1159PrimvarRefinerReal<REAL>::limitFVar(T const & src, U & dst, int channel) const {
\n+
1160
\n+
1161 Sdc::Scheme<SCHEME> scheme(_refiner._subdivOptions);
\n+
1162
\n+
1163 Vtr::internal::Level const & level = _refiner.getLevel(_refiner.GetMaxLevel());
\n+
1164 Vtr::internal::FVarLevel const & fvarChannel = level.getFVarLevel(channel);
\n+
1165
\n+
1166 int maxWeightsPerMask = 1 + 2 * level.getMaxValence();
\n+
1167
\n+
1168 Vtr::internal::StackBuffer<Weight,33> weightBuffer(maxWeightsPerMask);
\n+
1169 Vtr::internal::StackBuffer<Index,16> vEdgeBuffer(level.getMaxValence());
\n+
1170
\n+
1171 // This is a bit obscure -- assign both parent and child as last level
\n+
1172 Vtr::internal::VertexInterface vHood(level, level);
\n+
1173
\n+
1174 for (int vert = 0; vert < level.getNumVertices(); ++vert) {
\n+
1175
\n+
1176 ConstIndexArray vEdges = level.getVertexEdges(vert);
\n+
1177 ConstIndexArray vValues = fvarChannel.getVertexValues(vert);
\n+
1178
\n+
1179 // Incomplete vertices (present in sparse refinement) do not have their full
\n+
1180 // topological neighborhood to determine a proper limit -- just leave the
\n+
1181 // values (perhaps more than one per vertex) at the refined location.
\n+
1182 //
\n+
1183 // The same can be done if the face-varying channel is purely linear.
\n+
1184 //
\n+
1185 bool isIncomplete = (level.getVertexTag(vert)._incomplete || (vEdges.size() == 0));
\n+
1186 if (isIncomplete || fvarChannel.isLinear()) {
\n+
1187 for (int i = 0; i < vValues.size(); ++i) {
\n+
1188 Vtr::Index vValue = vValues[i];
\n+
1189
\n+
1190 dst[vValue].Clear();
\n+
1191 dst[vValue].AddWithWeight(src[vValue], 1.0f);
\n+
1192 }
\n+
1193 continue;
\n+
1194 }
\n+
1195
\n+
1196 bool fvarVertMatchesVertex = fvarChannel.valueTopologyMatches(vValues[0]);
\n+
1197 if (fvarVertMatchesVertex) {
\n+
1198
\n+
1199 // Assign the mask weights to the common buffer and compute the mask:
\n+
1200 //
\n+
1201 Weight * vWeights = weightBuffer,
\n+
1202 * eWeights = vWeights + 1,
\n+
1203 * fWeights = eWeights + vEdges.size();
\n+
1204
\n+
1205 Mask vMask(vWeights, eWeights, fWeights);
\n+
1206
\n+
1207 vHood.SetIndex(vert, vert);
\n+
1208
\n+
1209 scheme.ComputeVertexLimitMask(vHood, vMask, level.getVertexRule(vert));
\n+
1210
\n+
1211 //
\n+
1212 // Apply mask to corresponding FVar values for neighboring vertices:
\n+
1213 //
\n+
1214 Vtr::Index vValue = vValues[0];
\n+
1215
\n+
1216 dst[vValue].Clear();
\n+
1217 if (vMask.GetNumFaceWeights() > 0) {
\n+
1218 assert(!vMask.AreFaceWeightsForFaceCenters());
\n+
1219
\n+
1220 ConstIndexArray vFaces = level.getVertexFaces(vert);
\n+
1221 ConstLocalIndexArray vInFace = level.getVertexFaceLocalIndices(vert);
\n+
1222
\n+
1223 for (int i = 0; i < vFaces.size(); ++i) {
\n+
1224 ConstIndexArray faceValues = fvarChannel.getFaceValues(vFaces[i]);
\n+
1225 LocalIndex vOppInFace = vInFace[i] + 2;
\n+
1226 if (vOppInFace >= faceValues.size()) vOppInFace -= faceValues.size();
\n+
1227
\n+
1228 Index vValueOppositeFace = faceValues[vOppInFace];
\n+
1229
\n+
1230 dst[vValue].AddWithWeight(src[vValueOppositeFace], fWeights[i]);
\n+
1231 }
\n+
1232 }
\n+
1233 if (vMask.GetNumEdgeWeights() > 0) {
\n+
1234 Index * vEdgeValues = vEdgeBuffer;
\n+
1235 fvarChannel.getVertexEdgeValues(vert, vEdgeValues);
\n+
1236
\n+
1237 for (int i = 0; i < vEdges.size(); ++i) {
\n+
1238 dst[vValue].AddWithWeight(src[vEdgeValues[i]], eWeights[i]);
\n+
1239 }
\n+
1240 }
\n+
1241 dst[vValue].AddWithWeight(src[vValue], vWeights[0]);
\n+
1242 } else {
\n+
1243 //
\n+
1244 // Sibling FVar values associated with a vertex will be either a corner or a crease:
\n+
1245 //
\n+
1246 for (int i = 0; i < vValues.size(); ++i) {
\n+
1247 Vtr::Index vValue = vValues[i];
\n+
1248
\n+
1249 dst[vValue].Clear();
\n+
1250 if (fvarChannel.getValueTag(vValue).isCorner()) {
\n+
1251 dst[vValue].AddWithWeight(src[vValue], 1.0f);
\n+
1252 } else {
\n+
1253 Index vEndValues[2];
\n+
1254 fvarChannel.getVertexCreaseEndValues(vert, i, vEndValues);
\n+
1255
\n+
1256 dst[vValue].AddWithWeight(src[vEndValues[0]], 1.0f/6.0f);
\n+
1257 dst[vValue].AddWithWeight(src[vEndValues[1]], 1.0f/6.0f);
\n+
1258 dst[vValue].AddWithWeight(src[vValue], 2.0f/3.0f);
\n+
1259 }
\n+
1260 }
\n+
1261 }
\n+
1262 }
\n+
1263}
\n+
1264
\n+
1265class PrimvarRefiner : public PrimvarRefinerReal<float> {
\n+
1266public:
\n+
1267 PrimvarRefiner(TopologyRefiner const & refiner)
\n+
1268 : PrimvarRefinerReal<float>(refiner) { }
\n+
1269};
\n+
1270
\n+
1271} // end namespace Far
\n+
1272
\n+
1273} // end namespace OPENSUBDIV_VERSION
\n+
1274using namespace OPENSUBDIV_VERSION;
\n+
1275} // end namespace OpenSubdiv
\n+
1276
\n+
1277#endif /* OPENSUBDIV3_FAR_PRIMVAR_REFINER_H */
\n \n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LocalIndex
Vtr::LocalIndex LocalIndex
Definition: types.h:42
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::Error
void Error(ErrorType err, const char *format,...)
Sends an OSD error with a message (internal use only)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstIndexArray
Vtr::ConstIndexArray ConstIndexArray
Definition: types.h:47
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstLocalIndexArray
Vtr::ConstLocalIndexArray ConstLocalIndexArray
Definition: types.h:48
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::FAR_RUNTIME_ERROR
@ FAR_RUNTIME_ERROR
Issue a generic runtime error, but continue execution.
Definition: error.h:40
\n \n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::StencilReal
StencilReal(int *size, Index *indices, REAL *weights)
Constructor.
Definition: stencilTable.h:67
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::Next
void Next()
Advance to the next stencil in the table.
Definition: stencilTable.h:98
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::StencilReal
StencilReal(StencilReal const &other)
Copy constructor.
Definition: stencilTable.h:71
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetVertexIndices
Index const * GetVertexIndices() const
Returns the control vertices' indices.
Definition: stencilTable.h:88
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetSize
int GetSize() const
Returns the size of the stencil.
Definition: stencilTable.h:78
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetSizePtr
int * GetSizePtr() const
Returns the size of the stencil as a pointer.
Definition: stencilTable.h:83
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetWeights
REAL const * GetWeights() const
Returns the interpolation weights.
Definition: stencilTable.h:93
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil
Vertex stencil class wrapping the template for compatibility.
Definition: stencilTable.h:116
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil::Stencil
Stencil(int *size, Index *indices, float *weights)
Definition: stencilTable.h:123
\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumControlVertices
int GetNumControlVertices() const
Returns the number of control vertices indexed in the table.
Definition: stencilTable.h:161
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::update
void update(T1 const &srcBase, int numBase, T2 const &srcRef, U &dstValues, std::vector< REAL > const &valueWeights, Index start, Index end) const
Definition: stencilTable.h:618
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues
void UpdateValues(T1 const &srcBase, int numBase, T2 const &srcRef, U &dstValues, Index start=-1, Index end=-1) const
Definition: stencilTable.h:210
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetControlIndices
std::vector< Index > const & GetControlIndices() const
Returns the indices of the control vertices.
Definition: stencilTable.h:179
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::operator[]
StencilReal< REAL > operator[](Index index) const
Returns the stencil at index i in the table.
Definition: stencilTable.h:734
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::update
void update(T const &srcValues, U &dstValues, std::vector< REAL > const &valueWeights, Index start, Index end) const
Definition: stencilTable.h:650
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetSizes
std::vector< int > const & GetSizes() const
Returns the number of control vertices of each stencil in the table.
Definition: stencilTable.h:169
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumStencils
int GetNumStencils() const
Returns the number of stencils in the table.
Definition: stencilTable.h:156
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetOffsets
std::vector< Index > const & GetOffsets() const
Returns the offset to a given stencil (factory may leave empty)
Definition: stencilTable.h:174
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetStencil
StencilReal< REAL > GetStencil(Index i) const
Returns a Stencil at index i in the table.
Definition: stencilTable.h:722
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::Clear
void Clear()
Clears the stencils from the table.
\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetWeights
std::vector< REAL > const & GetWeights() const
Returns the stencil interpolation weights.
Definition: stencilTable.h:184
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues
void UpdateValues(T1 const *srcBase, int numBase, T2 const *srcRef, U *dst, Index start=-1, Index end=-1) const
Definition: stencilTable.h:221
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::StencilTableReal
StencilTableReal(int numControlVerts, std::vector< int > const &offsets, std::vector< int > const &sizes, std::vector< int > const &sources, std::vector< REAL > const &weights, bool includeCoarseVerts, size_t firstOffset)
\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues
void UpdateValues(T const *src, U *dst, Index start=-1, Index end=-1) const
Definition: stencilTable.h:217
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues
void UpdateValues(T const &srcValues, U &dstValues, Index start=-1, Index end=-1) const
Updates point values based on the control values.
Definition: stencilTable.h:205
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::StencilTable
StencilTable(int numControlVerts, std::vector< int > const &offsets, std::vector< int > const &sizes, std::vector< int > const &sources, std::vector< float > const &weights, bool includeCoarseVerts, size_t firstOffset)
Definition: stencilTable.h:288
\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuvWeights
REAL const * GetDuvWeights() const
Returns the uv derivative weights.
Definition: stencilTable.h:356
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDvWeights
REAL const * GetDvWeights() const
Returns the v derivative weights.
Definition: stencilTable.h:346
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::Next
void Next()
Advance to the next stencil in the table.
Definition: stencilTable.h:366
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDvvWeights
REAL const * GetDvvWeights() const
Returns the vv derivative weights.
Definition: stencilTable.h:361
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::LimitStencilReal
LimitStencilReal(int *size, Index *indices, REAL *weights, REAL *duWeights=0, REAL *dvWeights=0, REAL *duuWeights=0, REAL *duvWeights=0, REAL *dvvWeights=0)
Constructor.
Definition: stencilTable.h:324
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuWeights
REAL const * GetDuWeights() const
Returns the u derivative weights.
Definition: stencilTable.h:341
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuuWeights
REAL const * GetDuuWeights() const
Returns the uu derivative weights.
Definition: stencilTable.h:351
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil
Limit point stencil class wrapping the template for compatibility.
Definition: stencilTable.h:392
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil::LimitStencil
LimitStencil(int *size, Index *indices, float *weights, float *duWeights=0, float *dvWeights=0, float *duuWeights=0, float *duvWeights=0, float *dvvWeights=0)
Definition: stencilTable.h:398
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuuWeights
std::vector< REAL > const & GetDuuWeights() const
Returns the 'uu' derivative stencil interpolation weights.
Definition: stencilTable.h:444
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::operator[]
LimitStencilReal< REAL > operator[](Index index) const
Returns the limit stencil at index i in the table.
Definition: stencilTable.h:782
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDvvWeights
std::vector< REAL > const & GetDvvWeights() const
Returns the 'vv' derivative stencil interpolation weights.
Definition: stencilTable.h:454
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs
void UpdateDerivs(T const *src, U *uderivs, U *vderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:493
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs
void Update2ndDerivs(T const *src, T *uuderivs, U *uvderivs, U *vvderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:548
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDvWeights
std::vector< REAL > const & GetDvWeights() const
Returns the 'v' derivative stencil interpolation weights.
Definition: stencilTable.h:439
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs
void Update2ndDerivs(T1 const *srcBase, int numBase, T2 const *srcRef, U *uuderivs, U *uvderivs, U *vvderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:556
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs
void UpdateDerivs(T const &srcValues, U &uderivs, U &vderivs, int start=-1, int end=-1) const
Updates derivative values based on the control values.
Definition: stencilTable.h:476
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetLimitStencil
LimitStencilReal< REAL > GetLimitStencil(Index i) const
Returns a LimitStencil at index i in the table.
Definition: stencilTable.h:749
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::LimitStencilTableReal
LimitStencilTableReal(int numControlVerts, std::vector< int > const &offsets, std::vector< int > const &sizes, std::vector< int > const &sources, std::vector< REAL > const &weights, std::vector< REAL > const &duWeights, std::vector< REAL > const &dvWeights, std::vector< REAL > const &duuWeights, std::vector< REAL > const &duvWeights, std::vector< REAL > const &dvvWeights, bool includeCoarseVerts, size_t firstOffset)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs
void Update2ndDerivs(T const &srcValues, U &uuderivs, U &uvderivs, U &vvderivs, int start=-1, int end=-1) const
Updates 2nd derivative values based on the control values.
Definition: stencilTable.h:528
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuWeights
std::vector< REAL > const & GetDuWeights() const
Returns the 'u' derivative stencil interpolation weights.
Definition: stencilTable.h:434
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Clear
void Clear()
Clears the stencils from the table.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs
void Update2ndDerivs(T1 const &srcBase, int numBase, T2 const &srcRef, U &uuderivs, U &uvderivs, U &vvderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:538
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuvWeights
std::vector< REAL > const & GetDuvWeights() const
Returns the 'uv' derivative stencil interpolation weights.
Definition: stencilTable.h:449
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs
void UpdateDerivs(T1 const &srcBase, int numBase, T2 const &srcRef, U &uderivs, U &vderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:484
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs
void UpdateDerivs(T1 const *srcBase, int numBase, T2 const *srcRef, U *uderivs, U *vderivs, int start=-1, int end=-1) const
Definition: stencilTable.h:500
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable::LimitStencilTable
LimitStencilTable(int numControlVerts, std::vector< int > const &offsets, std::vector< int > const &sizes, std::vector< int > const &sources, std::vector< float > const &weights, std::vector< float > const &duWeights, std::vector< float > const &dvWeights, std::vector< float > const &duuWeights, std::vector< float > const &duvWeights, std::vector< float > const &dvvWeights, bool includeCoarseVerts, size_t firstOffset)
Definition: stencilTable.h:596
\n-\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Vtr::IndexIsValid
bool IndexIsValid(Index index)
Definition: types.h:58
\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal
Applies refinement operations to generic primvar data.
Definition: primvarRefiner.h:56
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Limit
void Limit(T const &src, U &dstPos) const
Apply limit weights to a primvar buffer.
Definition: primvarRefiner.h:319
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Limit
void Limit(T const &src, U &dstPos, U1 &dstTan1, U2 &dstTan2) const
Definition: primvarRefiner.h:344
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateFaceUniform
void InterpolateFaceUniform(int level, T const &src, U &dst) const
Refine uniform (per-face) primvar data between levels.
Definition: primvarRefiner.h:394
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateVarying
void InterpolateVarying(int level, T const &src, U &dst) const
Apply only varying interpolation weights to a primvar buffer for a single level of refinement.
Definition: primvarRefiner.h:412
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Interpolate
void Interpolate(int level, T const &src, U &dst) const
Apply vertex interpolation weights to a primvar buffer for a single level of refinement.
Definition: primvarRefiner.h:267
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateFaceVarying
void InterpolateFaceVarying(int level, T const &src, U &dst, int channel=0) const
Apply face-varying interpolation weights to a primvar buffer associated with a particular face-varyin...
Definition: primvarRefiner.h:293
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::LimitFaceVarying
void LimitFaceVarying(T const &src, U &dst, int channel=0) const
Definition: primvarRefiner.h:369
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner
Stores topology data for a specified set of refinement options.
Definition: topologyRefiner.h:51
\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme
Scheme is a class template which provides all implementation for the subdivision schemes supported by...
Definition: scheme.h:66
\n+\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-stencilTable.h\n+primvarRefiner.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2013 Pixar\n+ 2// Copyright 2015 DreamWorks Animation LLC.\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -29,993 +29,1398 @@\n 17//\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n- 24\n- 25#ifndef OPENSUBDIV3_FAR_STENCILTABLE_H\n- 26#define OPENSUBDIV3_FAR_STENCILTABLE_H\n- 27\n- 28#include \"../version.h\"\n- 29\n- 30#include \"../far/types.h\"\n- 31\n- 32#include \n- 33#include \n- 34#include \n- 35#include \n- 36\n- 37namespace OpenSubdiv {\n- 38namespace OPENSUBDIV_VERSION {\n- 39\n- 40namespace Far {\n- 41\n- 42// Forward declarations for friends:\n- 43class PatchTableBuilder;\n+ 24#ifndef OPENSUBDIV3_FAR_PRIMVAR_REFINER_H\n+ 25#define OPENSUBDIV3_FAR_PRIMVAR_REFINER_H\n+ 26\n+ 27#include \"../version.h\"\n+ 28\n+ 29#include \"../sdc/types.h\"\n+ 30#include \"../sdc/options.h\"\n+ 31#include \"../sdc/bilinearScheme.h\"\n+ 32#include \"../sdc/catmarkScheme.h\"\n+ 33#include \"../sdc/loopScheme.h\"\n+ 34#include \"../vtr/level.h\"\n+ 35#include \"../vtr/fvarLevel.h\"\n+ 36#include \"../vtr/refinement.h\"\n+ 37#include \"../vtr/fvarRefinement.h\"\n+ 38#include \"../vtr/stackBuffer.h\"\n+ 39#include \"../vtr/componentInterfaces.h\"\n+ 40#include \"../far/types.h\"\n+ 41#include \"../far/error.h\"\n+ 42#include \"../far/topologyLevel.h\"\n+ 43#include \"../far/topologyRefiner.h\"\n 44\n- 45template class StencilTableFactoryReal;\n- 46template class LimitStencilTableFactoryReal;\n- 47\n- 52template \n-53class StencilReal {\n- 54public:\n- 55\n-57 StencilReal() {}\n- 58\n-67 StencilReal(int * size, Index * indices, REAL * weights)\n- 68 : _size(size), _indices(indices), _weights(weights) { }\n- 69\n-71 StencilReal(StencilReal const & other) {\n- 72 _size = other._size;\n- 73 _indices = other._indices;\n- 74 _weights = other._weights;\n- 75 }\n- 76\n-78 int GetSize() const {\n- 79 return *_size;\n- 80 }\n- 81\n-83 int * GetSizePtr() const {\n- 84 return _size;\n- 85 }\n- 86\n-88 Index const * GetVertexIndices() const {\n- 89 return _indices;\n- 90 }\n- 91\n-93 REAL const * GetWeights() const {\n- 94 return _weights;\n- 95 }\n- 96\n-98 void Next() {\n- 99 int stride = *_size;\n- 100 ++_size;\n- 101 _indices += stride;\n- 102 _weights += stride;\n- 103 }\n- 104\n- 105protected:\n- 106 friend class StencilTableFactoryReal;\n- 107 friend class LimitStencilTableFactoryReal;\n- 108\n-109 int * _size;\n-110 Index * _indices;\n-111 REAL * _weights;\n- 112};\n- 113\n-116class Stencil : public StencilReal {\n- 117protected:\n-118 typedef StencilReal BaseStencil;\n- 119\n- 120public:\n-121 Stencil() : BaseStencil() { }\n-122 Stencil(BaseStencil const & other) : BaseStencil(other) { }\n-123 Stencil(int * size, Index * indices, float * weights)\n- 124 : BaseStencil(size, indices, weights) { }\n- 125};\n- 126\n- 127\n- 140template \n-141class StencilTableReal {\n- 142protected:\n-143 StencilTableReal(int numControlVerts,\n- 144 std::vector const& offsets,\n- 145 std::vector const& sizes,\n- 146 std::vector const& sources,\n- 147 std::vector const& weights,\n- 148 bool includeCoarseVerts,\n- 149 size_t firstOffset);\n- 150\n- 151public:\n- 152\n-153 virtual ~StencilTableReal() {};\n- 154\n-156 int GetNumStencils() const {\n- 157 return (int)_sizes.size();\n- 158 }\n- 159\n-161 int GetNumControlVertices() const {\n- 162 return _numControlVertices;\n- 163 }\n- 164\n-166 StencilReal GetStencil(Index i) const;\n- 167\n-169 std::vector const & GetSizes() const {\n- 170 return _sizes;\n- 171 }\n- 172\n-174 std::vector const & GetOffsets() const {\n- 175 return _offsets;\n- 176 }\n- 177\n-179 std::vector const & GetControlIndices() const {\n- 180 return _indices;\n- 181 }\n- 182\n-184 std::vector const & GetWeights() const {\n- 185 return _weights;\n- 186 }\n- 187\n-189 StencilReal operator[](Index index) const;\n+ 45#include \n+ 46\n+ 47namespace OpenSubdiv {\n+ 48namespace OPENSUBDIV_VERSION {\n+ 49\n+ 50namespace Far {\n+ 51\n+ 55template \n+56class PrimvarRefinerReal {\n+ 57\n+ 58public:\n+59 PrimvarRefinerReal(TopologyRefiner const & refiner) : _refiner(refiner) { }\n+60 ~PrimvarRefinerReal() { }\n+ 61\n+62 TopologyRefiner const & GetTopologyRefiner() const { return _refiner; }\n+ 63\n+ 65\n+ 100\n+113 template void Interpolate(int level, T const & src, U &\n+dst) const;\n+ 114\n+130 template void InterpolateVarying(int level, T const &\n+src, U & dst) const;\n+ 131\n+148 template void InterpolateFaceUniform(int level, T const\n+& src, U & dst) const;\n+ 149\n+161 template void InterpolateFaceVarying(int level, T const\n+& src, U & dst, int channel = 0) const;\n+ 162\n+ 163\n+175 template void Limit(T const & src, U & dstPos) const;\n+ 176\n+ 177 template \n+178 void Limit(T const & src, U & dstPos, U1 & dstTan1, U2 & dstTan2) const;\n+ 179\n+180 template void LimitFaceVarying(T const & src, U & dst,\n+int channel = 0) const;\n+ 181\n+ 183\n+ 184private:\n+ 185 typedef REAL Weight;\n+ 186\n+ 187 // Non-copyable:\n+ 188 PrimvarRefinerReal(PrimvarRefinerReal const & src) : _refiner\n+(src._refiner) { }\n+ 189 PrimvarRefinerReal & operator=(PrimvarRefinerReal const &) { return *this;\n+}\n 190\n- 204 template \n-205 void UpdateValues(T const &srcValues, U &dstValues, Index start=-1, Index\n-end=-1) const {\n- 206 this->update(srcValues, dstValues, _weights, start, end);\n- 207 }\n- 208\n- 209 template \n-210 void UpdateValues(T1 const &srcBase, int numBase, T2 const &srcRef,\n- 211 U &dstValues, Index start=-1, Index end=-1) const {\n- 212 this->update(srcBase, numBase, srcRef, dstValues, _weights, start, end);\n- 213 }\n- 214\n- 215 // Pointer interface for backward compatibility\n- 216 template \n-217 void UpdateValues(T const *src, U *dst, Index start=-1, Index end=-1) const\n-{\n- 218 this->update(src, dst, _weights, start, end);\n- 219 }\n- 220 template \n-221 void UpdateValues(T1 const *srcBase, int numBase, T2 const *srcRef,\n- 222 U *dst, Index start=-1, Index end=-1) const {\n- 223 this->update(srcBase, numBase, srcRef, dst, _weights, start, end);\n- 224 }\n- 225\n-227 void Clear();\n- 228\n- 229protected:\n- 230\n- 231 // Update values by applying cached stencil weights to new control values\n- 232 template \n-233 void update( T const &srcValues, U &dstValues,\n- 234 std::vector const & valueWeights, Index start, Index end) const;\n- 235 template \n-236 void update( T1 const &srcBase, int numBase, T2 const &srcRef, U\n-&dstValues,\n- 237 std::vector const & valueWeights, Index start, Index end) const;\n- 238\n- 239 // Populate the offsets table from the stencil sizes in _sizes (factory\n-helper)\n-240 void generateOffsets();\n+ 191 template void interpFromFaces\n+(int, T const &, U &) const;\n+ 192 template void interpFromEdges\n+(int, T const &, U &) const;\n+ 193 template void interpFromVerts\n+(int, T const &, U &) const;\n+ 194\n+ 195 template void\n+interpFVarFromFaces(int, T const &, U &, int) const;\n+ 196 template void\n+interpFVarFromEdges(int, T const &, U &, int) const;\n+ 197 template void\n+interpFVarFromVerts(int, T const &, U &, int) const;\n+ 198\n+ 199 template \n+ 200 void limit(T const & src, U & pos, U1 * tan1, U2 * tan2) const;\n+ 201\n+ 202 template \n+ 203 void limitFVar(T const & src, U & dst, int channel) const;\n+ 204\n+ 205private:\n+ 206 TopologyRefiner const & _refiner;\n+ 207\n+ 208private:\n+ 209 //\n+ 210 // Local class to fulfill interface for in the Scheme mask\n+queries:\n+ 211 //\n+ 212 class Mask {\n+ 213 public:\n+ 214 typedef REAL Weight; // Also part of the expected interface\n+ 215\n+ 216 public:\n+ 217 Mask(Weight* v, Weight* e, Weight* f) :\n+ 218 _vertWeights(v), _edgeWeights(e), _faceWeights(f),\n+ 219 _vertCount(0), _edgeCount(0), _faceCount(0),\n+ 220 _faceWeightsForFaceCenters(false)\n+ 221 { }\n+ 222\n+ 223 ~Mask() { }\n+ 224\n+ 225 public: // Generic interface expected of :\n+ 226 int GetNumVertexWeights() const { return _vertCount; }\n+ 227 int GetNumEdgeWeights() const { return _edgeCount; }\n+ 228 int GetNumFaceWeights() const { return _faceCount; }\n+ 229\n+ 230 void SetNumVertexWeights(int count) { _vertCount = count; }\n+ 231 void SetNumEdgeWeights( int count) { _edgeCount = count; }\n+ 232 void SetNumFaceWeights( int count) { _faceCount = count; }\n+ 233\n+ 234 Weight const& VertexWeight(int index) const { return _vertWeights[index];\n+}\n+ 235 Weight const& EdgeWeight( int index) const { return _edgeWeights[index]; }\n+ 236 Weight const& FaceWeight( int index) const { return _faceWeights[index]; }\n+ 237\n+ 238 Weight& VertexWeight(int index) { return _vertWeights[index]; }\n+ 239 Weight& EdgeWeight( int index) { return _edgeWeights[index]; }\n+ 240 Weight& FaceWeight( int index) { return _faceWeights[index]; }\n 241\n- 242 // Resize the table arrays (factory helper)\n-243 void resize(int nstencils, int nelems);\n+ 242 bool AreFaceWeightsForFaceCenters() const { return\n+_faceWeightsForFaceCenters; }\n+ 243 void SetFaceWeightsForFaceCenters(bool on) { _faceWeightsForFaceCenters =\n+on; }\n 244\n- 245 // Reserves the table arrays (factory helper)\n-246 void reserve(int nstencils, int nelems);\n- 247\n- 248 // Reallocates the table arrays to remove excess capacity (factory helper)\n-249 void shrinkToFit();\n- 250\n- 251 // Performs any final operations on internal tables (factory helper)\n-252 void finalize();\n+ 245 private:\n+ 246 Weight* _vertWeights;\n+ 247 Weight* _edgeWeights;\n+ 248 Weight* _faceWeights;\n+ 249\n+ 250 int _vertCount;\n+ 251 int _edgeCount;\n+ 252 int _faceCount;\n 253\n- 254protected:\n-255 StencilTableReal() : _numControlVertices(0) {}\n-256 StencilTableReal(int numControlVerts)\n- 257 : _numControlVertices(numControlVerts)\n- 258 { }\n- 259\n- 260 friend class StencilTableFactoryReal;\n-261 friend class Far::PatchTableBuilder;\n- 262\n-263 int _numControlVertices; // number of control vertices\n- 264\n-265 std::vector _sizes; // number of coefficients for each stencil\n-266 std::vector _offsets, // offset to the start of each stencil\n-267 _indices; // indices of contributing coarse vertices\n-268 std::vector _weights; // stencil weight coefficients\n- 269};\n+ 254 bool _faceWeightsForFaceCenters;\n+ 255 };\n+ 256};\n+ 257\n+ 258\n+ 259//\n+ 260// Public entry points to the methods. Queries of the scheme type and its\n+ 261// use as a template parameter in subsequent implementation will be\n+factored\n+ 262// out of a later release:\n+ 263//\n+ 264template \n+ 265template \n+ 266inline void\n+267PrimvarRefinerReal::Interpolate(int level, T const & src, U & dst)\n+const {\n+ 268\n+ 269 assert(level>0 && level<=(int)_refiner._refinements.size());\n 270\n-273class StencilTable : public StencilTableReal {\n- 274protected:\n-275 typedef StencilTableReal BaseTable;\n- 276\n- 277public:\n-278 Stencil GetStencil(Index index) const {\n- 279 return Stencil(BaseTable::GetStencil(index));\n- 280 }\n-281 Stencil operator[](Index index) const {\n- 282 return Stencil(BaseTable::GetStencil(index));\n- 283 }\n- 284\n- 285protected:\n-286 StencilTable() : BaseTable() { }\n-287 StencilTable(int numControlVerts) : BaseTable(numControlVerts) { }\n-288 StencilTable(int numControlVerts,\n- 289 std::vector const& offsets,\n- 290 std::vector const& sizes,\n- 291 std::vector const& sources,\n- 292 std::vector const& weights,\n- 293 bool includeCoarseVerts,\n- 294 size_t firstOffset)\n- 295 : BaseTable(numControlVerts, offsets,\n- 296 sizes, sources, weights, includeCoarseVerts, firstOffset) { }\n- 297};\n- 298\n- 299\n- 302template \n-303class LimitStencilReal : public StencilReal {\n- 304public:\n- 305\n-324 LimitStencilReal( int* size,\n- 325 Index * indices,\n- 326 REAL * weights,\n- 327 REAL * duWeights=0,\n- 328 REAL * dvWeights=0,\n- 329 REAL * duuWeights=0,\n- 330 REAL * duvWeights=0,\n- 331 REAL * dvvWeights=0)\n- 332 : StencilReal(size, indices, weights),\n- 333 _duWeights(duWeights),\n- 334 _dvWeights(dvWeights),\n- 335 _duuWeights(duuWeights),\n- 336 _duvWeights(duvWeights),\n- 337 _dvvWeights(dvvWeights) {\n+ 271 switch (_refiner._subdivType) {\n+ 272 case Sdc::SCHEME_CATMARK:\n+ 273 interpFromFaces(level, src, dst);\n+ 274 interpFromEdges(level, src, dst);\n+ 275 interpFromVerts(level, src, dst);\n+ 276 break;\n+ 277 case Sdc::SCHEME_LOOP:\n+ 278 interpFromFaces(level, src, dst);\n+ 279 interpFromEdges(level, src, dst);\n+ 280 interpFromVerts(level, src, dst);\n+ 281 break;\n+ 282 case Sdc::SCHEME_BILINEAR:\n+ 283 interpFromFaces(level, src, dst);\n+ 284 interpFromEdges(level, src, dst);\n+ 285 interpFromVerts(level, src, dst);\n+ 286 break;\n+ 287 }\n+ 288}\n+ 289\n+ 290template \n+ 291template \n+ 292inline void\n+293PrimvarRefinerReal::InterpolateFaceVarying(int level, T const & src, U\n+& dst, int channel) const {\n+ 294\n+ 295 assert(level>0 && level<=(int)_refiner._refinements.size());\n+ 296\n+ 297 switch (_refiner._subdivType) {\n+ 298 case Sdc::SCHEME_CATMARK:\n+ 299 interpFVarFromFaces(level, src, dst, channel);\n+ 300 interpFVarFromEdges(level, src, dst, channel);\n+ 301 interpFVarFromVerts(level, src, dst, channel);\n+ 302 break;\n+ 303 case Sdc::SCHEME_LOOP:\n+ 304 interpFVarFromFaces(level, src, dst, channel);\n+ 305 interpFVarFromEdges(level, src, dst, channel);\n+ 306 interpFVarFromVerts(level, src, dst, channel);\n+ 307 break;\n+ 308 case Sdc::SCHEME_BILINEAR:\n+ 309 interpFVarFromFaces(level, src, dst, channel);\n+ 310 interpFVarFromEdges(level, src, dst, channel);\n+ 311 interpFVarFromVerts(level, src, dst, channel);\n+ 312 break;\n+ 313 }\n+ 314}\n+ 315\n+ 316template \n+ 317template \n+ 318inline void\n+319PrimvarRefinerReal::Limit(T const & src, U & dst) const {\n+ 320\n+ 321 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() ==\n+0) {\n+ 322 Error(FAR_RUNTIME_ERROR,\n+ 323 \"Failure in PrimvarRefiner::Limit() -- \"\n+ 324 \"last level of refinement does not include full topology.\");\n+ 325 return;\n+ 326 }\n+ 327\n+ 328 switch (_refiner._subdivType) {\n+ 329 case Sdc::SCHEME_CATMARK:\n+ 330 limit(src, dst, (U*)0, (U*)0);\n+ 331 break;\n+ 332 case Sdc::SCHEME_LOOP:\n+ 333 limit(src, dst, (U*)0, (U*)0);\n+ 334 break;\n+ 335 case Sdc::SCHEME_BILINEAR:\n+ 336 limit(src, dst, (U*)0, (U*)0);\n+ 337 break;\n 338 }\n- 339\n-341 REAL const * GetDuWeights() const {\n- 342 return _duWeights;\n- 343 }\n- 344\n-346 REAL const * GetDvWeights() const {\n- 347 return _dvWeights;\n- 348 }\n- 349\n-351 REAL const * GetDuuWeights() const {\n- 352 return _duuWeights;\n- 353 }\n- 354\n-356 REAL const * GetDuvWeights() const {\n- 357 return _duvWeights;\n- 358 }\n- 359\n-361 REAL const * GetDvvWeights() const {\n- 362 return _dvvWeights;\n+ 339}\n+ 340\n+ 341template \n+ 342template \n+ 343inline void\n+344PrimvarRefinerReal::Limit(T const & src, U & dstPos, U1 & dstTan1, U2\n+& dstTan2) const {\n+ 345\n+ 346 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() ==\n+0) {\n+ 347 Error(FAR_RUNTIME_ERROR,\n+ 348 \"Failure in PrimvarRefiner::Limit() -- \"\n+ 349 \"last level of refinement does not include full topology.\");\n+ 350 return;\n+ 351 }\n+ 352\n+ 353 switch (_refiner._subdivType) {\n+ 354 case Sdc::SCHEME_CATMARK:\n+ 355 limit(src, dstPos, &dstTan1, &dstTan2);\n+ 356 break;\n+ 357 case Sdc::SCHEME_LOOP:\n+ 358 limit(src, dstPos, &dstTan1, &dstTan2);\n+ 359 break;\n+ 360 case Sdc::SCHEME_BILINEAR:\n+ 361 limit(src, dstPos, &dstTan1, &dstTan2);\n+ 362 break;\n 363 }\n- 364\n-366 void Next() {\n- 367 int stride = *this->_size;\n- 368 ++this->_size;\n- 369 this->_indices += stride;\n- 370 this->_weights += stride;\n- 371 if (_duWeights) _duWeights += stride;\n- 372 if (_dvWeights) _dvWeights += stride;\n- 373 if (_duuWeights) _duuWeights += stride;\n- 374 if (_duvWeights) _duvWeights += stride;\n- 375 if (_dvvWeights) _dvvWeights += stride;\n+ 364}\n+ 365\n+ 366template \n+ 367template \n+ 368inline void\n+369PrimvarRefinerReal::LimitFaceVarying(T const & src, U & dst, int\n+channel) const {\n+ 370\n+ 371 if (_refiner.getLevel(_refiner.GetMaxLevel()).getNumVertexEdgesTotal() ==\n+0) {\n+ 372 Error(FAR_RUNTIME_ERROR,\n+ 373 \"Failure in PrimvarRefiner::LimitFaceVarying() -- \"\n+ 374 \"last level of refinement does not include full topology.\");\n+ 375 return;\n 376 }\n 377\n- 378private:\n- 379\n- 380 friend class StencilTableFactoryReal;\n- 381 friend class LimitStencilTableFactoryReal;\n- 382\n- 383 REAL * _duWeights, // pointer to stencil u derivative limit weights\n- 384 * _dvWeights, // pointer to stencil v derivative limit weights\n- 385 * _duuWeights, // pointer to stencil uu derivative limit weights\n- 386 * _duvWeights, // pointer to stencil uv derivative limit weights\n- 387 * _dvvWeights; // pointer to stencil vv derivative limit weights\n- 388};\n- 389\n-392class LimitStencil : public LimitStencilReal {\n- 393protected:\n-394 typedef LimitStencilReal BaseStencil;\n+ 378 switch (_refiner._subdivType) {\n+ 379 case Sdc::SCHEME_CATMARK:\n+ 380 limitFVar(src, dst, channel);\n+ 381 break;\n+ 382 case Sdc::SCHEME_LOOP:\n+ 383 limitFVar(src, dst, channel);\n+ 384 break;\n+ 385 case Sdc::SCHEME_BILINEAR:\n+ 386 limitFVar(src, dst, channel);\n+ 387 break;\n+ 388 }\n+ 389}\n+ 390\n+ 391template \n+ 392template \n+ 393inline void\n+394PrimvarRefinerReal::InterpolateFaceUniform(int level, T const & src, U\n+& dst) const {\n 395\n- 396public:\n-397 LimitStencil(BaseStencil const & other) : BaseStencil(other) { }\n-398 LimitStencil(int* size, Index * indices, float * weights,\n- 399 float * duWeights=0, float * dvWeights=0,\n- 400 float * duuWeights=0, float * duvWeights=0, float * dvvWeights=0)\n- 401 : BaseStencil(size, indices, weights,\n- 402 duWeights, dvWeights, duuWeights, duvWeights, dvvWeights) { }\n- 403};\n+ 396 assert(level>0 && level<=(int)_refiner._refinements.size());\n+ 397\n+ 398 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n+(level-1);\n+ 399 Vtr::internal::Level const & child = refinement.child();\n+ 400\n+ 401 for (int cFace = 0; cFace < child.getNumFaces(); ++cFace) {\n+ 402\n+ 403 Vtr::Index pFace = refinement.getChildFaceParentFace(cFace);\n 404\n- 405\n- 408template \n-409class LimitStencilTableReal : public StencilTableReal {\n- 410protected:\n-411 LimitStencilTableReal(\n- 412 int numControlVerts,\n- 413 std::vector const& offsets,\n- 414 std::vector const& sizes,\n- 415 std::vector const& sources,\n- 416 std::vector const& weights,\n- 417 std::vector const& duWeights,\n- 418 std::vector const& dvWeights,\n- 419 std::vector const& duuWeights,\n- 420 std::vector const& duvWeights,\n- 421 std::vector const& dvvWeights,\n- 422 bool includeCoarseVerts,\n- 423 size_t firstOffset);\n+ 405 dst[cFace] = src[pFace];\n+ 406 }\n+ 407}\n+ 408\n+ 409template \n+ 410template \n+ 411inline void\n+412PrimvarRefinerReal::InterpolateVarying(int level, T const & src, U &\n+dst) const {\n+ 413\n+ 414 assert(level>0 && level<=(int)_refiner._refinements.size());\n+ 415\n+ 416 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n+(level-1);\n+ 417 Vtr::internal::Level const & parent = refinement.parent();\n+ 418\n+ 419 //\n+ 420 // Group values to interpolate based on origin -- note that there may\n+ 421 // be none originating from faces:\n+ 422 //\n+ 423 if (refinement.getNumChildVerticesFromFaces() > 0) {\n 424\n- 425public:\n+ 425 for (int face = 0; face < parent.getNumFaces(); ++face) {\n 426\n-428 LimitStencilReal GetLimitStencil(Index i) const;\n+ 427 Vtr::Index cVert = refinement.getFaceChildVertex(face);\n+ 428 if (Vtr::IndexIsValid(cVert)) {\n 429\n-431 LimitStencilReal operator[](Index index) const;\n+ 430 // Apply the weights to the parent face's vertices:\n+ 431 ConstIndexArray fVerts = parent.getFaceVertices(face);\n 432\n-434 std::vector const & GetDuWeights() const {\n- 435 return _duWeights;\n- 436 }\n- 437\n-439 std::vector const & GetDvWeights() const {\n- 440 return _dvWeights;\n+ 433 Weight fVaryingWeight = 1.0f / (Weight) fVerts.size();\n+ 434\n+ 435 dst[cVert].Clear();\n+ 436 for (int i = 0; i < fVerts.size(); ++i) {\n+ 437 dst[cVert].AddWithWeight(src[fVerts[i]], fVaryingWeight);\n+ 438 }\n+ 439 }\n+ 440 }\n 441 }\n- 442\n-444 std::vector const & GetDuuWeights() const {\n- 445 return _duuWeights;\n- 446 }\n- 447\n-449 std::vector const & GetDuvWeights() const {\n- 450 return _duvWeights;\n- 451 }\n- 452\n-454 std::vector const & GetDvvWeights() const {\n- 455 return _dvvWeights;\n- 456 }\n- 457\n- 475 template \n-476 void UpdateDerivs(T const & srcValues, U & uderivs, U & vderivs,\n- 477 int start=-1, int end=-1) const {\n- 478\n- 479 this->update(srcValues, uderivs, _duWeights, start, end);\n- 480 this->update(srcValues, vderivs, _dvWeights, start, end);\n- 481 }\n- 482\n- 483 template \n-484 void UpdateDerivs(T1 const & srcBase, int numBase, T2 const & srcRef,\n- 485 U & uderivs, U & vderivs, int start=-1, int end=-1) const {\n- 486\n- 487 this->update(srcBase, numBase, srcRef, uderivs, _duWeights, start, end);\n- 488 this->update(srcBase, numBase, srcRef, vderivs, _dvWeights, start, end);\n- 489 }\n- 490\n- 491 // Pointer interface for backward compatibility\n- 492 template \n-493 void UpdateDerivs(T const *src, U *uderivs, U *vderivs,\n- 494 int start=-1, int end=-1) const {\n- 495\n- 496 this->update(src, uderivs, _duWeights, start, end);\n- 497 this->update(src, vderivs, _dvWeights, start, end);\n- 498 }\n- 499 template \n-500 void UpdateDerivs(T1 const *srcBase, int numBase, T2 const *srcRef,\n- 501 U *uderivs, U *vderivs, int start=-1, int end=-1) const {\n+ 442 for (int edge = 0; edge < parent.getNumEdges(); ++edge) {\n+ 443\n+ 444 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);\n+ 445 if (Vtr::IndexIsValid(cVert)) {\n+ 446\n+ 447 // Apply the weights to the parent edges's vertices\n+ 448 ConstIndexArray eVerts = parent.getEdgeVertices(edge);\n+ 449\n+ 450 dst[cVert].Clear();\n+ 451 dst[cVert].AddWithWeight(src[eVerts[0]], 0.5f);\n+ 452 dst[cVert].AddWithWeight(src[eVerts[1]], 0.5f);\n+ 453 }\n+ 454 }\n+ 455 for (int vert = 0; vert < parent.getNumVertices(); ++vert) {\n+ 456\n+ 457 Vtr::Index cVert = refinement.getVertexChildVertex(vert);\n+ 458 if (Vtr::IndexIsValid(cVert)) {\n+ 459\n+ 460 // Essentially copy the parent vertex:\n+ 461 dst[cVert].Clear();\n+ 462 dst[cVert].AddWithWeight(src[vert], 1.0f);\n+ 463 }\n+ 464 }\n+ 465}\n+ 466\n+ 467\n+ 468//\n+ 469// Internal implementation methods -- grouping vertices to be interpolated\n+ 470// based on the type of parent component from which they originated:\n+ 471//\n+ 472template \n+ 473template \n+ 474inline void\n+ 475PrimvarRefinerReal::interpFromFaces(int level, T const & src, U &\n+dst) const {\n+ 476\n+ 477 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n+(level-1);\n+ 478 Vtr::internal::Level const & parent = refinement.parent();\n+ 479\n+ 480 if (refinement.getNumChildVerticesFromFaces() == 0) return;\n+ 481\n+ 482 Sdc::Scheme scheme(_refiner._subdivOptions);\n+ 483\n+ 484 Vtr::internal::StackBuffer fVertWeights(parent.getMaxValence\n+());\n+ 485\n+ 486 for (int face = 0; face < parent.getNumFaces(); ++face) {\n+ 487\n+ 488 Vtr::Index cVert = refinement.getFaceChildVertex(face);\n+ 489 if (!Vtr::IndexIsValid(cVert))\n+ 490 continue;\n+ 491\n+ 492 // Declare and compute mask weights for this vertex relative to its parent\n+face:\n+ 493 ConstIndexArray fVerts = parent.getFaceVertices(face);\n+ 494\n+ 495 Mask fMask(fVertWeights, 0, 0);\n+ 496 Vtr::internal::FaceInterface fHood(fVerts.size());\n+ 497\n+ 498 scheme.ComputeFaceVertexMask(fHood, fMask);\n+ 499\n+ 500 // Apply the weights to the parent face's vertices:\n+ 501 dst[cVert].Clear();\n 502\n- 503 this->update(srcBase, numBase, srcRef, uderivs, _duWeights, start, end);\n- 504 this->update(srcBase, numBase, srcRef, vderivs, _dvWeights, start, end);\n- 505 }\n- 506\n- 527 template \n-528 void Update2ndDerivs(T const & srcValues,\n- 529 U & uuderivs, U & uvderivs, U & vvderivs,\n- 530 int start=-1, int end=-1) const {\n+ 503 for (int i = 0; i < fVerts.size(); ++i) {\n+ 504\n+ 505 dst[cVert].AddWithWeight(src[fVerts[i]], fVertWeights[i]);\n+ 506 }\n+ 507 }\n+ 508}\n+ 509\n+ 510template \n+ 511template \n+ 512inline void\n+ 513PrimvarRefinerReal::interpFromEdges(int level, T const & src, U &\n+dst) const {\n+ 514\n+ 515 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n+(level-1);\n+ 516 Vtr::internal::Level const & parent = refinement.parent();\n+ 517 Vtr::internal::Level const & child = refinement.child();\n+ 518\n+ 519 Sdc::Scheme scheme(_refiner._subdivOptions);\n+ 520\n+ 521 Vtr::internal::EdgeInterface eHood(parent);\n+ 522\n+ 523 Weight eVertWeights[2];\n+ 524 Vtr::internal::StackBuffer eFaceWeights(parent.getMaxEdgeFaces\n+());\n+ 525\n+ 526 for (int edge = 0; edge < parent.getNumEdges(); ++edge) {\n+ 527\n+ 528 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);\n+ 529 if (!Vtr::IndexIsValid(cVert))\n+ 530 continue;\n 531\n- 532 this->update(srcValues, uuderivs, _duuWeights, start, end);\n- 533 this->update(srcValues, uvderivs, _duvWeights, start, end);\n- 534 this->update(srcValues, vvderivs, _dvvWeights, start, end);\n- 535 }\n- 536\n- 537 template \n-538 void Update2ndDerivs(T1 const & srcBase, int numBase, T2 const & srcRef,\n- 539 U & uuderivs, U & uvderivs, U & vvderivs, int start=-1, int end=-1) const\n-{\n- 540\n- 541 this->update(srcBase, numBase, srcRef, uuderivs, _duuWeights, start, end);\n- 542 this->update(srcBase, numBase, srcRef, uvderivs, _duvWeights, start, end);\n- 543 this->update(srcBase, numBase, srcRef, vvderivs, _dvvWeights, start, end);\n- 544 }\n- 545\n- 546 // Pointer interface for backward compatibility\n- 547 template \n-548 void Update2ndDerivs(T const *src, T *uuderivs, U *uvderivs, U *vvderivs,\n- 549 int start=-1, int end=-1) const {\n+ 532 // Declare and compute mask weights for this vertex relative to its parent\n+edge:\n+ 533 ConstIndexArray eVerts = parent.getEdgeVertices(edge),\n+ 534 eFaces = parent.getEdgeFaces(edge);\n+ 535\n+ 536 Mask eMask(eVertWeights, 0, eFaceWeights);\n+ 537\n+ 538 eHood.SetIndex(edge);\n+ 539\n+ 540 Sdc::Crease::Rule pRule = (parent.getEdgeSharpness(edge) > 0.0f) ? Sdc::\n+Crease::RULE_CREASE : Sdc::Crease::RULE_SMOOTH;\n+ 541 Sdc::Crease::Rule cRule = child.getVertexRule(cVert);\n+ 542\n+ 543 scheme.ComputeEdgeVertexMask(eHood, eMask, pRule, cRule);\n+ 544\n+ 545 // Apply the weights to the parent edges's vertices and (if applicable) to\n+ 546 // the child vertices of its incident faces:\n+ 547 dst[cVert].Clear();\n+ 548 dst[cVert].AddWithWeight(src[eVerts[0]], eVertWeights[0]);\n+ 549 dst[cVert].AddWithWeight(src[eVerts[1]], eVertWeights[1]);\n 550\n- 551 this->update(src, uuderivs, _duuWeights, start, end);\n- 552 this->update(src, uvderivs, _duvWeights, start, end);\n- 553 this->update(src, vvderivs, _dvvWeights, start, end);\n- 554 }\n- 555 template \n-556 void Update2ndDerivs(T1 const *srcBase, int numBase, T2 const *srcRef,\n- 557 U *uuderivs, U *uvderivs, U *vvderivs, int start=-1, int end=-1) const {\n+ 551 if (eMask.GetNumFaceWeights() > 0) {\n+ 552\n+ 553 for (int i = 0; i < eFaces.size(); ++i) {\n+ 554\n+ 555 if (eMask.AreFaceWeightsForFaceCenters()) {\n+ 556 assert(refinement.getNumChildVerticesFromFaces() > 0);\n+ 557 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(eFaces[i]);\n 558\n- 559 this->update(srcBase, numBase, srcRef, uuderivs, _duuWeights, start, end);\n- 560 this->update(srcBase, numBase, srcRef, uvderivs, _duvWeights, start, end);\n- 561 this->update(srcBase, numBase, srcRef, vvderivs, _dvvWeights, start, end);\n- 562 }\n- 563\n-565 void Clear();\n- 566\n- 567private:\n- 568 friend class LimitStencilTableFactoryReal;\n- 569\n- 570 // Resize the table arrays (factory helper)\n- 571 void resize(int nstencils, int nelems);\n- 572\n- 573private:\n- 574 std::vector _duWeights, // u derivative limit stencil weights\n- 575 _dvWeights, // v derivative limit stencil weights\n- 576 _duuWeights, // uu derivative limit stencil weights\n- 577 _duvWeights, // uv derivative limit stencil weights\n- 578 _dvvWeights; // vv derivative limit stencil weights\n- 579};\n- 580\n-583class LimitStencilTable : public LimitStencilTableReal {\n- 584protected:\n-585 typedef LimitStencilTableReal BaseTable;\n- 586\n- 587public:\n-588 LimitStencil GetLimitStencil(Index index) const {\n- 589 return LimitStencil(BaseTable::GetLimitStencil(index));\n- 590 }\n-591 LimitStencil operator[](Index index) const {\n- 592 return LimitStencil(BaseTable::GetLimitStencil(index));\n- 593 }\n+ 559 assert(Vtr::IndexIsValid(cVertOfFace));\n+ 560 dst[cVert].AddWithWeight(dst[cVertOfFace], eFaceWeights[i]);\n+ 561 } else {\n+ 562 Vtr::Index pFace = eFaces[i];\n+ 563 ConstIndexArray pFaceEdges = parent.getFaceEdges(pFace),\n+ 564 pFaceVerts = parent.getFaceVertices(pFace);\n+ 565\n+ 566 int eInFace = 0;\n+ 567 for ( ; pFaceEdges[eInFace] != edge; ++eInFace ) ;\n+ 568\n+ 569 int vInFace = eInFace + 2;\n+ 570 if (vInFace >= pFaceVerts.size()) vInFace -= pFaceVerts.size();\n+ 571\n+ 572 Vtr::Index pVertNext = pFaceVerts[vInFace];\n+ 573 dst[cVert].AddWithWeight(src[pVertNext], eFaceWeights[i]);\n+ 574 }\n+ 575 }\n+ 576 }\n+ 577 }\n+ 578}\n+ 579\n+ 580template \n+ 581template \n+ 582inline void\n+ 583PrimvarRefinerReal::interpFromVerts(int level, T const & src, U &\n+dst) const {\n+ 584\n+ 585 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n+(level-1);\n+ 586 Vtr::internal::Level const & parent = refinement.parent();\n+ 587 Vtr::internal::Level const & child = refinement.child();\n+ 588\n+ 589 Sdc::Scheme scheme(_refiner._subdivOptions);\n+ 590\n+ 591 Vtr::internal::VertexInterface vHood(parent, child);\n+ 592\n+ 593 Vtr::internal::StackBuffer weightBuffer(2*parent.getMaxValence\n+());\n 594\n- 595protected:\n-596 LimitStencilTable(int numControlVerts,\n- 597 std::vector const& offsets,\n- 598 std::vector const& sizes,\n- 599 std::vector const& sources,\n- 600 std::vector const& weights,\n- 601 std::vector const& duWeights,\n- 602 std::vector const& dvWeights,\n- 603 std::vector const& duuWeights,\n- 604 std::vector const& duvWeights,\n- 605 std::vector const& dvvWeights,\n- 606 bool includeCoarseVerts,\n- 607 size_t firstOffset)\n- 608 : BaseTable(numControlVerts,\n- 609 offsets, sizes, sources, weights,\n- 610 duWeights, dvWeights, duuWeights, duvWeights, dvvWeights,\n- 611 includeCoarseVerts, firstOffset) { }\n- 612};\n- 613\n- 614\n- 615// Update values by applying cached stencil weights to new control values\n- 616template \n- 617template void\n-618StencilTableReal::update(T1 const &srcBase, int numBase,\n- 619 T2 const &srcRef, U &dstValues,\n- 620 std::vector const &valueWeights, Index start, Index end) const {\n- 621\n- 622 int const * sizes = &_sizes.at(0);\n- 623 Index const * indices = &_indices.at(0);\n- 624 REAL const * weights = &valueWeights.at(0);\n+ 595 for (int vert = 0; vert < parent.getNumVertices(); ++vert) {\n+ 596\n+ 597 Vtr::Index cVert = refinement.getVertexChildVertex(vert);\n+ 598 if (!Vtr::IndexIsValid(cVert))\n+ 599 continue;\n+ 600\n+ 601 // Declare and compute mask weights for this vertex relative to its parent\n+edge:\n+ 602 ConstIndexArray vEdges = parent.getVertexEdges(vert),\n+ 603 vFaces = parent.getVertexFaces(vert);\n+ 604\n+ 605 Weight vVertWeight,\n+ 606 * vEdgeWeights = weightBuffer,\n+ 607 * vFaceWeights = vEdgeWeights + vEdges.size();\n+ 608\n+ 609 Mask vMask(&vVertWeight, vEdgeWeights, vFaceWeights);\n+ 610\n+ 611 vHood.SetIndex(vert, cVert);\n+ 612\n+ 613 Sdc::Crease::Rule pRule = parent.getVertexRule(vert);\n+ 614 Sdc::Crease::Rule cRule = child.getVertexRule(cVert);\n+ 615\n+ 616 scheme.ComputeVertexVertexMask(vHood, vMask, pRule, cRule);\n+ 617\n+ 618 // Apply the weights to the parent vertex, the vertices opposite its\n+incident\n+ 619 // edges, and the child vertices of its incident faces:\n+ 620 //\n+ 621 // In order to improve numerical precision, it's better to apply smaller\n+weights\n+ 622 // first, so begin with the face-weights followed by the edge-weights and\n+the\n+ 623 // vertex weight last.\n+ 624 dst[cVert].Clear();\n 625\n- 626 if (start > 0) {\n- 627 assert(start < (Index)_offsets.size());\n- 628 sizes += start;\n- 629 indices += _offsets[start];\n- 630 weights += _offsets[start];\n- 631 } else {\n- 632 start = 0;\n- 633 }\n- 634\n- 635 int nstencils = ((end < start) ? GetNumStencils() : end) - start;\n- 636\n- 637 for (int i = 0; i < nstencils; ++i, ++sizes) {\n- 638 dstValues[start + i].Clear();\n- 639 for (int j = 0; j < *sizes; ++j, ++indices, ++weights) {\n- 640 if (*indices < numBase) {\n- 641 dstValues[start + i].AddWithWeight(srcBase[*indices], *weights);\n- 642 } else {\n- 643 dstValues[start + i].AddWithWeight(srcRef[*indices - numBase], *weights);\n+ 626 if (vMask.GetNumFaceWeights() > 0) {\n+ 627 assert(vMask.AreFaceWeightsForFaceCenters());\n+ 628\n+ 629 for (int i = 0; i < vFaces.size(); ++i) {\n+ 630\n+ 631 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(vFaces[i]);\n+ 632 assert(Vtr::IndexIsValid(cVertOfFace));\n+ 633 dst[cVert].AddWithWeight(dst[cVertOfFace], vFaceWeights[i]);\n+ 634 }\n+ 635 }\n+ 636 if (vMask.GetNumEdgeWeights() > 0) {\n+ 637\n+ 638 for (int i = 0; i < vEdges.size(); ++i) {\n+ 639\n+ 640 ConstIndexArray eVerts = parent.getEdgeVertices(vEdges[i]);\n+ 641 Vtr::Index pVertOppositeEdge = (eVerts[0] == vert) ? eVerts[1] : eVerts\n+[0];\n+ 642\n+ 643 dst[cVert].AddWithWeight(src[pVertOppositeEdge], vEdgeWeights[i]);\n 644 }\n 645 }\n- 646 }\n- 647}\n- 648template \n- 649template void\n-650StencilTableReal::update(T const &srcValues, U &dstValues,\n- 651 std::vector const &valueWeights, Index start, Index end) const {\n- 652\n- 653 int const * sizes = &_sizes.at(0);\n- 654 Index const * indices = &_indices.at(0);\n- 655 REAL const * weights = &valueWeights.at(0);\n- 656\n- 657 if (start > 0) {\n- 658 assert(start < (Index)_offsets.size());\n- 659 sizes += start;\n- 660 indices += _offsets[start];\n- 661 weights += _offsets[start];\n- 662 } else {\n- 663 start = 0;\n- 664 }\n- 665\n- 666 int nstencils = ((end < start) ? GetNumStencils() : end) - start;\n+ 646 dst[cVert].AddWithWeight(src[vert], vVertWeight);\n+ 647 }\n+ 648}\n+ 649\n+ 650\n+ 651//\n+ 652// Internal face-varying implementation details:\n+ 653//\n+ 654template \n+ 655template \n+ 656inline void\n+ 657PrimvarRefinerReal::interpFVarFromFaces(int level, T const & src, U &\n+dst, int channel) const {\n+ 658\n+ 659 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n+(level-1);\n+ 660\n+ 661 if (refinement.getNumChildVerticesFromFaces() == 0) return;\n+ 662\n+ 663 Sdc::Scheme scheme(_refiner._subdivOptions);\n+ 664\n+ 665 Vtr::internal::Level const & parentLevel = refinement.parent();\n+ 666 Vtr::internal::Level const & childLevel = refinement.child();\n 667\n- 668 for (int i = 0; i < nstencils; ++i, ++sizes) {\n- 669 dstValues[start + i].Clear();\n- 670 for (int j = 0; j < *sizes; ++j, ++indices, ++weights) {\n- 671 dstValues[start + i].AddWithWeight(srcValues[*indices], *weights);\n- 672 }\n- 673 }\n- 674}\n- 675\n- 676template \n- 677inline void\n-678StencilTableReal::generateOffsets() {\n- 679 Index offset=0;\n- 680 int noffsets = (int)_sizes.size();\n- 681 _offsets.resize(noffsets);\n- 682 for (int i=0; i<(int)_sizes.size(); ++i ) {\n- 683 _offsets[i]=offset;\n- 684 offset+=_sizes[i];\n- 685 }\n- 686}\n- 687\n- 688template \n- 689inline void\n-690StencilTableReal::resize(int nstencils, int nelems) {\n- 691 _sizes.resize(nstencils);\n- 692 _indices.resize(nelems);\n- 693 _weights.resize(nelems);\n- 694}\n- 695\n- 696template \n- 697inline void\n-698StencilTableReal::reserve(int nstencils, int nelems) {\n- 699 _sizes.reserve(nstencils);\n- 700 _indices.reserve(nelems);\n- 701 _weights.reserve(nelems);\n- 702}\n- 703\n- 704template \n+ 668 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel\n+(channel);\n+ 669 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel\n+(channel);\n+ 670\n+ 671 Vtr::internal::StackBuffer fValueWeights\n+(parentLevel.getMaxValence());\n+ 672\n+ 673 for (int face = 0; face < parentLevel.getNumFaces(); ++face) {\n+ 674\n+ 675 Vtr::Index cVert = refinement.getFaceChildVertex(face);\n+ 676 if (!Vtr::IndexIsValid(cVert))\n+ 677 continue;\n+ 678\n+ 679 Vtr::Index cVertValue = childFVar.getVertexValueOffset(cVert);\n+ 680\n+ 681 // The only difference for face-varying here is that we get the values\n+associated\n+ 682 // with each face-vertex directly from the FVarLevel, rather than using\n+the parent\n+ 683 // face-vertices directly. If any face-vertex has any sibling values, then\n+we may\n+ 684 // get the wrong one using the face-vertex index directly.\n+ 685\n+ 686 // Declare and compute mask weights for this vertex relative to its parent\n+face:\n+ 687 ConstIndexArray fValues = parentFVar.getFaceValues(face);\n+ 688\n+ 689 Mask fMask(fValueWeights, 0, 0);\n+ 690 Vtr::internal::FaceInterface fHood(fValues.size());\n+ 691\n+ 692 scheme.ComputeFaceVertexMask(fHood, fMask);\n+ 693\n+ 694 // Apply the weights to the parent face's vertices:\n+ 695 dst[cVertValue].Clear();\n+ 696\n+ 697 for (int i = 0; i < fValues.size(); ++i) {\n+ 698 dst[cVertValue].AddWithWeight(src[fValues[i]], fValueWeights[i]);\n+ 699 }\n+ 700 }\n+ 701}\n+ 702\n+ 703template \n+ 704template \n 705inline void\n-706StencilTableReal::shrinkToFit() {\n- 707 std::vector(_sizes).swap(_sizes);\n- 708 std::vector(_indices).swap(_indices);\n- 709 std::vector(_weights).swap(_weights);\n- 710}\n+ 706PrimvarRefinerReal::interpFVarFromEdges(int level, T const & src, U &\n+dst, int channel) const {\n+ 707\n+ 708 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n+(level-1);\n+ 709\n+ 710 Sdc::Scheme scheme(_refiner._subdivOptions);\n 711\n- 712template \n- 713inline void\n-714StencilTableReal::finalize() {\n- 715 shrinkToFit();\n- 716 generateOffsets();\n- 717}\n+ 712 Vtr::internal::Level const & parentLevel = refinement.parent();\n+ 713 Vtr::internal::Level const & childLevel = refinement.child();\n+ 714\n+ 715 Vtr::internal::FVarRefinement const & refineFVar =\n+refinement.getFVarRefinement(channel);\n+ 716 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel\n+(channel);\n+ 717 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel\n+(channel);\n 718\n- 719// Returns a Stencil at index i in the table\n- 720template \n- 721inline StencilReal\n-722StencilTableReal::GetStencil(Index i) const {\n- 723 assert((! _offsets.empty()) && i<(int)_offsets.size());\n- 724\n- 725 Index ofs = _offsets[i];\n- 726\n- 727 return StencilReal(const_cast(&_sizes[i]),\n- 728 const_cast(&_indices[ofs]),\n- 729 const_cast(&_weights[ofs]));\n- 730}\n- 731\n- 732template \n- 733inline StencilReal\n-734StencilTableReal::operator[](Index index) const {\n- 735 return GetStencil(index);\n- 736}\n+ 719 //\n+ 720 // Allocate and initialize (if linearly interpolated) interpolation\n+weights for\n+ 721 // the edge mask:\n+ 722 //\n+ 723 Weight eVertWeights[2];\n+ 724 Vtr::internal::StackBuffer eFaceWeights\n+(parentLevel.getMaxEdgeFaces());\n+ 725\n+ 726 Mask eMask(eVertWeights, 0, eFaceWeights);\n+ 727\n+ 728 bool isLinearFVar = parentFVar.isLinear() || (_refiner._subdivType ==\n+Sdc::SCHEME_BILINEAR);\n+ 729 if (isLinearFVar) {\n+ 730 eMask.SetNumVertexWeights(2);\n+ 731 eMask.SetNumEdgeWeights(0);\n+ 732 eMask.SetNumFaceWeights(0);\n+ 733\n+ 734 eVertWeights[0] = 0.5f;\n+ 735 eVertWeights[1] = 0.5f;\n+ 736 }\n 737\n- 738template \n- 739inline void\n- 740LimitStencilTableReal::resize(int nstencils, int nelems) {\n- 741 StencilTableReal::resize(nstencils, nelems);\n- 742 _duWeights.resize(nelems);\n- 743 _dvWeights.resize(nelems);\n- 744}\n+ 738 Vtr::internal::EdgeInterface eHood(parentLevel);\n+ 739\n+ 740 for (int edge = 0; edge < parentLevel.getNumEdges(); ++edge) {\n+ 741\n+ 742 Vtr::Index cVert = refinement.getEdgeChildVertex(edge);\n+ 743 if (!Vtr::IndexIsValid(cVert))\n+ 744 continue;\n 745\n- 746// Returns a LimitStencil at index i in the table\n- 747template \n- 748inline LimitStencilReal\n-749LimitStencilTableReal::GetLimitStencil(Index i) const {\n- 750 assert((! this->GetOffsets().empty()) && i<(int)this->GetOffsets().size\n-());\n- 751\n- 752 Index ofs = this->GetOffsets()[i];\n- 753\n- 754 if (!_duWeights.empty() && !_dvWeights.empty() &&\n- 755 !_duuWeights.empty() && !_duvWeights.empty() && !_dvvWeights.empty()) {\n- 756 return LimitStencilReal(\n- 757 const_cast(&this->GetSizes()[i]),\n- 758 const_cast(&this->GetControlIndices()[ofs]),\n- 759 const_cast(&this->GetWeights()[ofs]),\n- 760 const_cast(&GetDuWeights()[ofs]),\n- 761 const_cast(&GetDvWeights()[ofs]),\n- 762 const_cast(&GetDuuWeights()[ofs]),\n- 763 const_cast(&GetDuvWeights()[ofs]),\n- 764 const_cast(&GetDvvWeights()[ofs]) );\n- 765 } else if (!_duWeights.empty() && !_dvWeights.empty()) {\n- 766 return LimitStencilReal(\n- 767 const_cast(&this->GetSizes()[i]),\n- 768 const_cast(&this->GetControlIndices()[ofs]),\n- 769 const_cast(&this->GetWeights()[ofs]),\n- 770 const_cast(&GetDuWeights()[ofs]),\n- 771 const_cast(&GetDvWeights()[ofs]) );\n- 772 } else {\n- 773 return LimitStencilReal(\n- 774 const_cast(&this->GetSizes()[i]),\n- 775 const_cast(&this->GetControlIndices()[ofs]),\n- 776 const_cast(&this->GetWeights()[ofs]) );\n- 777 }\n- 778}\n- 779\n- 780template \n- 781inline LimitStencilReal\n-782LimitStencilTableReal::operator[](Index index) const {\n- 783 return GetLimitStencil(index);\n- 784}\n- 785\n- 786} // end namespace Far\n- 787\n- 788} // end namespace OPENSUBDIV_VERSION\n- 789using namespace OPENSUBDIV_VERSION;\n- 790\n- 791} // end namespace OpenSubdiv\n+ 746 ConstIndexArray cVertValues = childFVar.getVertexValues(cVert);\n+ 747\n+ 748 bool fvarEdgeVertMatchesVertex = childFVar.valueTopologyMatches\n+(cVertValues[0]);\n+ 749 if (fvarEdgeVertMatchesVertex) {\n+ 750 //\n+ 751 // If smoothly interpolated, compute new weights for the edge mask:\n+ 752 //\n+ 753 if (!isLinearFVar) {\n+ 754 eHood.SetIndex(edge);\n+ 755\n+ 756 Sdc::Crease::Rule pRule = (parentLevel.getEdgeSharpness(edge) > 0.0f)\n+ 757 ? Sdc::Crease::RULE_CREASE : Sdc::Crease::RULE_SMOOTH;\n+ 758 Sdc::Crease::Rule cRule = childLevel.getVertexRule(cVert);\n+ 759\n+ 760 scheme.ComputeEdgeVertexMask(eHood, eMask, pRule, cRule);\n+ 761 }\n+ 762\n+ 763 // Apply the weights to the parent edge's vertices and (if applicable) to\n+ 764 // the child vertices of its incident faces:\n+ 765 //\n+ 766 // Even though the face-varying topology matches the vertex topology, we\n+need\n+ 767 // to be careful here when getting values corresponding to the two end-\n+vertices.\n+ 768 // While the edge may be continuous, the vertices at their ends may have\n+ 769 // discontinuities elsewhere in their neighborhood (i.e. on the \"other\n+side\"\n+ 770 // of the end-vertex) and so have sibling values associated with them. In\n+most\n+ 771 // cases the topology for an end-vertex will match and we can use it\n+directly,\n+ 772 // but we must still check and retrieve as needed.\n+ 773 //\n+ 774 // Indices for values corresponding to face-vertices are guaranteed to\n+match,\n+ 775 // so we can use the child-vertex indices directly.\n+ 776 //\n+ 777 // And by \"directly\", we always use getVertexValue(vertexIndex) to\n+reference\n+ 778 // values in the \"src\" to account for the possible indirection that may\n+exist at\n+ 779 // level 0 -- where there may be fewer values than vertices and an\n+additional\n+ 780 // indirection is necessary. We can use a vertex index directly for \"dst\"\n+when\n+ 781 // it matches.\n+ 782 //\n+ 783 Vtr::Index eVertValues[2];\n+ 784\n+ 785 parentFVar.getEdgeFaceValues(edge, 0, eVertValues);\n+ 786\n+ 787 Index cVertValue = cVertValues[0];\n+ 788\n+ 789 dst[cVertValue].Clear();\n+ 790 dst[cVertValue].AddWithWeight(src[eVertValues[0]], eVertWeights[0]);\n+ 791 dst[cVertValue].AddWithWeight(src[eVertValues[1]], eVertWeights[1]);\n 792\n- 793#endif // OPENSUBDIV3_FAR_STENCILTABLE_H\n+ 793 if (eMask.GetNumFaceWeights() > 0) {\n+ 794\n+ 795 ConstIndexArray eFaces = parentLevel.getEdgeFaces(edge);\n+ 796\n+ 797 for (int i = 0; i < eFaces.size(); ++i) {\n+ 798 if (eMask.AreFaceWeightsForFaceCenters()) {\n+ 799\n+ 800 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(eFaces[i]);\n+ 801 assert(Vtr::IndexIsValid(cVertOfFace));\n+ 802\n+ 803 Vtr::Index cValueOfFace = childFVar.getVertexValueOffset(cVertOfFace);\n+ 804 dst[cVertValue].AddWithWeight(dst[cValueOfFace], eFaceWeights[i]);\n+ 805 } else {\n+ 806 Vtr::Index pFace = eFaces[i];\n+ 807 ConstIndexArray pFaceEdges = parentLevel.getFaceEdges(pFace),\n+ 808 pFaceVerts = parentLevel.getFaceVertices(pFace);\n+ 809\n+ 810 int eInFace = 0;\n+ 811 for ( ; pFaceEdges[eInFace] != edge; ++eInFace ) ;\n+ 812\n+ 813 // Edge \"i\" spans vertices [i,i+1] so we want i+2...\n+ 814 int vInFace = eInFace + 2;\n+ 815 if (vInFace >= pFaceVerts.size()) vInFace -= pFaceVerts.size();\n+ 816\n+ 817 Vtr::Index pValueNext = parentFVar.getFaceValues(pFace)[vInFace];\n+ 818 dst[cVertValue].AddWithWeight(src[pValueNext], eFaceWeights[i]);\n+ 819 }\n+ 820 }\n+ 821 }\n+ 822 } else {\n+ 823 //\n+ 824 // Mismatched edge-verts should just be linearly interpolated between the\n+pairs of\n+ 825 // values for each sibling of the child edge-vertex -- the question is:\n+which face\n+ 826 // holds that pair of values for a given sibling?\n+ 827 //\n+ 828 // In the manifold case, the sibling and edge-face indices will\n+correspond. We\n+ 829 // will eventually need to update this to account for > 3 incident faces.\n+ 830 //\n+ 831 for (int i = 0; i < cVertValues.size(); ++i) {\n+ 832 Vtr::Index eVertValues[2];\n+ 833 int eFaceIndex = refineFVar.getChildValueParentSource(cVert, i);\n+ 834 assert(eFaceIndex == i);\n+ 835\n+ 836 parentFVar.getEdgeFaceValues(edge, eFaceIndex, eVertValues);\n+ 837\n+ 838 Index cVertValue = cVertValues[i];\n+ 839\n+ 840 dst[cVertValue].Clear();\n+ 841 dst[cVertValue].AddWithWeight(src[eVertValues[0]], 0.5);\n+ 842 dst[cVertValue].AddWithWeight(src[eVertValues[1]], 0.5);\n+ 843 }\n+ 844 }\n+ 845 }\n+ 846}\n+ 847\n+ 848template \n+ 849template \n+ 850inline void\n+ 851PrimvarRefinerReal::interpFVarFromVerts(int level, T const & src, U &\n+dst, int channel) const {\n+ 852\n+ 853 Vtr::internal::Refinement const & refinement = _refiner.getRefinement\n+(level-1);\n+ 854\n+ 855 Sdc::Scheme scheme(_refiner._subdivOptions);\n+ 856\n+ 857 Vtr::internal::Level const & parentLevel = refinement.parent();\n+ 858 Vtr::internal::Level const & childLevel = refinement.child();\n+ 859\n+ 860 Vtr::internal::FVarRefinement const & refineFVar =\n+refinement.getFVarRefinement(channel);\n+ 861 Vtr::internal::FVarLevel const & parentFVar = parentLevel.getFVarLevel\n+(channel);\n+ 862 Vtr::internal::FVarLevel const & childFVar = childLevel.getFVarLevel\n+(channel);\n+ 863\n+ 864 bool isLinearFVar = parentFVar.isLinear() || (_refiner._subdivType ==\n+Sdc::SCHEME_BILINEAR);\n+ 865\n+ 866 Vtr::internal::StackBuffer weightBuffer\n+(2*parentLevel.getMaxValence());\n+ 867\n+ 868 Vtr::internal::StackBuffer vEdgeValues\n+(parentLevel.getMaxValence());\n+ 869\n+ 870 Vtr::internal::VertexInterface vHood(parentLevel, childLevel);\n+ 871\n+ 872 for (int vert = 0; vert < parentLevel.getNumVertices(); ++vert) {\n+ 873\n+ 874 Vtr::Index cVert = refinement.getVertexChildVertex(vert);\n+ 875 if (!Vtr::IndexIsValid(cVert))\n+ 876 continue;\n+ 877\n+ 878 ConstIndexArray pVertValues = parentFVar.getVertexValues(vert),\n+ 879 cVertValues = childFVar.getVertexValues(cVert);\n+ 880\n+ 881 bool fvarVertVertMatchesVertex = childFVar.valueTopologyMatches\n+(cVertValues[0]);\n+ 882 if (isLinearFVar && fvarVertVertMatchesVertex) {\n+ 883 dst[cVertValues[0]].Clear();\n+ 884 dst[cVertValues[0]].AddWithWeight(src[pVertValues[0]], 1.0f);\n+ 885 continue;\n+ 886 }\n+ 887\n+ 888 if (fvarVertVertMatchesVertex) {\n+ 889 //\n+ 890 // Declare and compute mask weights for this vertex relative to its parent\n+edge:\n+ 891 //\n+ 892 // (We really need to encapsulate this somewhere else for use here and in\n+the\n+ 893 // general case)\n+ 894 //\n+ 895 ConstIndexArray vEdges = parentLevel.getVertexEdges(vert);\n+ 896\n+ 897 Weight vVertWeight;\n+ 898 Weight * vEdgeWeights = weightBuffer;\n+ 899 Weight * vFaceWeights = vEdgeWeights + vEdges.size();\n+ 900\n+ 901 Mask vMask(&vVertWeight, vEdgeWeights, vFaceWeights);\n+ 902\n+ 903 vHood.SetIndex(vert, cVert);\n+ 904\n+ 905 Sdc::Crease::Rule pRule = parentLevel.getVertexRule(vert);\n+ 906 Sdc::Crease::Rule cRule = childLevel.getVertexRule(cVert);\n+ 907\n+ 908 scheme.ComputeVertexVertexMask(vHood, vMask, pRule, cRule);\n+ 909\n+ 910 // Apply the weights to the parent vertex, the vertices opposite its\n+incident\n+ 911 // edges, and the child vertices of its incident faces:\n+ 912 //\n+ 913 // Even though the face-varying topology matches the vertex topology, we\n+need\n+ 914 // to be careful here when getting values corresponding to vertices at the\n+ 915 // ends of edges. While the edge may be continuous, the end vertex may\n+have\n+ 916 // discontinuities elsewhere in their neighborhood (i.e. on the \"other\n+side\"\n+ 917 // of the end-vertex) and so have sibling values associated with them. In\n+most\n+ 918 // cases the topology for an end-vertex will match and we can use it\n+directly,\n+ 919 // but we must still check and retrieve as needed.\n+ 920 //\n+ 921 // Indices for values corresponding to face-vertices are guaranteed to\n+match,\n+ 922 // so we can use the child-vertex indices directly.\n+ 923 //\n+ 924 // And by \"directly\", we always use getVertexValue(vertexIndex) to\n+reference\n+ 925 // values in the \"src\" to account for the possible indirection that may\n+exist at\n+ 926 // level 0 -- where there may be fewer values than vertices and an\n+additional\n+ 927 // indirection is necessary. We can use a vertex index directly for \"dst\"\n+when\n+ 928 // it matches.\n+ 929 //\n+ 930 // As with applying the mask to vertex data, in order to improve numerical\n+ 931 // precision, it's better to apply smaller weights first, so begin with\n+the\n+ 932 // face-weights followed by the edge-weights and the vertex weight last.\n+ 933 //\n+ 934 Vtr::Index pVertValue = pVertValues[0];\n+ 935 Vtr::Index cVertValue = cVertValues[0];\n+ 936\n+ 937 dst[cVertValue].Clear();\n+ 938 if (vMask.GetNumFaceWeights() > 0) {\n+ 939 assert(vMask.AreFaceWeightsForFaceCenters());\n+ 940\n+ 941 ConstIndexArray vFaces = parentLevel.getVertexFaces(vert);\n+ 942\n+ 943 for (int i = 0; i < vFaces.size(); ++i) {\n+ 944\n+ 945 Vtr::Index cVertOfFace = refinement.getFaceChildVertex(vFaces[i]);\n+ 946 assert(Vtr::IndexIsValid(cVertOfFace));\n+ 947\n+ 948 Vtr::Index cValueOfFace = childFVar.getVertexValueOffset(cVertOfFace);\n+ 949 dst[cVertValue].AddWithWeight(dst[cValueOfFace], vFaceWeights[i]);\n+ 950 }\n+ 951 }\n+ 952 if (vMask.GetNumEdgeWeights() > 0) {\n+ 953\n+ 954 parentFVar.getVertexEdgeValues(vert, vEdgeValues);\n+ 955\n+ 956 for (int i = 0; i < vEdges.size(); ++i) {\n+ 957 dst[cVertValue].AddWithWeight(src[vEdgeValues[i]], vEdgeWeights[i]);\n+ 958 }\n+ 959 }\n+ 960 dst[cVertValue].AddWithWeight(src[pVertValue], vVertWeight);\n+ 961 } else {\n+ 962 //\n+ 963 // Each FVar value associated with a vertex will be either a corner or a\n+crease,\n+ 964 // or potentially in transition from corner to crease:\n+ 965 // - if the CHILD is a corner, there can be no transition so we have a\n+corner\n+ 966 // - otherwise if the PARENT is a crease, both will be creases (no\n+transition)\n+ 967 // - otherwise the parent must be a corner and the child a crease\n+(transition)\n+ 968 //\n+ 969 Vtr::internal::FVarLevel::ConstValueTagArray pValueTags =\n+parentFVar.getVertexValueTags(vert);\n+ 970 Vtr::internal::FVarLevel::ConstValueTagArray cValueTags =\n+childFVar.getVertexValueTags(cVert);\n+ 971\n+ 972 for (int cSiblingIndex = 0; cSiblingIndex < cVertValues.size();\n+++cSiblingIndex) {\n+ 973 int pSiblingIndex = refineFVar.getChildValueParentSource(cVert,\n+cSiblingIndex);\n+ 974 assert(pSiblingIndex == cSiblingIndex);\n+ 975\n+ 976 typedef Vtr::internal::FVarLevel::Sibling SiblingIntType;\n+ 977\n+ 978 SiblingIntType cSibling = (SiblingIntType) cSiblingIndex;\n+ 979 SiblingIntType pSibling = (SiblingIntType) pSiblingIndex;\n+ 980\n+ 981 Vtr::Index pVertValue = pVertValues[pSibling];\n+ 982 Vtr::Index cVertValue = cVertValues[cSibling];\n+ 983\n+ 984 dst[cVertValue].Clear();\n+ 985 if (isLinearFVar || cValueTags[cSibling].isCorner()) {\n+ 986 dst[cVertValue].AddWithWeight(src[pVertValue], 1.0f);\n+ 987 } else {\n+ 988 //\n+ 989 // We have either a crease or a transition from corner to crease -- in\n+ 990 // either case, we need the end values for the full/fractional crease:\n+ 991 //\n+ 992 Index pEndValues[2];\n+ 993 parentFVar.getVertexCreaseEndValues(vert, pSibling, pEndValues);\n+ 994\n+ 995 Weight vWeight = 0.75f;\n+ 996 Weight eWeight = 0.125f;\n+ 997\n+ 998 //\n+ 999 // If semi-sharp we need to apply fractional weighting -- if made sharp\n+because\n+ 1000 // of the other sibling (dependent-sharp) use the fractional weight from\n+that\n+ 1001 // other sibling (should only occur when there are 2):\n+ 1002 //\n+ 1003 if (pValueTags[pSibling].isSemiSharp()) {\n+ 1004 Weight wCorner = pValueTags[pSibling].isDepSharp()\n+ 1005 ? refineFVar.getFractionalWeight(vert, !pSibling, cVert, !cSibling)\n+ 1006 : refineFVar.getFractionalWeight(vert, pSibling, cVert, cSibling);\n+ 1007 Weight wCrease = 1.0f - wCorner;\n+ 1008\n+ 1009 vWeight = wCrease * 0.75f + wCorner;\n+ 1010 eWeight = wCrease * 0.125f;\n+ 1011 }\n+ 1012 dst[cVertValue].AddWithWeight(src[pEndValues[0]], eWeight);\n+ 1013 dst[cVertValue].AddWithWeight(src[pEndValues[1]], eWeight);\n+ 1014 dst[cVertValue].AddWithWeight(src[pVertValue], vWeight);\n+ 1015 }\n+ 1016 }\n+ 1017 }\n+ 1018 }\n+ 1019}\n+ 1020\n+ 1021template \n+ 1022template \n+ 1023inline void\n+ 1024PrimvarRefinerReal::limit(T const & src, U & dstPos, U1 *\n+dstTan1Ptr, U2 * dstTan2Ptr) const {\n+ 1025\n+ 1026 Sdc::Scheme scheme(_refiner._subdivOptions);\n+ 1027\n+ 1028 Vtr::internal::Level const & level = _refiner.getLevel\n+(_refiner.GetMaxLevel());\n+ 1029\n+ 1030 int maxWeightsPerMask = 1 + 2 * level.getMaxValence();\n+ 1031 bool hasTangents = (dstTan1Ptr && dstTan2Ptr);\n+ 1032 int numMasks = 1 + (hasTangents ? 2 : 0);\n+ 1033\n+ 1034 Vtr::internal::StackBuffer indexBuffer(maxWeightsPerMask);\n+ 1035 Vtr::internal::StackBuffer weightBuffer(numMasks *\n+maxWeightsPerMask);\n+ 1036\n+ 1037 Weight * vPosWeights = weightBuffer,\n+ 1038 * ePosWeights = vPosWeights + 1,\n+ 1039 * fPosWeights = ePosWeights + level.getMaxValence();\n+ 1040 Weight * vTan1Weights = vPosWeights + maxWeightsPerMask,\n+ 1041 * eTan1Weights = ePosWeights + maxWeightsPerMask,\n+ 1042 * fTan1Weights = fPosWeights + maxWeightsPerMask;\n+ 1043 Weight * vTan2Weights = vTan1Weights + maxWeightsPerMask,\n+ 1044 * eTan2Weights = eTan1Weights + maxWeightsPerMask,\n+ 1045 * fTan2Weights = fTan1Weights + maxWeightsPerMask;\n+ 1046\n+ 1047 Mask posMask( vPosWeights, ePosWeights, fPosWeights);\n+ 1048 Mask tan1Mask(vTan1Weights, eTan1Weights, fTan1Weights);\n+ 1049 Mask tan2Mask(vTan2Weights, eTan2Weights, fTan2Weights);\n+ 1050\n+ 1051 // This is a bit obscure -- assigning both parent and child as last level\n+-- but\n+ 1052 // this mask type was intended for another purpose. Consider one for the\n+limit:\n+ 1053 Vtr::internal::VertexInterface vHood(level, level);\n+ 1054\n+ 1055 for (int vert = 0; vert < level.getNumVertices(); ++vert) {\n+ 1056 ConstIndexArray vEdges = level.getVertexEdges(vert);\n+ 1057\n+ 1058 // Incomplete vertices (present in sparse refinement) do not have their\n+full\n+ 1059 // topological neighborhood to determine a proper limit -- just leave the\n+ 1060 // vertex at the refined location and continue to the next:\n+ 1061 if (level.getVertexTag(vert)._incomplete || (vEdges.size() == 0)) {\n+ 1062 dstPos[vert].Clear();\n+ 1063 dstPos[vert].AddWithWeight(src[vert], 1.0);\n+ 1064 if (hasTangents) {\n+ 1065 (*dstTan1Ptr)[vert].Clear();\n+ 1066 (*dstTan2Ptr)[vert].Clear();\n+ 1067 }\n+ 1068 continue;\n+ 1069 }\n+ 1070\n+ 1071 //\n+ 1072 // Limit masks require the subdivision Rule for the vertex in order to\n+deal\n+ 1073 // with infinitely sharp features correctly -- including boundaries and\n+corners.\n+ 1074 // The vertex neighborhood is minimally defined with vertex and edge\n+counts.\n+ 1075 //\n+ 1076 Sdc::Crease::Rule vRule = level.getVertexRule(vert);\n+ 1077\n+ 1078 // This is a bit obscure -- child vertex index will be ignored here\n+ 1079 vHood.SetIndex(vert, vert);\n+ 1080\n+ 1081 if (hasTangents) {\n+ 1082 scheme.ComputeVertexLimitMask(vHood, posMask, tan1Mask, tan2Mask, vRule);\n+ 1083 } else {\n+ 1084 scheme.ComputeVertexLimitMask(vHood, posMask, vRule);\n+ 1085 }\n+ 1086\n+ 1087 //\n+ 1088 // Gather the neighboring vertices of this vertex -- the vertices\n+opposite its\n+ 1089 // incident edges, and the opposite vertices of its incident faces:\n+ 1090 //\n+ 1091 Index * eIndices = indexBuffer;\n+ 1092 Index * fIndices = indexBuffer + vEdges.size();\n+ 1093\n+ 1094 for (int i = 0; i < vEdges.size(); ++i) {\n+ 1095 ConstIndexArray eVerts = level.getEdgeVertices(vEdges[i]);\n+ 1096\n+ 1097 eIndices[i] = (eVerts[0] == vert) ? eVerts[1] : eVerts[0];\n+ 1098 }\n+ 1099 if (posMask.GetNumFaceWeights() || (hasTangents &&\n+tan1Mask.GetNumFaceWeights())) {\n+ 1100 ConstIndexArray vFaces = level.getVertexFaces(vert);\n+ 1101 ConstLocalIndexArray vInFace = level.getVertexFaceLocalIndices(vert);\n+ 1102\n+ 1103 for (int i = 0; i < vFaces.size(); ++i) {\n+ 1104 ConstIndexArray fVerts = level.getFaceVertices(vFaces[i]);\n+ 1105\n+ 1106 LocalIndex vOppInFace = (vInFace[i] + 2);\n+ 1107 if (vOppInFace >= fVerts.size()) vOppInFace -= (LocalIndex)fVerts.size();\n+ 1108\n+ 1109 fIndices[i] = level.getFaceVertices(vFaces[i])[vOppInFace];\n+ 1110 }\n+ 1111 }\n+ 1112\n+ 1113 //\n+ 1114 // Combine the weights and indices for position and tangents. As with\n+applying\n+ 1115 // refinement masks to vertex data, in order to improve numerical\n+precision, it's\n+ 1116 // better to apply smaller weights first, so begin with the face-weights\n+followed\n+ 1117 // by the edge-weights and the vertex weight last.\n+ 1118 //\n+ 1119 dstPos[vert].Clear();\n+ 1120 for (int i = 0; i < posMask.GetNumFaceWeights(); ++i) {\n+ 1121 dstPos[vert].AddWithWeight(src[fIndices[i]], fPosWeights[i]);\n+ 1122 }\n+ 1123 for (int i = 0; i < posMask.GetNumEdgeWeights(); ++i) {\n+ 1124 dstPos[vert].AddWithWeight(src[eIndices[i]], ePosWeights[i]);\n+ 1125 }\n+ 1126 dstPos[vert].AddWithWeight(src[vert], vPosWeights[0]);\n+ 1127\n+ 1128 //\n+ 1129 // Apply the tangent masks -- both will have the same number of weights\n+and\n+ 1130 // indices (one tangent may be \"padded\" to accommodate the other), but\n+these\n+ 1131 // may differ from those of the position:\n+ 1132 //\n+ 1133 if (hasTangents) {\n+ 1134 assert(tan1Mask.GetNumFaceWeights() == tan2Mask.GetNumFaceWeights());\n+ 1135 assert(tan1Mask.GetNumEdgeWeights() == tan2Mask.GetNumEdgeWeights());\n+ 1136\n+ 1137 U1 & dstTan1 = *dstTan1Ptr;\n+ 1138 U2 & dstTan2 = *dstTan2Ptr;\n+ 1139\n+ 1140 dstTan1[vert].Clear();\n+ 1141 dstTan2[vert].Clear();\n+ 1142 for (int i = 0; i < tan1Mask.GetNumFaceWeights(); ++i) {\n+ 1143 dstTan1[vert].AddWithWeight(src[fIndices[i]], fTan1Weights[i]);\n+ 1144 dstTan2[vert].AddWithWeight(src[fIndices[i]], fTan2Weights[i]);\n+ 1145 }\n+ 1146 for (int i = 0; i < tan1Mask.GetNumEdgeWeights(); ++i) {\n+ 1147 dstTan1[vert].AddWithWeight(src[eIndices[i]], eTan1Weights[i]);\n+ 1148 dstTan2[vert].AddWithWeight(src[eIndices[i]], eTan2Weights[i]);\n+ 1149 }\n+ 1150 dstTan1[vert].AddWithWeight(src[vert], vTan1Weights[0]);\n+ 1151 dstTan2[vert].AddWithWeight(src[vert], vTan2Weights[0]);\n+ 1152 }\n+ 1153 }\n+ 1154}\n+ 1155\n+ 1156template \n+ 1157template \n+ 1158inline void\n+ 1159PrimvarRefinerReal::limitFVar(T const & src, U & dst, int channel)\n+const {\n+ 1160\n+ 1161 Sdc::Scheme scheme(_refiner._subdivOptions);\n+ 1162\n+ 1163 Vtr::internal::Level const & level = _refiner.getLevel\n+(_refiner.GetMaxLevel());\n+ 1164 Vtr::internal::FVarLevel const & fvarChannel = level.getFVarLevel\n+(channel);\n+ 1165\n+ 1166 int maxWeightsPerMask = 1 + 2 * level.getMaxValence();\n+ 1167\n+ 1168 Vtr::internal::StackBuffer weightBuffer(maxWeightsPerMask);\n+ 1169 Vtr::internal::StackBuffer vEdgeBuffer(level.getMaxValence());\n+ 1170\n+ 1171 // This is a bit obscure -- assign both parent and child as last level\n+ 1172 Vtr::internal::VertexInterface vHood(level, level);\n+ 1173\n+ 1174 for (int vert = 0; vert < level.getNumVertices(); ++vert) {\n+ 1175\n+ 1176 ConstIndexArray vEdges = level.getVertexEdges(vert);\n+ 1177 ConstIndexArray vValues = fvarChannel.getVertexValues(vert);\n+ 1178\n+ 1179 // Incomplete vertices (present in sparse refinement) do not have their\n+full\n+ 1180 // topological neighborhood to determine a proper limit -- just leave the\n+ 1181 // values (perhaps more than one per vertex) at the refined location.\n+ 1182 //\n+ 1183 // The same can be done if the face-varying channel is purely linear.\n+ 1184 //\n+ 1185 bool isIncomplete = (level.getVertexTag(vert)._incomplete || (vEdges.size\n+() == 0));\n+ 1186 if (isIncomplete || fvarChannel.isLinear()) {\n+ 1187 for (int i = 0; i < vValues.size(); ++i) {\n+ 1188 Vtr::Index vValue = vValues[i];\n+ 1189\n+ 1190 dst[vValue].Clear();\n+ 1191 dst[vValue].AddWithWeight(src[vValue], 1.0f);\n+ 1192 }\n+ 1193 continue;\n+ 1194 }\n+ 1195\n+ 1196 bool fvarVertMatchesVertex = fvarChannel.valueTopologyMatches(vValues\n+[0]);\n+ 1197 if (fvarVertMatchesVertex) {\n+ 1198\n+ 1199 // Assign the mask weights to the common buffer and compute the mask:\n+ 1200 //\n+ 1201 Weight * vWeights = weightBuffer,\n+ 1202 * eWeights = vWeights + 1,\n+ 1203 * fWeights = eWeights + vEdges.size();\n+ 1204\n+ 1205 Mask vMask(vWeights, eWeights, fWeights);\n+ 1206\n+ 1207 vHood.SetIndex(vert, vert);\n+ 1208\n+ 1209 scheme.ComputeVertexLimitMask(vHood, vMask, level.getVertexRule(vert));\n+ 1210\n+ 1211 //\n+ 1212 // Apply mask to corresponding FVar values for neighboring vertices:\n+ 1213 //\n+ 1214 Vtr::Index vValue = vValues[0];\n+ 1215\n+ 1216 dst[vValue].Clear();\n+ 1217 if (vMask.GetNumFaceWeights() > 0) {\n+ 1218 assert(!vMask.AreFaceWeightsForFaceCenters());\n+ 1219\n+ 1220 ConstIndexArray vFaces = level.getVertexFaces(vert);\n+ 1221 ConstLocalIndexArray vInFace = level.getVertexFaceLocalIndices(vert);\n+ 1222\n+ 1223 for (int i = 0; i < vFaces.size(); ++i) {\n+ 1224 ConstIndexArray faceValues = fvarChannel.getFaceValues(vFaces[i]);\n+ 1225 LocalIndex vOppInFace = vInFace[i] + 2;\n+ 1226 if (vOppInFace >= faceValues.size()) vOppInFace -= faceValues.size();\n+ 1227\n+ 1228 Index vValueOppositeFace = faceValues[vOppInFace];\n+ 1229\n+ 1230 dst[vValue].AddWithWeight(src[vValueOppositeFace], fWeights[i]);\n+ 1231 }\n+ 1232 }\n+ 1233 if (vMask.GetNumEdgeWeights() > 0) {\n+ 1234 Index * vEdgeValues = vEdgeBuffer;\n+ 1235 fvarChannel.getVertexEdgeValues(vert, vEdgeValues);\n+ 1236\n+ 1237 for (int i = 0; i < vEdges.size(); ++i) {\n+ 1238 dst[vValue].AddWithWeight(src[vEdgeValues[i]], eWeights[i]);\n+ 1239 }\n+ 1240 }\n+ 1241 dst[vValue].AddWithWeight(src[vValue], vWeights[0]);\n+ 1242 } else {\n+ 1243 //\n+ 1244 // Sibling FVar values associated with a vertex will be either a corner\n+or a crease:\n+ 1245 //\n+ 1246 for (int i = 0; i < vValues.size(); ++i) {\n+ 1247 Vtr::Index vValue = vValues[i];\n+ 1248\n+ 1249 dst[vValue].Clear();\n+ 1250 if (fvarChannel.getValueTag(vValue).isCorner()) {\n+ 1251 dst[vValue].AddWithWeight(src[vValue], 1.0f);\n+ 1252 } else {\n+ 1253 Index vEndValues[2];\n+ 1254 fvarChannel.getVertexCreaseEndValues(vert, i, vEndValues);\n+ 1255\n+ 1256 dst[vValue].AddWithWeight(src[vEndValues[0]], 1.0f/6.0f);\n+ 1257 dst[vValue].AddWithWeight(src[vEndValues[1]], 1.0f/6.0f);\n+ 1258 dst[vValue].AddWithWeight(src[vValue], 2.0f/3.0f);\n+ 1259 }\n+ 1260 }\n+ 1261 }\n+ 1262 }\n+ 1263}\n+ 1264\n+1265class PrimvarRefiner : public PrimvarRefinerReal {\n+ 1266public:\n+1267 PrimvarRefiner(TopologyRefiner const & refiner)\n+ 1268 : PrimvarRefinerReal(refiner) { }\n+ 1269};\n+ 1270\n+ 1271} // end namespace Far\n+ 1272\n+ 1273} // end namespace OPENSUBDIV_VERSION\n+ 1274using namespace OPENSUBDIV_VERSION;\n+ 1275} // end namespace OpenSubdiv\n+ 1276\n+ 1277#endif /* OPENSUBDIV3_FAR_PRIMVAR_REFINER_H */\n OpenSubdiv\n Definition: limits.h:32\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LocalIndex\n+Vtr::LocalIndex LocalIndex\n+Definition: types.h:42\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::Error\n+void Error(ErrorType err, const char *format,...)\n+Sends an OSD error with a message (internal use only)\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstIndexArray\n+Vtr::ConstIndexArray ConstIndexArray\n+Definition: types.h:47\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstLocalIndexArray\n+Vtr::ConstLocalIndexArray ConstLocalIndexArray\n+Definition: types.h:48\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::FAR_RUNTIME_ERROR\n+@ FAR_RUNTIME_ERROR\n+Issue a generic runtime error, but continue execution.\n+Definition: error.h:40\n OpenSubdiv::OPENSUBDIV_VERSION::Far::Index\n Vtr::Index Index\n Definition: types.h:41\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal\n-A specialized factory for StencilTable.\n-Definition: stencilTableFactory.h:51\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal\n-A specialized factory for LimitStencilTable.\n-Definition: stencilTableFactory.h:209\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal\n-Vertex stencil descriptor.\n-Definition: stencilTable.h:53\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::StencilReal\n-StencilReal()\n-Default constructor.\n-Definition: stencilTable.h:57\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::_size\n-int * _size\n-Definition: stencilTable.h:109\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::StencilReal\n-StencilReal(int *size, Index *indices, REAL *weights)\n-Constructor.\n-Definition: stencilTable.h:67\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::Next\n-void Next()\n-Advance to the next stencil in the table.\n-Definition: stencilTable.h:98\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::StencilReal\n-StencilReal(StencilReal const &other)\n-Copy constructor.\n-Definition: stencilTable.h:71\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::_indices\n-Index * _indices\n-Definition: stencilTable.h:110\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetVertexIndices\n-Index const * GetVertexIndices() const\n-Returns the control vertices' indices.\n-Definition: stencilTable.h:88\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetSize\n-int GetSize() const\n-Returns the size of the stencil.\n-Definition: stencilTable.h:78\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::_weights\n-REAL * _weights\n-Definition: stencilTable.h:111\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetSizePtr\n-int * GetSizePtr() const\n-Returns the size of the stencil as a pointer.\n-Definition: stencilTable.h:83\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::GetWeights\n-REAL const * GetWeights() const\n-Returns the interpolation weights.\n-Definition: stencilTable.h:93\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil\n-Vertex stencil class wrapping the template for compatibility.\n-Definition: stencilTable.h:116\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil::Stencil\n-Stencil(BaseStencil const &other)\n-Definition: stencilTable.h:122\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil::Stencil\n-Stencil(int *size, Index *indices, float *weights)\n-Definition: stencilTable.h:123\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil::BaseStencil\n-StencilReal< float > BaseStencil\n-Definition: stencilTable.h:118\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::Stencil::Stencil\n-Stencil()\n-Definition: stencilTable.h:121\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal\n-Table of subdivision stencils.\n-Definition: stencilTable.h:141\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumControlVertices\n-int GetNumControlVertices() const\n-Returns the number of control vertices indexed in the table.\n-Definition: stencilTable.h:161\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::_sizes\n-std::vector< int > _sizes\n-Definition: stencilTable.h:265\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::PatchTableBuilder\n-friend class Far::PatchTableBuilder\n-Definition: stencilTable.h:261\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::update\n-void update(T1 const &srcBase, int numBase, T2 const &srcRef, U &dstValues,\n-std::vector< REAL > const &valueWeights, Index start, Index end) const\n-Definition: stencilTable.h:618\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues\n-void UpdateValues(T1 const &srcBase, int numBase, T2 const &srcRef, U\n-&dstValues, Index start=-1, Index end=-1) const\n-Definition: stencilTable.h:210\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetControlIndices\n-std::vector< Index > const & GetControlIndices() const\n-Returns the indices of the control vertices.\n-Definition: stencilTable.h:179\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::finalize\n-void finalize()\n-Definition: stencilTable.h:714\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::operator[]\n-StencilReal< REAL > operator[](Index index) const\n-Returns the stencil at index i in the table.\n-Definition: stencilTable.h:734\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::_numControlVertices\n-int _numControlVertices\n-Definition: stencilTable.h:263\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::update\n-void update(T const &srcValues, U &dstValues, std::vector< REAL > const\n-&valueWeights, Index start, Index end) const\n-Definition: stencilTable.h:650\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetSizes\n-std::vector< int > const & GetSizes() const\n-Returns the number of control vertices of each stencil in the table.\n-Definition: stencilTable.h:169\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumStencils\n-int GetNumStencils() const\n-Returns the number of stencils in the table.\n-Definition: stencilTable.h:156\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::~StencilTableReal\n-virtual ~StencilTableReal()\n-Definition: stencilTable.h:153\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetOffsets\n-std::vector< Index > const & GetOffsets() const\n-Returns the offset to a given stencil (factory may leave empty)\n-Definition: stencilTable.h:174\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetStencil\n-StencilReal< REAL > GetStencil(Index i) const\n-Returns a Stencil at index i in the table.\n-Definition: stencilTable.h:722\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::shrinkToFit\n-void shrinkToFit()\n-Definition: stencilTable.h:706\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::generateOffsets\n-void generateOffsets()\n-Definition: stencilTable.h:678\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::Clear\n-void Clear()\n-Clears the stencils from the table.\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::_offsets\n-std::vector< Index > _offsets\n-Definition: stencilTable.h:266\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::_weights\n-std::vector< REAL > _weights\n-Definition: stencilTable.h:268\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::resize\n-void resize(int nstencils, int nelems)\n-Definition: stencilTable.h:690\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetWeights\n-std::vector< REAL > const & GetWeights() const\n-Returns the stencil interpolation weights.\n-Definition: stencilTable.h:184\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues\n-void UpdateValues(T1 const *srcBase, int numBase, T2 const *srcRef, U *dst,\n-Index start=-1, Index end=-1) const\n-Definition: stencilTable.h:221\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::StencilTableReal\n-StencilTableReal(int numControlVerts, std::vector< int > const &offsets, std::\n-vector< int > const &sizes, std::vector< int > const &sources, std::vector<\n-REAL > const &weights, bool includeCoarseVerts, size_t firstOffset)\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::_indices\n-std::vector< Index > _indices\n-Definition: stencilTable.h:267\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::StencilTableReal\n-StencilTableReal()\n-Definition: stencilTable.h:255\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::reserve\n-void reserve(int nstencils, int nelems)\n-Definition: stencilTable.h:698\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues\n-void UpdateValues(T const *src, U *dst, Index start=-1, Index end=-1) const\n-Definition: stencilTable.h:217\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::UpdateValues\n-void UpdateValues(T const &srcValues, U &dstValues, Index start=-1, Index end=-\n-1) const\n-Updates point values based on the control values.\n-Definition: stencilTable.h:205\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n-Stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:273\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::StencilTable\n-StencilTable(int numControlVerts, std::vector< int > const &offsets, std::\n-vector< int > const &sizes, std::vector< int > const &sources, std::vector<\n-float > const &weights, bool includeCoarseVerts, size_t firstOffset)\n-Definition: stencilTable.h:288\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::StencilTable\n-StencilTable(int numControlVerts)\n-Definition: stencilTable.h:287\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::BaseTable\n-StencilTableReal< float > BaseTable\n-Definition: stencilTable.h:275\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::GetStencil\n-Stencil GetStencil(Index index) const\n-Definition: stencilTable.h:278\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::operator[]\n-Stencil operator[](Index index) const\n-Definition: stencilTable.h:281\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::StencilTable\n-StencilTable()\n-Definition: stencilTable.h:286\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal\n-Limit point stencil descriptor.\n-Definition: stencilTable.h:303\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuvWeights\n-REAL const * GetDuvWeights() const\n-Returns the uv derivative weights.\n-Definition: stencilTable.h:356\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDvWeights\n-REAL const * GetDvWeights() const\n-Returns the v derivative weights.\n-Definition: stencilTable.h:346\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::Next\n-void Next()\n-Advance to the next stencil in the table.\n-Definition: stencilTable.h:366\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDvvWeights\n-REAL const * GetDvvWeights() const\n-Returns the vv derivative weights.\n-Definition: stencilTable.h:361\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::LimitStencilReal\n-LimitStencilReal(int *size, Index *indices, REAL *weights, REAL *duWeights=0,\n-REAL *dvWeights=0, REAL *duuWeights=0, REAL *duvWeights=0, REAL *dvvWeights=0)\n-Constructor.\n-Definition: stencilTable.h:324\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuWeights\n-REAL const * GetDuWeights() const\n-Returns the u derivative weights.\n-Definition: stencilTable.h:341\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::GetDuuWeights\n-REAL const * GetDuuWeights() const\n-Returns the uu derivative weights.\n-Definition: stencilTable.h:351\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil\n-Limit point stencil class wrapping the template for compatibility.\n-Definition: stencilTable.h:392\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil::LimitStencil\n-LimitStencil(BaseStencil const &other)\n-Definition: stencilTable.h:397\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil::BaseStencil\n-LimitStencilReal< float > BaseStencil\n-Definition: stencilTable.h:394\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil::LimitStencil\n-LimitStencil(int *size, Index *indices, float *weights, float *duWeights=0,\n-float *dvWeights=0, float *duuWeights=0, float *duvWeights=0, float\n-*dvvWeights=0)\n-Definition: stencilTable.h:398\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal\n-Table of limit subdivision stencils.\n-Definition: stencilTable.h:409\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuuWeights\n-std::vector< REAL > const & GetDuuWeights() const\n-Returns the 'uu' derivative stencil interpolation weights.\n-Definition: stencilTable.h:444\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::operator[]\n-LimitStencilReal< REAL > operator[](Index index) const\n-Returns the limit stencil at index i in the table.\n-Definition: stencilTable.h:782\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDvvWeights\n-std::vector< REAL > const & GetDvvWeights() const\n-Returns the 'vv' derivative stencil interpolation weights.\n-Definition: stencilTable.h:454\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs\n-void UpdateDerivs(T const *src, U *uderivs, U *vderivs, int start=-1, int end=-\n-1) const\n-Definition: stencilTable.h:493\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs\n-void Update2ndDerivs(T const *src, T *uuderivs, U *uvderivs, U *vvderivs, int\n-start=-1, int end=-1) const\n-Definition: stencilTable.h:548\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDvWeights\n-std::vector< REAL > const & GetDvWeights() const\n-Returns the 'v' derivative stencil interpolation weights.\n-Definition: stencilTable.h:439\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs\n-void Update2ndDerivs(T1 const *srcBase, int numBase, T2 const *srcRef, U\n-*uuderivs, U *uvderivs, U *vvderivs, int start=-1, int end=-1) const\n-Definition: stencilTable.h:556\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs\n-void UpdateDerivs(T const &srcValues, U &uderivs, U &vderivs, int start=-1, int\n-end=-1) const\n-Updates derivative values based on the control values.\n-Definition: stencilTable.h:476\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetLimitStencil\n-LimitStencilReal< REAL > GetLimitStencil(Index i) const\n-Returns a LimitStencil at index i in the table.\n-Definition: stencilTable.h:749\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::\n-LimitStencilTableReal\n-LimitStencilTableReal(int numControlVerts, std::vector< int > const &offsets,\n-std::vector< int > const &sizes, std::vector< int > const &sources, std::\n-vector< REAL > const &weights, std::vector< REAL > const &duWeights, std::\n-vector< REAL > const &dvWeights, std::vector< REAL > const &duuWeights, std::\n-vector< REAL > const &duvWeights, std::vector< REAL > const &dvvWeights, bool\n-includeCoarseVerts, size_t firstOffset)\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs\n-void Update2ndDerivs(T const &srcValues, U &uuderivs, U &uvderivs, U &vvderivs,\n-int start=-1, int end=-1) const\n-Updates 2nd derivative values based on the control values.\n-Definition: stencilTable.h:528\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuWeights\n-std::vector< REAL > const & GetDuWeights() const\n-Returns the 'u' derivative stencil interpolation weights.\n-Definition: stencilTable.h:434\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Clear\n-void Clear()\n-Clears the stencils from the table.\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Update2ndDerivs\n-void Update2ndDerivs(T1 const &srcBase, int numBase, T2 const &srcRef, U\n-&uuderivs, U &uvderivs, U &vvderivs, int start=-1, int end=-1) const\n-Definition: stencilTable.h:538\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::GetDuvWeights\n-std::vector< REAL > const & GetDuvWeights() const\n-Returns the 'uv' derivative stencil interpolation weights.\n-Definition: stencilTable.h:449\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs\n-void UpdateDerivs(T1 const &srcBase, int numBase, T2 const &srcRef, U &uderivs,\n-U &vderivs, int start=-1, int end=-1) const\n-Definition: stencilTable.h:484\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::UpdateDerivs\n-void UpdateDerivs(T1 const *srcBase, int numBase, T2 const *srcRef, U *uderivs,\n-U *vderivs, int start=-1, int end=-1) const\n-Definition: stencilTable.h:500\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n-Limit stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:583\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable::BaseTable\n-LimitStencilTableReal< float > BaseTable\n-Definition: stencilTable.h:585\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable::GetLimitStencil\n-LimitStencil GetLimitStencil(Index index) const\n-Definition: stencilTable.h:588\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable::LimitStencilTable\n-LimitStencilTable(int numControlVerts, std::vector< int > const &offsets, std::\n-vector< int > const &sizes, std::vector< int > const &sources, std::vector<\n-float > const &weights, std::vector< float > const &duWeights, std::vector<\n-float > const &dvWeights, std::vector< float > const &duuWeights, std::vector<\n-float > const &duvWeights, std::vector< float > const &dvvWeights, bool\n-includeCoarseVerts, size_t firstOffset)\n-Definition: stencilTable.h:596\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable::operator[]\n-LimitStencil operator[](Index index) const\n-Definition: stencilTable.h:591\n+OpenSubdiv::OPENSUBDIV_VERSION::Vtr::Index\n+int Index\n+Definition: types.h:54\n+OpenSubdiv::OPENSUBDIV_VERSION::Vtr::IndexIsValid\n+bool IndexIsValid(Index index)\n+Definition: types.h:58\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SCHEME_BILINEAR\n+@ SCHEME_BILINEAR\n+Definition: types.h:38\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SCHEME_CATMARK\n+@ SCHEME_CATMARK\n+Definition: types.h:39\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SCHEME_LOOP\n+@ SCHEME_LOOP\n+Definition: types.h:40\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal\n+Applies refinement operations to generic primvar data.\n+Definition: primvarRefiner.h:56\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Limit\n+void Limit(T const &src, U &dstPos) const\n+Apply limit weights to a primvar buffer.\n+Definition: primvarRefiner.h:319\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::PrimvarRefinerReal\n+PrimvarRefinerReal(TopologyRefiner const &refiner)\n+Definition: primvarRefiner.h:59\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Limit\n+void Limit(T const &src, U &dstPos, U1 &dstTan1, U2 &dstTan2) const\n+Definition: primvarRefiner.h:344\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateFaceUniform\n+void InterpolateFaceUniform(int level, T const &src, U &dst) const\n+Refine uniform (per-face) primvar data between levels.\n+Definition: primvarRefiner.h:394\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::GetTopologyRefiner\n+TopologyRefiner const & GetTopologyRefiner() const\n+Definition: primvarRefiner.h:62\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateVarying\n+void InterpolateVarying(int level, T const &src, U &dst) const\n+Apply only varying interpolation weights to a primvar buffer for a single level\n+of refinement.\n+Definition: primvarRefiner.h:412\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Interpolate\n+void Interpolate(int level, T const &src, U &dst) const\n+Apply vertex interpolation weights to a primvar buffer for a single level of\n+refinement.\n+Definition: primvarRefiner.h:267\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::InterpolateFaceVarying\n+void InterpolateFaceVarying(int level, T const &src, U &dst, int channel=0)\n+const\n+Apply face-varying interpolation weights to a primvar buffer associated with a\n+particular face-varyin...\n+Definition: primvarRefiner.h:293\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::~PrimvarRefinerReal\n+~PrimvarRefinerReal()\n+Definition: primvarRefiner.h:60\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::LimitFaceVarying\n+void LimitFaceVarying(T const &src, U &dst, int channel=0) const\n+Definition: primvarRefiner.h:369\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefiner\n+Definition: primvarRefiner.h:1265\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefiner::PrimvarRefiner\n+PrimvarRefiner(TopologyRefiner const &refiner)\n+Definition: primvarRefiner.h:1267\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner\n+Stores topology data for a specified set of refinement options.\n+Definition: topologyRefiner.h:51\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::Rule\n+Rule\n+Definition: crease.h:82\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::RULE_SMOOTH\n+@ RULE_SMOOTH\n+Definition: crease.h:84\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::RULE_CREASE\n+@ RULE_CREASE\n+Definition: crease.h:86\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme\n+Scheme is a class template which provides all implementation for the\n+subdivision schemes supported by...\n+Definition: scheme.h:66\n+OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstArray\n+Definition: array.h:53\n+OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstArray::size\n+size_type size() const\n+Definition: array.h:72\n * opensubdiv\n * far\n- * stencilTable.h\n+ * primvarRefiner.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00731.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00731.html", "unified_diff": "@@ -87,17 +87,17 @@\n
\n Classes |\n Namespaces
\n
patchTable.h File Reference
\n \n
\n
#include "../version.h"
\n-#include "../far/patchDescriptor.h"
\n-#include "../far/patchParam.h"
\n-#include "../far/stencilTable.h"
\n+#include "../far/patchDescriptor.h"
\n+#include "../far/patchParam.h"
\n+#include "../far/stencilTable.h"
\n #include "../sdc/options.h"
\n #include <vector>
\n
\n

Go to the source code of this file.

\n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00731_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00731_source.html", "unified_diff": "@@ -693,20 +693,20 @@\n
856} // end namespace OPENSUBDIV_VERSION
\n
857using namespace OPENSUBDIV_VERSION;
\n
858
\n
859} // end namespace OpenSubdiv
\n
860
\n
861#endif /* OPENSUBDIV3_FAR_PATCH_TABLE */
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParamTable
std::vector< PatchParam > PatchParamTable
Definition: patchParam.h:243
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParamTable
std::vector< PatchParam > PatchParamTable
Definition: patchParam.h:243
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n-\n+\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchMap
An quadtree-based map connecting coarse faces to their sub-patches.
Definition: patchMap.h:49
\n-\n+\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::GetPatchArrayVaryingVertices
ConstIndexArray GetPatchArrayVaryingVertices(int array) const
Returns the varying vertex indices for the patches in array.
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::GetPatchControlVerticesTable
PatchVertsTable const & GetPatchControlVerticesTable() const
Get the table of patch control vertices.
Definition: patchTable.h:378
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::ComputeLocalPointValuesFaceVarying
void ComputeLocalPointValuesFaceVarying(T const *src, T *dst, int channel=0) const
Updates local point face-varying values.
Definition: patchTable.h:796
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::GetVaryingVertices
ConstIndexArray GetVaryingVertices() const
Returns an array of varying vertex indices for the patches.
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::ComputeLocalPointValuesVarying
void ComputeLocalPointValuesVarying(T const *src, T *dst) const
Updates local point varying values.
Definition: patchTable.h:787
\n@@ -779,16 +779,16 @@\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::LocalPointStencilPrecisionMatchesType
bool LocalPointStencilPrecisionMatchesType() const
Tests if the precision of the stencil table to compute local point vertex values matches the given fl...
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::GetLocalPointVaryingStencilTable
StencilTableReal< REAL > const * GetLocalPointVaryingStencilTable() const
Returns the stencil table to compute local point varying values.
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::GetNumLocalPointsFaceVarying
int GetNumLocalPointsFaceVarying(int channel=0) const
Returns the number of local face-varying points for channel.
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::PatchHandle
Handle that can be used as unique patch identifier within PatchTable.
Definition: patchTable.h:60
\n \n \n \n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n \n \n \n \n \n \n
\n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00737.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00737.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/patchParam.h File Reference\n+OpenSubdiv: opensubdiv/far/patchDescriptor.h File Reference\n \n \n \n \n \n \n \n@@ -84,50 +84,48 @@\n
\n \n
\n
\n Classes |\n Namespaces |\n Typedefs
\n-
patchParam.h File Reference
\n+
patchDescriptor.h File Reference
\n
\n
\n
#include "../version.h"
\n #include "../far/types.h"
\n+#include "../sdc/types.h"
\n+#include <vector>
\n
\n

Go to the source code of this file.

\n

\n Classes

\n \n-\n-\n+\n+\n \n

\n Classes

struct  PatchParam
 Patch parameterization. More...
class  PatchDescriptor
 Describes the type of a patch. More...
 
\n \n \n \n \n \n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
\n \n-\n-\n-\n-\n-\n-\n+\n+\n

\n Typedefs

typedef std::vector< PatchParam > PatchParamTable
 
typedef Vtr::Array< PatchParam > PatchParamArray
 
typedef Vtr::ConstArray< PatchParam > ConstPatchParamArray
 
typedef Vtr::ConstArray< PatchDescriptor > ConstPatchDescriptorArray
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,33 +5,31 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces | Typedefs\n-patchParam.h File Reference\n+patchDescriptor.h File Reference\n #include \"../version.h\"\n #include \"../far/types.h\"\n+#include \"../sdc/types.h\"\n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n-struct \u00a0PatchParam\n-\u00a0 Patch parameterization. More...\n+class \u00a0PatchDescriptor\n+\u00a0 Describes the type of a patch. More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n Typedefs\n- typedef std::vector< PatchParam >\u00a0PatchParamTable\n-\u00a0\n- typedef Vtr::Array< PatchParam >\u00a0PatchParamArray\n-\u00a0\n-typedef Vtr::ConstArray< PatchParam >\u00a0ConstPatchParamArray\n+typedef Vtr::ConstArray< PatchDescriptor >\u00a0ConstPatchDescriptorArray\n \u00a0\n * opensubdiv\n * far\n- * patchParam.h\n+ * patchDescriptor.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00737.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00737.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,6 +1,4 @@\n var a00737 = [\n- [\"PatchParam\", \"a00989.html\", \"a00989\"],\n- [\"ConstPatchParamArray\", \"a00737.html#a3139c7c52ddf26b2403e8e95ff63eeb3\", null],\n- [\"PatchParamArray\", \"a00737.html#abe07210a029a01571c54276ae28b4d88\", null],\n- [\"PatchParamTable\", \"a00737.html#a46ea68db89cc3d55c4191ef7648281ce\", null]\n+ [\"PatchDescriptor\", \"a00973.html\", \"a00973\"],\n+ [\"ConstPatchDescriptorArray\", \"a00737.html#ad258c2d8f7871bc0d40cf3a9ded6e08a\", null]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00737_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00737_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/patchParam.h Source File\n+OpenSubdiv: opensubdiv/far/patchDescriptor.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
patchParam.h
\n+
patchDescriptor.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,198 +107,196 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_FAR_PATCH_PARAM_H
\n-
26#define OPENSUBDIV3_FAR_PATCH_PARAM_H
\n+
25#ifndef OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H
\n+
26#define OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H
\n
27
\n
28#include "../version.h"
\n
29
\n
30#include "../far/types.h"
\n-
31
\n-
32namespace OpenSubdiv {
\n-
33namespace OPENSUBDIV_VERSION {
\n+
31#include "../sdc/types.h"
\n+
32
\n+
33#include <vector>
\n
34
\n-
35namespace Far {
\n-
36
\n-
101
\n-
152struct PatchParam {
\n-
168 void Set(Index faceid, short u, short v,
\n-
169 unsigned short depth, bool nonquad,
\n-
170 unsigned short boundary, unsigned short transition,
\n-
171 bool regular = false);
\n-
172
\n-
174 void Clear() { field0 = field1 = 0; }
\n-
175
\n-
177 Index GetFaceId() const { return Index(unpack(field0,28,0)); }
\n-
178
\n-
181 unsigned short GetU() const { return (unsigned short)unpack(field1,10,22); }
\n-
182
\n-
185 unsigned short GetV() const { return (unsigned short)unpack(field1,10,12); }
\n-
186
\n-
188 unsigned short GetTransition() const { return (unsigned short)unpack(field0,4,28); }
\n-
189
\n-
191 unsigned short GetBoundary() const { return (unsigned short)unpack(field1,5,7); }
\n+
35namespace OpenSubdiv {
\n+
36namespace OPENSUBDIV_VERSION {
\n+
37
\n+
38namespace Far {
\n+
39
\n+
44class PatchDescriptor {
\n+
45
\n+
46public:
\n+
47
\n+
48 enum Type {
\n+
49 NON_PATCH = 0,
\n+
50
\n+
51 POINTS,
\n+
52 LINES,
\n+
53
\n+
54 QUADS,
\n+
55 TRIANGLES,
\n+
56
\n+
57 LOOP,
\n+
58
\n+
59 REGULAR,
\n+
60 GREGORY,
\n+
61 GREGORY_BOUNDARY,
\n+
62 GREGORY_BASIS,
\n+
63 GREGORY_TRIANGLE
\n+
64 };
\n+
65
\n+
66public:
\n+
67
\n+
69 PatchDescriptor() :
\n+
70 _type(NON_PATCH) { }
\n+
71
\n+
73 PatchDescriptor(int type) :
\n+
74 _type(type) { }
\n+
75
\n+
77 PatchDescriptor( PatchDescriptor const & d ) :
\n+
78 _type(d.GetType()) { }
\n+
79
\n+
81 PatchDescriptor & operator=( PatchDescriptor const & d ) {
\n+
82 _type = d.GetType();
\n+
83 return *this;
\n+
84 }
\n+
85
\n+
87 Type GetType() const {
\n+
88 return (Type)_type;
\n+
89 }
\n+
90
\n+
92 static inline bool IsAdaptive(Type type) {
\n+
93 return type > TRIANGLES;
\n+
94 }
\n+
95
\n+
97 bool IsAdaptive() const {
\n+
98 return IsAdaptive( this->GetType() );
\n+
99 }
\n+
100
\n+
103 static inline short GetNumControlVertices( Type t );
\n+
104
\n+
106 static inline short GetNumFVarControlVertices( Type t );
\n+
107
\n+
110 short GetNumControlVertices() const {
\n+
111 return GetNumControlVertices( this->GetType() );
\n+
112 }
\n+
113
\n+
115 short GetNumFVarControlVertices() const {
\n+
116 return GetNumFVarControlVertices( this->GetType() );
\n+
117 }
\n+
118
\n+
120 static short GetRegularPatchSize() { return 16; }
\n+
121
\n+
123 static short GetGregoryPatchSize() { return 4; }
\n+
124
\n+
126 static short GetGregoryBasisPatchSize() { return 20; }
\n+
127
\n+
128
\n+
131 static Vtr::ConstArray<PatchDescriptor> GetAdaptivePatchDescriptors(Sdc::SchemeType type);
\n+
132
\n+
134 inline bool operator < ( PatchDescriptor const other ) const;
\n+
135
\n+
137 inline bool operator == ( PatchDescriptor const other ) const;
\n+
138
\n+
139 // debug helper
\n+
140 void print() const;
\n+
141
\n+
142private:
\n+
143 unsigned int _type;
\n+
144};
\n+
145
\n+
146typedef Vtr::ConstArray<PatchDescriptor> ConstPatchDescriptorArray;
\n+
147
\n+
148// Returns the number of control vertices expected for a patch of this type
\n+
149inline short
\n+
150PatchDescriptor::GetNumControlVertices( Type type ) {
\n+
151 switch (type) {
\n+
152 case REGULAR : return GetRegularPatchSize();
\n+
153 case LOOP : return 12;
\n+
154 case QUADS : return 4;
\n+
155 case GREGORY :
\n+
156 case GREGORY_BOUNDARY : return GetGregoryPatchSize();
\n+
157 case GREGORY_BASIS : return GetGregoryBasisPatchSize();
\n+
158 case GREGORY_TRIANGLE : return 18;
\n+
159 case TRIANGLES : return 3;
\n+
160 case LINES : return 2;
\n+
161 case POINTS : return 1;
\n+
162 default : return -1;
\n+
163 }
\n+
164}
\n+
165
\n+
166// Returns the number of face-varying control vertices expected for a patch of this type
\n+
167inline short
\n+
168PatchDescriptor::GetNumFVarControlVertices( Type type ) {
\n+
169 return PatchDescriptor::GetNumControlVertices(type);
\n+
170}
\n+
171
\n+
172// Allows ordering of patches by type
\n+
173inline bool
\n+
174PatchDescriptor::operator < ( PatchDescriptor const other ) const {
\n+
175 return (_type < other._type);
\n+
176}
\n+
177
\n+
178// True if the descriptors are identical
\n+
179inline bool
\n+
180PatchDescriptor::operator == ( PatchDescriptor const other ) const {
\n+
181 return _type == other._type;
\n+
182}
\n+
183
\n+
184
\n+
185
\n+
186} // end namespace Far
\n+
187
\n+
188} // end namespace OPENSUBDIV_VERSION
\n+
189using namespace OPENSUBDIV_VERSION;
\n+
190
\n+
191} // end namespace OpenSubdiv
\n
192
\n-
194 bool NonQuadRoot() const { return (unpack(field1,1,4) != 0); }
\n-
195
\n-
197 unsigned short GetDepth() const { return (unsigned short)unpack(field1,4,0); }
\n-
198
\n-
200 float GetParamFraction() const;
\n-
201
\n-
208 template <typename REAL>
\n-
209 void Normalize( REAL & u, REAL & v ) const;
\n-
210 template <typename REAL>
\n-
211 void NormalizeTriangle( REAL & u, REAL & v ) const;
\n-
212
\n-
219 template <typename REAL>
\n-
220 void Unnormalize( REAL & u, REAL & v ) const;
\n-
221 template <typename REAL>
\n-
222 void UnnormalizeTriangle( REAL & u, REAL & v ) const;
\n-
223
\n-
225 bool IsTriangleRotated() const;
\n-
226
\n-
228 bool IsRegular() const { return (unpack(field1,1,5) != 0); }
\n-
229
\n-
230 unsigned int field0:32;
\n-
231 unsigned int field1:32;
\n-
232
\n-
233private:
\n-
234 unsigned int pack(unsigned int value, int width, int offset) const {
\n-
235 return (unsigned int)((value & ((1<<width)-1)) << offset);
\n-
236 }
\n-
237
\n-
238 unsigned int unpack(unsigned int value, int width, int offset) const {
\n-
239 return (unsigned int)((value >> offset) & ((1<<width)-1));
\n-
240 }
\n-
241};
\n-
242
\n-
243typedef std::vector<PatchParam> PatchParamTable;
\n-
244
\n-
245typedef Vtr::Array<PatchParam> PatchParamArray;
\n-
246typedef Vtr::ConstArray<PatchParam> ConstPatchParamArray;
\n-
247
\n-
248inline void
\n-
249PatchParam::Set(Index faceid, short u, short v,
\n-
250 unsigned short depth, bool nonquad,
\n-
251 unsigned short boundary, unsigned short transition,
\n-
252 bool regular) {
\n-
253 field0 = pack(faceid, 28, 0) |
\n-
254 pack(transition, 4, 28);
\n-
255
\n-
256 field1 = pack(u, 10, 22) |
\n-
257 pack(v, 10, 12) |
\n-
258 pack(boundary, 5, 7) |
\n-
259 pack(regular, 1, 5) |
\n-
260 pack(nonquad, 1, 4) |
\n-
261 pack(depth, 4, 0);
\n-
262}
\n-
263
\n-
264inline float
\n-
265PatchParam::GetParamFraction( ) const {
\n-
266 return 1.0f / (float)(1 << (GetDepth() - NonQuadRoot()));
\n-
267}
\n-
268
\n-
269template <typename REAL>
\n-
270inline void
\n-
271PatchParam::Normalize( REAL & u, REAL & v ) const {
\n-
272
\n-
273 REAL fracInv = (REAL)(1.0f / GetParamFraction());
\n-
274
\n-
275 u = u * fracInv - (REAL)GetU();
\n-
276 v = v * fracInv - (REAL)GetV();
\n-
277}
\n-
278
\n-
279template <typename REAL>
\n-
280inline void
\n-
281PatchParam::Unnormalize( REAL & u, REAL & v ) const {
\n-
282
\n-
283 REAL frac = (REAL)GetParamFraction();
\n-
284
\n-
285 u = (u + (REAL)GetU()) * frac;
\n-
286 v = (v + (REAL)GetV()) * frac;
\n-
287}
\n-
288
\n-
289inline bool
\n-
290PatchParam::IsTriangleRotated() const {
\n-
291
\n-
292 return (GetU() + GetV()) >= (1 << GetDepth());
\n-
293}
\n-
294
\n-
295template <typename REAL>
\n-
296inline void
\n-
297PatchParam::NormalizeTriangle( REAL & u, REAL & v ) const {
\n-
298
\n-
299 if (IsTriangleRotated()) {
\n-
300 REAL fracInv = (REAL)(1.0f / GetParamFraction());
\n-
301
\n-
302 int depthFactor = 1 << GetDepth();
\n-
303 u = (REAL)(depthFactor - GetU()) - (u * fracInv);
\n-
304 v = (REAL)(depthFactor - GetV()) - (v * fracInv);
\n-
305 } else {
\n-
306 Normalize(u, v);
\n-
307 }
\n-
308}
\n-
309
\n-
310template <typename REAL>
\n-
311inline void
\n-
312PatchParam::UnnormalizeTriangle( REAL & u, REAL & v ) const {
\n-
313
\n-
314 if (IsTriangleRotated()) {
\n-
315 REAL frac = GetParamFraction();
\n-
316
\n-
317 int depthFactor = 1 << GetDepth();
\n-
318 u = ((REAL)(depthFactor - GetU()) - u) * frac;
\n-
319 v = ((REAL)(depthFactor - GetV()) - v) * frac;
\n-
320 } else {
\n-
321 Unnormalize(u, v);
\n-
322 }
\n-
323}
\n-
324
\n-
325} // end namespace Far
\n-
326
\n-
327} // end namespace OPENSUBDIV_VERSION
\n-
328using namespace OPENSUBDIV_VERSION;
\n-
329
\n-
330} // end namespace OpenSubdiv
\n-
331
\n-
332#endif /* OPENSUBDIV3_FAR_PATCH_PARAM */
\n+
193#endif /* OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H */
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstPatchParamArray
Vtr::ConstArray< PatchParam > ConstPatchParamArray
Definition: patchParam.h:246
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParamTable
std::vector< PatchParam > PatchParamTable
Definition: patchParam.h:243
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParamArray
Vtr::Array< PatchParam > PatchParamArray
Definition: patchParam.h:245
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::IsTriangleRotated
bool IsTriangleRotated() const
Returns if a triangular patch is parametrically rotated 180 degrees.
Definition: patchParam.h:290
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::IsRegular
bool IsRegular() const
Returns whether the patch is regular.
Definition: patchParam.h:228
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::NormalizeTriangle
void NormalizeTriangle(REAL &u, REAL &v) const
Definition: patchParam.h:297
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetBoundary
unsigned short GetBoundary() const
Returns the boundary edge encoding for the patch.
Definition: patchParam.h:191
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetTransition
unsigned short GetTransition() const
Returns the transition edge encoding for the patch.
Definition: patchParam.h:188
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetParamFraction
float GetParamFraction() const
Returns the fraction of unit parametric space covered by this face.
Definition: patchParam.h:265
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetU
unsigned short GetU() const
Returns the log2 value of the u parameter at the first corner of the patch.
Definition: patchParam.h:181
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetFaceId
Index GetFaceId() const
Returns the faceid.
Definition: patchParam.h:177
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetDepth
unsigned short GetDepth() const
Returns the level of subdivision of the patch.
Definition: patchParam.h:197
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::NonQuadRoot
bool NonQuadRoot() const
True if the parent base face is a non-quad.
Definition: patchParam.h:194
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Set
void Set(Index faceid, short u, short v, unsigned short depth, bool nonquad, unsigned short boundary, unsigned short transition, bool regular=false)
Sets the values of the bit fields.
Definition: patchParam.h:249
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Clear
void Clear()
Resets everything to 0.
Definition: patchParam.h:174
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Unnormalize
void Unnormalize(REAL &u, REAL &v) const
A (u,v) pair in a normalized parametric space is mapped back into the fraction of parametric space co...
Definition: patchParam.h:281
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::UnnormalizeTriangle
void UnnormalizeTriangle(REAL &u, REAL &v) const
Definition: patchParam.h:312
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetV
unsigned short GetV() const
Returns the log2 value of the v parameter at the first corner of the patch.
Definition: patchParam.h:185
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Normalize
void Normalize(REAL &u, REAL &v) const
A (u,v) pair in the fraction of parametric space covered by this face is mapped into a normalized par...
Definition: patchParam.h:271
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstPatchDescriptorArray
Vtr::ConstArray< PatchDescriptor > ConstPatchDescriptorArray
Definition: patchDescriptor.h:146
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SchemeType
SchemeType
Enumerated type for all subdivision schemes supported by OpenSubdiv.
Definition: types.h:37
\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::LOOP
@ LOOP
regular triangular patch for the Loop scheme
Definition: patchDescriptor.h:57
\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::REGULAR
@ REGULAR
regular B-Spline patch for the Catmark scheme
Definition: patchDescriptor.h:59
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::IsAdaptive
bool IsAdaptive() const
Returns true if the type is an adaptive patch.
Definition: patchDescriptor.h:97
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetType
Type GetType() const
Returns the type of the patch.
Definition: patchDescriptor.h:87
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetGregoryBasisPatchSize
static short GetGregoryBasisPatchSize()
Number of control vertices of Gregory patch basis (20)
Definition: patchDescriptor.h:126
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetNumControlVertices
short GetNumControlVertices() const
Returns the number of control vertices expected for a patch of the type described.
Definition: patchDescriptor.h:110
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::PatchDescriptor
PatchDescriptor(PatchDescriptor const &d)
Copy Constructor.
Definition: patchDescriptor.h:77
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator==
bool operator==(PatchDescriptor const other) const
True if the descriptors are identical.
Definition: patchDescriptor.h:180
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetRegularPatchSize
static short GetRegularPatchSize()
Number of control vertices of Regular Patches in table.
Definition: patchDescriptor.h:120
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator=
PatchDescriptor & operator=(PatchDescriptor const &d)
Assignment operator.
Definition: patchDescriptor.h:81
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::IsAdaptive
static bool IsAdaptive(Type type)
Returns true if the type is an adaptive (non-linear) patch.
Definition: patchDescriptor.h:92
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator<
bool operator<(PatchDescriptor const other) const
Allows ordering of patches by type.
Definition: patchDescriptor.h:174
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetGregoryPatchSize
static short GetGregoryPatchSize()
Number of control vertices of Gregory (and Gregory Boundary) Patches in table.
Definition: patchDescriptor.h:123
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetAdaptivePatchDescriptors
static Vtr::ConstArray< PatchDescriptor > GetAdaptivePatchDescriptors(Sdc::SchemeType type)
Returns a vector of all the legal patch descriptors for the given adaptive subdivision scheme.
\n+\n \n-\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-patchParam.h\n+patchDescriptor.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,262 +30,280 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_FAR_PATCH_PARAM_H\n- 26#define OPENSUBDIV3_FAR_PATCH_PARAM_H\n+ 25#ifndef OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H\n+ 26#define OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H\n 27\n 28#include \"../version.h\"\n 29\n 30#include \"../far/types.h\"\n- 31\n- 32namespace OpenSubdiv {\n- 33namespace OPENSUBDIV_VERSION {\n+ 31#include \"../sdc/types.h\"\n+ 32\n+ 33#include \n 34\n- 35namespace Far {\n- 36\n- 101\n-152struct PatchParam {\n- 168 void Set(Index faceid, short u, short v,\n- 169 unsigned short depth, bool nonquad,\n- 170 unsigned short boundary, unsigned short transition,\n- 171 bool regular = false);\n- 172\n-174 void Clear() { field0 = field1 = 0; }\n- 175\n-177 Index GetFaceId() const { return Index(unpack(field0,28,0)); }\n- 178\n-181 unsigned short GetU() const { return (unsigned short)unpack(field1,10,22);\n-}\n- 182\n-185 unsigned short GetV() const { return (unsigned short)unpack(field1,10,12);\n-}\n- 186\n-188 unsigned short GetTransition() const { return (unsigned short)unpack\n-(field0,4,28); }\n- 189\n-191 unsigned short GetBoundary() const { return (unsigned short)unpack\n-(field1,5,7); }\n+ 35namespace OpenSubdiv {\n+ 36namespace OPENSUBDIV_VERSION {\n+ 37\n+ 38namespace Far {\n+ 39\n+44class PatchDescriptor {\n+ 45\n+ 46public:\n+ 47\n+48 enum Type {\n+49 NON_PATCH = 0,\n+ 50\n+51 POINTS,\n+52 LINES,\n+ 53\n+54 QUADS,\n+55 TRIANGLES,\n+ 56\n+57 LOOP,\n+ 58\n+59 REGULAR,\n+60 GREGORY,\n+61 GREGORY_BOUNDARY,\n+62 GREGORY_BASIS,\n+ 63 GREGORY_TRIANGLE\n+64 };\n+ 65\n+ 66public:\n+ 67\n+69 PatchDescriptor() :\n+ 70 _type(NON_PATCH) { }\n+ 71\n+73 PatchDescriptor(int type) :\n+ 74 _type(type) { }\n+ 75\n+77 PatchDescriptor( PatchDescriptor const & d ) :\n+ 78 _type(d.GetType()) { }\n+ 79\n+81 PatchDescriptor & operator=( PatchDescriptor const & d ) {\n+ 82 _type = d.GetType();\n+ 83 return *this;\n+ 84 }\n+ 85\n+87 Type GetType() const {\n+ 88 return (Type)_type;\n+ 89 }\n+ 90\n+92 static inline bool IsAdaptive(Type type) {\n+ 93 return type > TRIANGLES;\n+ 94 }\n+ 95\n+97 bool IsAdaptive() const {\n+ 98 return IsAdaptive( this->GetType() );\n+ 99 }\n+ 100\n+ 103 static inline short GetNumControlVertices( Type t );\n+ 104\n+ 106 static inline short GetNumFVarControlVertices( Type t );\n+ 107\n+110 short GetNumControlVertices() const {\n+ 111 return GetNumControlVertices( this->GetType() );\n+ 112 }\n+ 113\n+115 short GetNumFVarControlVertices() const {\n+ 116 return GetNumFVarControlVertices( this->GetType() );\n+ 117 }\n+ 118\n+120 static short GetRegularPatchSize() { return 16; }\n+ 121\n+123 static short GetGregoryPatchSize() { return 4; }\n+ 124\n+126 static short GetGregoryBasisPatchSize() { return 20; }\n+ 127\n+ 128\n+131 static Vtr::ConstArray GetAdaptivePatchDescriptors(Sdc::\n+SchemeType type);\n+ 132\n+ 134 inline bool operator_<( PatchDescriptor const other ) const;\n+ 135\n+ 137 inline bool operator_==( PatchDescriptor const other ) const;\n+ 138\n+ 139 // debug helper\n+140 void print() const;\n+ 141\n+ 142private:\n+ 143 unsigned int _type;\n+ 144};\n+ 145\n+146typedef Vtr::ConstArray ConstPatchDescriptorArray;\n+ 147\n+ 148// Returns the number of control vertices expected for a patch of this type\n+ 149inline short\n+150PatchDescriptor::GetNumControlVertices( Type type ) {\n+ 151 switch (type) {\n+ 152 case REGULAR : return GetRegularPatchSize();\n+ 153 case LOOP : return 12;\n+ 154 case QUADS : return 4;\n+ 155 case GREGORY :\n+ 156 case GREGORY_BOUNDARY : return GetGregoryPatchSize();\n+ 157 case GREGORY_BASIS : return GetGregoryBasisPatchSize();\n+ 158 case GREGORY_TRIANGLE : return 18;\n+ 159 case TRIANGLES : return 3;\n+ 160 case LINES : return 2;\n+ 161 case POINTS : return 1;\n+ 162 default : return -1;\n+ 163 }\n+ 164}\n+ 165\n+ 166// Returns the number of face-varying control vertices expected for a patch\n+of this type\n+ 167inline short\n+168PatchDescriptor::GetNumFVarControlVertices( Type type ) {\n+ 169 return PatchDescriptor::GetNumControlVertices(type);\n+ 170}\n+ 171\n+ 172// Allows ordering of patches by type\n+173inline bool\n+ 174PatchDescriptor::operator_<( PatchDescriptor const other ) const {\n+ 175 return (_type < other._type);\n+ 176}\n+ 177\n+ 178// True if the descriptors are identical\n+ 179inline bool\n+180PatchDescriptor::operator_==( PatchDescriptor const other ) const {\n+ 181 return _type == other._type;\n+ 182}\n+ 183\n+ 184\n+ 185\n+ 186} // end namespace Far\n+ 187\n+ 188} // end namespace OPENSUBDIV_VERSION\n+ 189using namespace OPENSUBDIV_VERSION;\n+ 190\n+ 191} // end namespace OpenSubdiv\n 192\n-194 bool NonQuadRoot() const { return (unpack(field1,1,4) != 0); }\n- 195\n-197 unsigned short GetDepth() const { return (unsigned short)unpack\n-(field1,4,0); }\n- 198\n- 200 float GetParamFraction() const;\n- 201\n- 208 template \n- 209 void Normalize( REAL & u, REAL & v ) const;\n- 210 template \n- 211 void NormalizeTriangle( REAL & u, REAL & v ) const;\n- 212\n- 219 template \n- 220 void Unnormalize( REAL & u, REAL & v ) const;\n- 221 template \n- 222 void UnnormalizeTriangle( REAL & u, REAL & v ) const;\n- 223\n- 225 bool IsTriangleRotated() const;\n- 226\n-228 bool IsRegular() const { return (unpack(field1,1,5) != 0); }\n- 229\n-230 unsigned int field0:32;\n-231 unsigned int field1:32;\n- 232\n- 233private:\n- 234 unsigned int pack(unsigned int value, int width, int offset) const {\n- 235 return (unsigned int)((value & ((1<> offset) & ((1< PatchParamTable;\n- 244\n-245typedef Vtr::Array PatchParamArray;\n-246typedef Vtr::ConstArray ConstPatchParamArray;\n- 247\n- 248inline void\n-249PatchParam::Set(Index faceid, short u, short v,\n- 250 unsigned short depth, bool nonquad,\n- 251 unsigned short boundary, unsigned short transition,\n- 252 bool regular) {\n- 253 field0 = pack(faceid, 28, 0) |\n- 254 pack(transition, 4, 28);\n- 255\n- 256 field1 = pack(u, 10, 22) |\n- 257 pack(v, 10, 12) |\n- 258 pack(boundary, 5, 7) |\n- 259 pack(regular, 1, 5) |\n- 260 pack(nonquad, 1, 4) |\n- 261 pack(depth, 4, 0);\n- 262}\n- 263\n- 264inline float\n-265PatchParam::GetParamFraction( ) const {\n- 266 return 1.0f / (float)(1 << (GetDepth() - NonQuadRoot()));\n- 267}\n- 268\n- 269template \n- 270inline void\n-271PatchParam::Normalize( REAL & u, REAL & v ) const {\n- 272\n- 273 REAL fracInv = (REAL)(1.0f / GetParamFraction());\n- 274\n- 275 u = u * fracInv - (REAL)GetU();\n- 276 v = v * fracInv - (REAL)GetV();\n- 277}\n- 278\n- 279template \n- 280inline void\n-281PatchParam::Unnormalize( REAL & u, REAL & v ) const {\n- 282\n- 283 REAL frac = (REAL)GetParamFraction();\n- 284\n- 285 u = (u + (REAL)GetU()) * frac;\n- 286 v = (v + (REAL)GetV()) * frac;\n- 287}\n- 288\n- 289inline bool\n-290PatchParam::IsTriangleRotated() const {\n- 291\n- 292 return (GetU() + GetV()) >= (1 << GetDepth());\n- 293}\n- 294\n- 295template \n- 296inline void\n-297PatchParam::NormalizeTriangle( REAL & u, REAL & v ) const {\n- 298\n- 299 if (IsTriangleRotated()) {\n- 300 REAL fracInv = (REAL)(1.0f / GetParamFraction());\n- 301\n- 302 int depthFactor = 1 << GetDepth();\n- 303 u = (REAL)(depthFactor - GetU()) - (u * fracInv);\n- 304 v = (REAL)(depthFactor - GetV()) - (v * fracInv);\n- 305 } else {\n- 306 Normalize(u, v);\n- 307 }\n- 308}\n- 309\n- 310template \n- 311inline void\n-312PatchParam::UnnormalizeTriangle( REAL & u, REAL & v ) const {\n- 313\n- 314 if (IsTriangleRotated()) {\n- 315 REAL frac = GetParamFraction();\n- 316\n- 317 int depthFactor = 1 << GetDepth();\n- 318 u = ((REAL)(depthFactor - GetU()) - u) * frac;\n- 319 v = ((REAL)(depthFactor - GetV()) - v) * frac;\n- 320 } else {\n- 321 Unnormalize(u, v);\n- 322 }\n- 323}\n- 324\n- 325} // end namespace Far\n- 326\n- 327} // end namespace OPENSUBDIV_VERSION\n- 328using namespace OPENSUBDIV_VERSION;\n- 329\n- 330} // end namespace OpenSubdiv\n- 331\n- 332#endif /* OPENSUBDIV3_FAR_PATCH_PARAM */\n+ 193#endif /* OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H */\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstPatchParamArray\n-Vtr::ConstArray< PatchParam > ConstPatchParamArray\n-Definition: patchParam.h:246\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParamTable\n-std::vector< PatchParam > PatchParamTable\n-Definition: patchParam.h:243\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::Index\n-Vtr::Index Index\n-Definition: types.h:41\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParamArray\n-Vtr::Array< PatchParam > PatchParamArray\n-Definition: patchParam.h:245\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam\n-Patch parameterization.\n-Definition: patchParam.h:152\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::IsTriangleRotated\n-bool IsTriangleRotated() const\n-Returns if a triangular patch is parametrically rotated 180 degrees.\n-Definition: patchParam.h:290\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::IsRegular\n-bool IsRegular() const\n-Returns whether the patch is regular.\n-Definition: patchParam.h:228\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::NormalizeTriangle\n-void NormalizeTriangle(REAL &u, REAL &v) const\n-Definition: patchParam.h:297\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetBoundary\n-unsigned short GetBoundary() const\n-Returns the boundary edge encoding for the patch.\n-Definition: patchParam.h:191\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetTransition\n-unsigned short GetTransition() const\n-Returns the transition edge encoding for the patch.\n-Definition: patchParam.h:188\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetParamFraction\n-float GetParamFraction() const\n-Returns the fraction of unit parametric space covered by this face.\n-Definition: patchParam.h:265\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetU\n-unsigned short GetU() const\n-Returns the log2 value of the u parameter at the first corner of the patch.\n-Definition: patchParam.h:181\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetFaceId\n-Index GetFaceId() const\n-Returns the faceid.\n-Definition: patchParam.h:177\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetDepth\n-unsigned short GetDepth() const\n-Returns the level of subdivision of the patch.\n-Definition: patchParam.h:197\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::NonQuadRoot\n-bool NonQuadRoot() const\n-True if the parent base face is a non-quad.\n-Definition: patchParam.h:194\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Set\n-void Set(Index faceid, short u, short v, unsigned short depth, bool nonquad,\n-unsigned short boundary, unsigned short transition, bool regular=false)\n-Sets the values of the bit fields.\n-Definition: patchParam.h:249\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Clear\n-void Clear()\n-Resets everything to 0.\n-Definition: patchParam.h:174\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::field1\n-unsigned int field1\n-Definition: patchParam.h:231\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Unnormalize\n-void Unnormalize(REAL &u, REAL &v) const\n-A (u,v) pair in a normalized parametric space is mapped back into the fraction\n-of parametric space co...\n-Definition: patchParam.h:281\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::UnnormalizeTriangle\n-void UnnormalizeTriangle(REAL &u, REAL &v) const\n-Definition: patchParam.h:312\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetV\n-unsigned short GetV() const\n-Returns the log2 value of the v parameter at the first corner of the patch.\n-Definition: patchParam.h:185\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::field0\n-unsigned int field0\n-Definition: patchParam.h:230\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Normalize\n-void Normalize(REAL &u, REAL &v) const\n-A (u,v) pair in the fraction of parametric space covered by this face is mapped\n-into a normalized par...\n-Definition: patchParam.h:271\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstPatchDescriptorArray\n+Vtr::ConstArray< PatchDescriptor > ConstPatchDescriptorArray\n+Definition: patchDescriptor.h:146\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SchemeType\n+SchemeType\n+Enumerated type for all subdivision schemes supported by OpenSubdiv.\n+Definition: types.h:37\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor\n+Describes the type of a patch.\n+Definition: patchDescriptor.h:44\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n+Type\n+Definition: patchDescriptor.h:48\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::LINES\n+@ LINES\n+lines (useful for cage drawing)\n+Definition: patchDescriptor.h:52\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::QUADS\n+@ QUADS\n+4-sided quadrilateral (bilinear)\n+Definition: patchDescriptor.h:54\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::NON_PATCH\n+@ NON_PATCH\n+undefined\n+Definition: patchDescriptor.h:49\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::LOOP\n+@ LOOP\n+regular triangular patch for the Loop scheme\n+Definition: patchDescriptor.h:57\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_TRIANGLE\n+@ GREGORY_TRIANGLE\n+Definition: patchDescriptor.h:63\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::TRIANGLES\n+@ TRIANGLES\n+3-sided triangle\n+Definition: patchDescriptor.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_BOUNDARY\n+@ GREGORY_BOUNDARY\n+Definition: patchDescriptor.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_BASIS\n+@ GREGORY_BASIS\n+Definition: patchDescriptor.h:62\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::POINTS\n+@ POINTS\n+points (useful for cage drawing)\n+Definition: patchDescriptor.h:51\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY\n+@ GREGORY\n+Definition: patchDescriptor.h:60\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::REGULAR\n+@ REGULAR\n+regular B-Spline patch for the Catmark scheme\n+Definition: patchDescriptor.h:59\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::IsAdaptive\n+bool IsAdaptive() const\n+Returns true if the type is an adaptive patch.\n+Definition: patchDescriptor.h:97\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetType\n+Type GetType() const\n+Returns the type of the patch.\n+Definition: patchDescriptor.h:87\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetGregoryBasisPatchSize\n+static short GetGregoryBasisPatchSize()\n+Number of control vertices of Gregory patch basis (20)\n+Definition: patchDescriptor.h:126\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetNumControlVertices\n+short GetNumControlVertices() const\n+Returns the number of control vertices expected for a patch of the type\n+described.\n+Definition: patchDescriptor.h:110\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::PatchDescriptor\n+PatchDescriptor(PatchDescriptor const &d)\n+Copy Constructor.\n+Definition: patchDescriptor.h:77\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator==\n+bool operator==(PatchDescriptor const other) const\n+True if the descriptors are identical.\n+Definition: patchDescriptor.h:180\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetRegularPatchSize\n+static short GetRegularPatchSize()\n+Number of control vertices of Regular Patches in table.\n+Definition: patchDescriptor.h:120\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetNumFVarControlVertices\n+short GetNumFVarControlVertices() const\n+Deprecated.\n+Definition: patchDescriptor.h:115\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::PatchDescriptor\n+PatchDescriptor(int type)\n+Constructor.\n+Definition: patchDescriptor.h:73\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator=\n+PatchDescriptor & operator=(PatchDescriptor const &d)\n+Assignment operator.\n+Definition: patchDescriptor.h:81\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::IsAdaptive\n+static bool IsAdaptive(Type type)\n+Returns true if the type is an adaptive (non-linear) patch.\n+Definition: patchDescriptor.h:92\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator<\n+bool operator<(PatchDescriptor const other) const\n+Allows ordering of patches by type.\n+Definition: patchDescriptor.h:174\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetGregoryPatchSize\n+static short GetGregoryPatchSize()\n+Number of control vertices of Gregory (and Gregory Boundary) Patches in table.\n+Definition: patchDescriptor.h:123\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::print\n+void print() const\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::\n+GetAdaptivePatchDescriptors\n+static Vtr::ConstArray< PatchDescriptor > GetAdaptivePatchDescriptors(Sdc::\n+SchemeType type)\n+Returns a vector of all the legal patch descriptors for the given adaptive\n+subdivision scheme.\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::PatchDescriptor\n+PatchDescriptor()\n+Default constructor.\n+Definition: patchDescriptor.h:69\n OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstArray\n Definition: array.h:53\n-OpenSubdiv::OPENSUBDIV_VERSION::Vtr::Array\n-Definition: array.h:105\n * opensubdiv\n * far\n- * patchParam.h\n+ * patchDescriptor.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00740.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00740.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/patchDescriptor.h File Reference\n+OpenSubdiv: opensubdiv/far/patchParam.h File Reference\n \n \n \n \n \n \n \n@@ -84,48 +84,50 @@\n \n \n
\n
\n Classes |\n Namespaces |\n Typedefs
\n-
patchDescriptor.h File Reference
\n+
patchParam.h File Reference
\n
\n
\n
#include "../version.h"
\n #include "../far/types.h"
\n-#include "../sdc/types.h"
\n-#include <vector>
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n+\n+\n \n

\n Classes

class  PatchDescriptor
 Describes the type of a patch. More...
struct  PatchParam
 Patch parameterization. More...
 
\n \n \n \n \n \n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
\n \n-\n-\n+\n+\n+\n+\n+\n+\n

\n Typedefs

typedef Vtr::ConstArray< PatchDescriptor > ConstPatchDescriptorArray
 
typedef std::vector< PatchParam > PatchParamTable
 
typedef Vtr::Array< PatchParam > PatchParamArray
 
typedef Vtr::ConstArray< PatchParam > ConstPatchParamArray
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,31 +5,33 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces | Typedefs\n-patchDescriptor.h File Reference\n+patchParam.h File Reference\n #include \"../version.h\"\n #include \"../far/types.h\"\n-#include \"../sdc/types.h\"\n-#include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0PatchDescriptor\n-\u00a0 Describes the type of a patch. More...\n+struct \u00a0PatchParam\n+\u00a0 Patch parameterization. More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n Typedefs\n-typedef Vtr::ConstArray< PatchDescriptor >\u00a0ConstPatchDescriptorArray\n+ typedef std::vector< PatchParam >\u00a0PatchParamTable\n+\u00a0\n+ typedef Vtr::Array< PatchParam >\u00a0PatchParamArray\n+\u00a0\n+typedef Vtr::ConstArray< PatchParam >\u00a0ConstPatchParamArray\n \u00a0\n * opensubdiv\n * far\n- * patchDescriptor.h\n+ * patchParam.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00740.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00740.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,6 @@\n var a00740 = [\n- [\"PatchDescriptor\", \"a00973.html\", \"a00973\"],\n- [\"ConstPatchDescriptorArray\", \"a00740.html#ad258c2d8f7871bc0d40cf3a9ded6e08a\", null]\n+ [\"PatchParam\", \"a00989.html\", \"a00989\"],\n+ [\"ConstPatchParamArray\", \"a00740.html#a3139c7c52ddf26b2403e8e95ff63eeb3\", null],\n+ [\"PatchParamArray\", \"a00740.html#abe07210a029a01571c54276ae28b4d88\", null],\n+ [\"PatchParamTable\", \"a00740.html#a46ea68db89cc3d55c4191ef7648281ce\", null]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00740_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00740_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/far/patchDescriptor.h Source File\n+OpenSubdiv: opensubdiv/far/patchParam.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
patchDescriptor.h
\n+
patchParam.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,196 +107,198 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H
\n-
26#define OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H
\n+
25#ifndef OPENSUBDIV3_FAR_PATCH_PARAM_H
\n+
26#define OPENSUBDIV3_FAR_PATCH_PARAM_H
\n
27
\n
28#include "../version.h"
\n
29
\n
30#include "../far/types.h"
\n-
31#include "../sdc/types.h"
\n-
32
\n-
33#include <vector>
\n+
31
\n+
32namespace OpenSubdiv {
\n+
33namespace OPENSUBDIV_VERSION {
\n
34
\n-
35namespace OpenSubdiv {
\n-
36namespace OPENSUBDIV_VERSION {
\n-
37
\n-
38namespace Far {
\n-
39
\n-
44class PatchDescriptor {
\n-
45
\n-
46public:
\n-
47
\n-
48 enum Type {
\n-
49 NON_PATCH = 0,
\n-
50
\n-
51 POINTS,
\n-
52 LINES,
\n-
53
\n-
54 QUADS,
\n-
55 TRIANGLES,
\n-
56
\n-
57 LOOP,
\n-
58
\n-
59 REGULAR,
\n-
60 GREGORY,
\n-
61 GREGORY_BOUNDARY,
\n-
62 GREGORY_BASIS,
\n-
63 GREGORY_TRIANGLE
\n-
64 };
\n-
65
\n-
66public:
\n-
67
\n-
69 PatchDescriptor() :
\n-
70 _type(NON_PATCH) { }
\n-
71
\n-
73 PatchDescriptor(int type) :
\n-
74 _type(type) { }
\n-
75
\n-
77 PatchDescriptor( PatchDescriptor const & d ) :
\n-
78 _type(d.GetType()) { }
\n-
79
\n-
81 PatchDescriptor & operator=( PatchDescriptor const & d ) {
\n-
82 _type = d.GetType();
\n-
83 return *this;
\n-
84 }
\n-
85
\n-
87 Type GetType() const {
\n-
88 return (Type)_type;
\n-
89 }
\n-
90
\n-
92 static inline bool IsAdaptive(Type type) {
\n-
93 return type > TRIANGLES;
\n-
94 }
\n-
95
\n-
97 bool IsAdaptive() const {
\n-
98 return IsAdaptive( this->GetType() );
\n-
99 }
\n-
100
\n-
103 static inline short GetNumControlVertices( Type t );
\n-
104
\n-
106 static inline short GetNumFVarControlVertices( Type t );
\n-
107
\n-
110 short GetNumControlVertices() const {
\n-
111 return GetNumControlVertices( this->GetType() );
\n-
112 }
\n-
113
\n-
115 short GetNumFVarControlVertices() const {
\n-
116 return GetNumFVarControlVertices( this->GetType() );
\n-
117 }
\n-
118
\n-
120 static short GetRegularPatchSize() { return 16; }
\n-
121
\n-
123 static short GetGregoryPatchSize() { return 4; }
\n-
124
\n-
126 static short GetGregoryBasisPatchSize() { return 20; }
\n-
127
\n-
128
\n-
131 static Vtr::ConstArray<PatchDescriptor> GetAdaptivePatchDescriptors(Sdc::SchemeType type);
\n-
132
\n-
134 inline bool operator < ( PatchDescriptor const other ) const;
\n-
135
\n-
137 inline bool operator == ( PatchDescriptor const other ) const;
\n-
138
\n-
139 // debug helper
\n-
140 void print() const;
\n-
141
\n-
142private:
\n-
143 unsigned int _type;
\n-
144};
\n-
145
\n-
146typedef Vtr::ConstArray<PatchDescriptor> ConstPatchDescriptorArray;
\n-
147
\n-
148// Returns the number of control vertices expected for a patch of this type
\n-
149inline short
\n-
150PatchDescriptor::GetNumControlVertices( Type type ) {
\n-
151 switch (type) {
\n-
152 case REGULAR : return GetRegularPatchSize();
\n-
153 case LOOP : return 12;
\n-
154 case QUADS : return 4;
\n-
155 case GREGORY :
\n-
156 case GREGORY_BOUNDARY : return GetGregoryPatchSize();
\n-
157 case GREGORY_BASIS : return GetGregoryBasisPatchSize();
\n-
158 case GREGORY_TRIANGLE : return 18;
\n-
159 case TRIANGLES : return 3;
\n-
160 case LINES : return 2;
\n-
161 case POINTS : return 1;
\n-
162 default : return -1;
\n-
163 }
\n-
164}
\n-
165
\n-
166// Returns the number of face-varying control vertices expected for a patch of this type
\n-
167inline short
\n-
168PatchDescriptor::GetNumFVarControlVertices( Type type ) {
\n-
169 return PatchDescriptor::GetNumControlVertices(type);
\n-
170}
\n-
171
\n-
172// Allows ordering of patches by type
\n-
173inline bool
\n-
174PatchDescriptor::operator < ( PatchDescriptor const other ) const {
\n-
175 return (_type < other._type);
\n-
176}
\n-
177
\n-
178// True if the descriptors are identical
\n-
179inline bool
\n-
180PatchDescriptor::operator == ( PatchDescriptor const other ) const {
\n-
181 return _type == other._type;
\n-
182}
\n-
183
\n-
184
\n-
185
\n-
186} // end namespace Far
\n-
187
\n-
188} // end namespace OPENSUBDIV_VERSION
\n-
189using namespace OPENSUBDIV_VERSION;
\n-
190
\n-
191} // end namespace OpenSubdiv
\n+
35namespace Far {
\n+
36
\n+
101
\n+
152struct PatchParam {
\n+
168 void Set(Index faceid, short u, short v,
\n+
169 unsigned short depth, bool nonquad,
\n+
170 unsigned short boundary, unsigned short transition,
\n+
171 bool regular = false);
\n+
172
\n+
174 void Clear() { field0 = field1 = 0; }
\n+
175
\n+
177 Index GetFaceId() const { return Index(unpack(field0,28,0)); }
\n+
178
\n+
181 unsigned short GetU() const { return (unsigned short)unpack(field1,10,22); }
\n+
182
\n+
185 unsigned short GetV() const { return (unsigned short)unpack(field1,10,12); }
\n+
186
\n+
188 unsigned short GetTransition() const { return (unsigned short)unpack(field0,4,28); }
\n+
189
\n+
191 unsigned short GetBoundary() const { return (unsigned short)unpack(field1,5,7); }
\n
192
\n-
193#endif /* OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H */
\n+
194 bool NonQuadRoot() const { return (unpack(field1,1,4) != 0); }
\n+
195
\n+
197 unsigned short GetDepth() const { return (unsigned short)unpack(field1,4,0); }
\n+
198
\n+
200 float GetParamFraction() const;
\n+
201
\n+
208 template <typename REAL>
\n+
209 void Normalize( REAL & u, REAL & v ) const;
\n+
210 template <typename REAL>
\n+
211 void NormalizeTriangle( REAL & u, REAL & v ) const;
\n+
212
\n+
219 template <typename REAL>
\n+
220 void Unnormalize( REAL & u, REAL & v ) const;
\n+
221 template <typename REAL>
\n+
222 void UnnormalizeTriangle( REAL & u, REAL & v ) const;
\n+
223
\n+
225 bool IsTriangleRotated() const;
\n+
226
\n+
228 bool IsRegular() const { return (unpack(field1,1,5) != 0); }
\n+
229
\n+
230 unsigned int field0:32;
\n+
231 unsigned int field1:32;
\n+
232
\n+
233private:
\n+
234 unsigned int pack(unsigned int value, int width, int offset) const {
\n+
235 return (unsigned int)((value & ((1<<width)-1)) << offset);
\n+
236 }
\n+
237
\n+
238 unsigned int unpack(unsigned int value, int width, int offset) const {
\n+
239 return (unsigned int)((value >> offset) & ((1<<width)-1));
\n+
240 }
\n+
241};
\n+
242
\n+
243typedef std::vector<PatchParam> PatchParamTable;
\n+
244
\n+
245typedef Vtr::Array<PatchParam> PatchParamArray;
\n+
246typedef Vtr::ConstArray<PatchParam> ConstPatchParamArray;
\n+
247
\n+
248inline void
\n+
249PatchParam::Set(Index faceid, short u, short v,
\n+
250 unsigned short depth, bool nonquad,
\n+
251 unsigned short boundary, unsigned short transition,
\n+
252 bool regular) {
\n+
253 field0 = pack(faceid, 28, 0) |
\n+
254 pack(transition, 4, 28);
\n+
255
\n+
256 field1 = pack(u, 10, 22) |
\n+
257 pack(v, 10, 12) |
\n+
258 pack(boundary, 5, 7) |
\n+
259 pack(regular, 1, 5) |
\n+
260 pack(nonquad, 1, 4) |
\n+
261 pack(depth, 4, 0);
\n+
262}
\n+
263
\n+
264inline float
\n+
265PatchParam::GetParamFraction( ) const {
\n+
266 return 1.0f / (float)(1 << (GetDepth() - NonQuadRoot()));
\n+
267}
\n+
268
\n+
269template <typename REAL>
\n+
270inline void
\n+
271PatchParam::Normalize( REAL & u, REAL & v ) const {
\n+
272
\n+
273 REAL fracInv = (REAL)(1.0f / GetParamFraction());
\n+
274
\n+
275 u = u * fracInv - (REAL)GetU();
\n+
276 v = v * fracInv - (REAL)GetV();
\n+
277}
\n+
278
\n+
279template <typename REAL>
\n+
280inline void
\n+
281PatchParam::Unnormalize( REAL & u, REAL & v ) const {
\n+
282
\n+
283 REAL frac = (REAL)GetParamFraction();
\n+
284
\n+
285 u = (u + (REAL)GetU()) * frac;
\n+
286 v = (v + (REAL)GetV()) * frac;
\n+
287}
\n+
288
\n+
289inline bool
\n+
290PatchParam::IsTriangleRotated() const {
\n+
291
\n+
292 return (GetU() + GetV()) >= (1 << GetDepth());
\n+
293}
\n+
294
\n+
295template <typename REAL>
\n+
296inline void
\n+
297PatchParam::NormalizeTriangle( REAL & u, REAL & v ) const {
\n+
298
\n+
299 if (IsTriangleRotated()) {
\n+
300 REAL fracInv = (REAL)(1.0f / GetParamFraction());
\n+
301
\n+
302 int depthFactor = 1 << GetDepth();
\n+
303 u = (REAL)(depthFactor - GetU()) - (u * fracInv);
\n+
304 v = (REAL)(depthFactor - GetV()) - (v * fracInv);
\n+
305 } else {
\n+
306 Normalize(u, v);
\n+
307 }
\n+
308}
\n+
309
\n+
310template <typename REAL>
\n+
311inline void
\n+
312PatchParam::UnnormalizeTriangle( REAL & u, REAL & v ) const {
\n+
313
\n+
314 if (IsTriangleRotated()) {
\n+
315 REAL frac = GetParamFraction();
\n+
316
\n+
317 int depthFactor = 1 << GetDepth();
\n+
318 u = ((REAL)(depthFactor - GetU()) - u) * frac;
\n+
319 v = ((REAL)(depthFactor - GetV()) - v) * frac;
\n+
320 } else {
\n+
321 Unnormalize(u, v);
\n+
322 }
\n+
323}
\n+
324
\n+
325} // end namespace Far
\n+
326
\n+
327} // end namespace OPENSUBDIV_VERSION
\n+
328using namespace OPENSUBDIV_VERSION;
\n+
329
\n+
330} // end namespace OpenSubdiv
\n+
331
\n+
332#endif /* OPENSUBDIV3_FAR_PATCH_PARAM */
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstPatchDescriptorArray
Vtr::ConstArray< PatchDescriptor > ConstPatchDescriptorArray
Definition: patchDescriptor.h:146
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SchemeType
SchemeType
Enumerated type for all subdivision schemes supported by OpenSubdiv.
Definition: types.h:37
\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::LOOP
@ LOOP
regular triangular patch for the Loop scheme
Definition: patchDescriptor.h:57
\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::REGULAR
@ REGULAR
regular B-Spline patch for the Catmark scheme
Definition: patchDescriptor.h:59
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::IsAdaptive
bool IsAdaptive() const
Returns true if the type is an adaptive patch.
Definition: patchDescriptor.h:97
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetType
Type GetType() const
Returns the type of the patch.
Definition: patchDescriptor.h:87
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetGregoryBasisPatchSize
static short GetGregoryBasisPatchSize()
Number of control vertices of Gregory patch basis (20)
Definition: patchDescriptor.h:126
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetNumControlVertices
short GetNumControlVertices() const
Returns the number of control vertices expected for a patch of the type described.
Definition: patchDescriptor.h:110
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::PatchDescriptor
PatchDescriptor(PatchDescriptor const &d)
Copy Constructor.
Definition: patchDescriptor.h:77
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator==
bool operator==(PatchDescriptor const other) const
True if the descriptors are identical.
Definition: patchDescriptor.h:180
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetRegularPatchSize
static short GetRegularPatchSize()
Number of control vertices of Regular Patches in table.
Definition: patchDescriptor.h:120
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator=
PatchDescriptor & operator=(PatchDescriptor const &d)
Assignment operator.
Definition: patchDescriptor.h:81
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::IsAdaptive
static bool IsAdaptive(Type type)
Returns true if the type is an adaptive (non-linear) patch.
Definition: patchDescriptor.h:92
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator<
bool operator<(PatchDescriptor const other) const
Allows ordering of patches by type.
Definition: patchDescriptor.h:174
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetGregoryPatchSize
static short GetGregoryPatchSize()
Number of control vertices of Gregory (and Gregory Boundary) Patches in table.
Definition: patchDescriptor.h:123
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetAdaptivePatchDescriptors
static Vtr::ConstArray< PatchDescriptor > GetAdaptivePatchDescriptors(Sdc::SchemeType type)
Returns a vector of all the legal patch descriptors for the given adaptive subdivision scheme.
\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstPatchParamArray
Vtr::ConstArray< PatchParam > ConstPatchParamArray
Definition: patchParam.h:246
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParamTable
std::vector< PatchParam > PatchParamTable
Definition: patchParam.h:243
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParamArray
Vtr::Array< PatchParam > PatchParamArray
Definition: patchParam.h:245
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::IsTriangleRotated
bool IsTriangleRotated() const
Returns if a triangular patch is parametrically rotated 180 degrees.
Definition: patchParam.h:290
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::IsRegular
bool IsRegular() const
Returns whether the patch is regular.
Definition: patchParam.h:228
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::NormalizeTriangle
void NormalizeTriangle(REAL &u, REAL &v) const
Definition: patchParam.h:297
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetBoundary
unsigned short GetBoundary() const
Returns the boundary edge encoding for the patch.
Definition: patchParam.h:191
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetTransition
unsigned short GetTransition() const
Returns the transition edge encoding for the patch.
Definition: patchParam.h:188
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetParamFraction
float GetParamFraction() const
Returns the fraction of unit parametric space covered by this face.
Definition: patchParam.h:265
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetU
unsigned short GetU() const
Returns the log2 value of the u parameter at the first corner of the patch.
Definition: patchParam.h:181
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetFaceId
Index GetFaceId() const
Returns the faceid.
Definition: patchParam.h:177
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetDepth
unsigned short GetDepth() const
Returns the level of subdivision of the patch.
Definition: patchParam.h:197
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::NonQuadRoot
bool NonQuadRoot() const
True if the parent base face is a non-quad.
Definition: patchParam.h:194
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Set
void Set(Index faceid, short u, short v, unsigned short depth, bool nonquad, unsigned short boundary, unsigned short transition, bool regular=false)
Sets the values of the bit fields.
Definition: patchParam.h:249
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Clear
void Clear()
Resets everything to 0.
Definition: patchParam.h:174
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Unnormalize
void Unnormalize(REAL &u, REAL &v) const
A (u,v) pair in a normalized parametric space is mapped back into the fraction of parametric space co...
Definition: patchParam.h:281
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::UnnormalizeTriangle
void UnnormalizeTriangle(REAL &u, REAL &v) const
Definition: patchParam.h:312
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetV
unsigned short GetV() const
Returns the log2 value of the v parameter at the first corner of the patch.
Definition: patchParam.h:185
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Normalize
void Normalize(REAL &u, REAL &v) const
A (u,v) pair in the fraction of parametric space covered by this face is mapped into a normalized par...
Definition: patchParam.h:271
\n \n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-patchDescriptor.h\n+patchParam.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,280 +30,262 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H\n- 26#define OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H\n+ 25#ifndef OPENSUBDIV3_FAR_PATCH_PARAM_H\n+ 26#define OPENSUBDIV3_FAR_PATCH_PARAM_H\n 27\n 28#include \"../version.h\"\n 29\n 30#include \"../far/types.h\"\n- 31#include \"../sdc/types.h\"\n- 32\n- 33#include \n+ 31\n+ 32namespace OpenSubdiv {\n+ 33namespace OPENSUBDIV_VERSION {\n 34\n- 35namespace OpenSubdiv {\n- 36namespace OPENSUBDIV_VERSION {\n- 37\n- 38namespace Far {\n- 39\n-44class PatchDescriptor {\n- 45\n- 46public:\n- 47\n-48 enum Type {\n-49 NON_PATCH = 0,\n- 50\n-51 POINTS,\n-52 LINES,\n- 53\n-54 QUADS,\n-55 TRIANGLES,\n- 56\n-57 LOOP,\n- 58\n-59 REGULAR,\n-60 GREGORY,\n-61 GREGORY_BOUNDARY,\n-62 GREGORY_BASIS,\n- 63 GREGORY_TRIANGLE\n-64 };\n- 65\n- 66public:\n- 67\n-69 PatchDescriptor() :\n- 70 _type(NON_PATCH) { }\n- 71\n-73 PatchDescriptor(int type) :\n- 74 _type(type) { }\n- 75\n-77 PatchDescriptor( PatchDescriptor const & d ) :\n- 78 _type(d.GetType()) { }\n- 79\n-81 PatchDescriptor & operator=( PatchDescriptor const & d ) {\n- 82 _type = d.GetType();\n- 83 return *this;\n- 84 }\n- 85\n-87 Type GetType() const {\n- 88 return (Type)_type;\n- 89 }\n- 90\n-92 static inline bool IsAdaptive(Type type) {\n- 93 return type > TRIANGLES;\n- 94 }\n- 95\n-97 bool IsAdaptive() const {\n- 98 return IsAdaptive( this->GetType() );\n- 99 }\n- 100\n- 103 static inline short GetNumControlVertices( Type t );\n- 104\n- 106 static inline short GetNumFVarControlVertices( Type t );\n- 107\n-110 short GetNumControlVertices() const {\n- 111 return GetNumControlVertices( this->GetType() );\n- 112 }\n- 113\n-115 short GetNumFVarControlVertices() const {\n- 116 return GetNumFVarControlVertices( this->GetType() );\n- 117 }\n- 118\n-120 static short GetRegularPatchSize() { return 16; }\n- 121\n-123 static short GetGregoryPatchSize() { return 4; }\n- 124\n-126 static short GetGregoryBasisPatchSize() { return 20; }\n- 127\n- 128\n-131 static Vtr::ConstArray GetAdaptivePatchDescriptors(Sdc::\n-SchemeType type);\n- 132\n- 134 inline bool operator_<( PatchDescriptor const other ) const;\n- 135\n- 137 inline bool operator_==( PatchDescriptor const other ) const;\n- 138\n- 139 // debug helper\n-140 void print() const;\n- 141\n- 142private:\n- 143 unsigned int _type;\n- 144};\n- 145\n-146typedef Vtr::ConstArray ConstPatchDescriptorArray;\n- 147\n- 148// Returns the number of control vertices expected for a patch of this type\n- 149inline short\n-150PatchDescriptor::GetNumControlVertices( Type type ) {\n- 151 switch (type) {\n- 152 case REGULAR : return GetRegularPatchSize();\n- 153 case LOOP : return 12;\n- 154 case QUADS : return 4;\n- 155 case GREGORY :\n- 156 case GREGORY_BOUNDARY : return GetGregoryPatchSize();\n- 157 case GREGORY_BASIS : return GetGregoryBasisPatchSize();\n- 158 case GREGORY_TRIANGLE : return 18;\n- 159 case TRIANGLES : return 3;\n- 160 case LINES : return 2;\n- 161 case POINTS : return 1;\n- 162 default : return -1;\n- 163 }\n- 164}\n- 165\n- 166// Returns the number of face-varying control vertices expected for a patch\n-of this type\n- 167inline short\n-168PatchDescriptor::GetNumFVarControlVertices( Type type ) {\n- 169 return PatchDescriptor::GetNumControlVertices(type);\n- 170}\n- 171\n- 172// Allows ordering of patches by type\n-173inline bool\n- 174PatchDescriptor::operator_<( PatchDescriptor const other ) const {\n- 175 return (_type < other._type);\n- 176}\n- 177\n- 178// True if the descriptors are identical\n- 179inline bool\n-180PatchDescriptor::operator_==( PatchDescriptor const other ) const {\n- 181 return _type == other._type;\n- 182}\n- 183\n- 184\n- 185\n- 186} // end namespace Far\n- 187\n- 188} // end namespace OPENSUBDIV_VERSION\n- 189using namespace OPENSUBDIV_VERSION;\n- 190\n- 191} // end namespace OpenSubdiv\n+ 35namespace Far {\n+ 36\n+ 101\n+152struct PatchParam {\n+ 168 void Set(Index faceid, short u, short v,\n+ 169 unsigned short depth, bool nonquad,\n+ 170 unsigned short boundary, unsigned short transition,\n+ 171 bool regular = false);\n+ 172\n+174 void Clear() { field0 = field1 = 0; }\n+ 175\n+177 Index GetFaceId() const { return Index(unpack(field0,28,0)); }\n+ 178\n+181 unsigned short GetU() const { return (unsigned short)unpack(field1,10,22);\n+}\n+ 182\n+185 unsigned short GetV() const { return (unsigned short)unpack(field1,10,12);\n+}\n+ 186\n+188 unsigned short GetTransition() const { return (unsigned short)unpack\n+(field0,4,28); }\n+ 189\n+191 unsigned short GetBoundary() const { return (unsigned short)unpack\n+(field1,5,7); }\n 192\n- 193#endif /* OPENSUBDIV3_FAR_PATCH_DESCRIPTOR_H */\n+194 bool NonQuadRoot() const { return (unpack(field1,1,4) != 0); }\n+ 195\n+197 unsigned short GetDepth() const { return (unsigned short)unpack\n+(field1,4,0); }\n+ 198\n+ 200 float GetParamFraction() const;\n+ 201\n+ 208 template \n+ 209 void Normalize( REAL & u, REAL & v ) const;\n+ 210 template \n+ 211 void NormalizeTriangle( REAL & u, REAL & v ) const;\n+ 212\n+ 219 template \n+ 220 void Unnormalize( REAL & u, REAL & v ) const;\n+ 221 template \n+ 222 void UnnormalizeTriangle( REAL & u, REAL & v ) const;\n+ 223\n+ 225 bool IsTriangleRotated() const;\n+ 226\n+228 bool IsRegular() const { return (unpack(field1,1,5) != 0); }\n+ 229\n+230 unsigned int field0:32;\n+231 unsigned int field1:32;\n+ 232\n+ 233private:\n+ 234 unsigned int pack(unsigned int value, int width, int offset) const {\n+ 235 return (unsigned int)((value & ((1<> offset) & ((1< PatchParamTable;\n+ 244\n+245typedef Vtr::Array PatchParamArray;\n+246typedef Vtr::ConstArray ConstPatchParamArray;\n+ 247\n+ 248inline void\n+249PatchParam::Set(Index faceid, short u, short v,\n+ 250 unsigned short depth, bool nonquad,\n+ 251 unsigned short boundary, unsigned short transition,\n+ 252 bool regular) {\n+ 253 field0 = pack(faceid, 28, 0) |\n+ 254 pack(transition, 4, 28);\n+ 255\n+ 256 field1 = pack(u, 10, 22) |\n+ 257 pack(v, 10, 12) |\n+ 258 pack(boundary, 5, 7) |\n+ 259 pack(regular, 1, 5) |\n+ 260 pack(nonquad, 1, 4) |\n+ 261 pack(depth, 4, 0);\n+ 262}\n+ 263\n+ 264inline float\n+265PatchParam::GetParamFraction( ) const {\n+ 266 return 1.0f / (float)(1 << (GetDepth() - NonQuadRoot()));\n+ 267}\n+ 268\n+ 269template \n+ 270inline void\n+271PatchParam::Normalize( REAL & u, REAL & v ) const {\n+ 272\n+ 273 REAL fracInv = (REAL)(1.0f / GetParamFraction());\n+ 274\n+ 275 u = u * fracInv - (REAL)GetU();\n+ 276 v = v * fracInv - (REAL)GetV();\n+ 277}\n+ 278\n+ 279template \n+ 280inline void\n+281PatchParam::Unnormalize( REAL & u, REAL & v ) const {\n+ 282\n+ 283 REAL frac = (REAL)GetParamFraction();\n+ 284\n+ 285 u = (u + (REAL)GetU()) * frac;\n+ 286 v = (v + (REAL)GetV()) * frac;\n+ 287}\n+ 288\n+ 289inline bool\n+290PatchParam::IsTriangleRotated() const {\n+ 291\n+ 292 return (GetU() + GetV()) >= (1 << GetDepth());\n+ 293}\n+ 294\n+ 295template \n+ 296inline void\n+297PatchParam::NormalizeTriangle( REAL & u, REAL & v ) const {\n+ 298\n+ 299 if (IsTriangleRotated()) {\n+ 300 REAL fracInv = (REAL)(1.0f / GetParamFraction());\n+ 301\n+ 302 int depthFactor = 1 << GetDepth();\n+ 303 u = (REAL)(depthFactor - GetU()) - (u * fracInv);\n+ 304 v = (REAL)(depthFactor - GetV()) - (v * fracInv);\n+ 305 } else {\n+ 306 Normalize(u, v);\n+ 307 }\n+ 308}\n+ 309\n+ 310template \n+ 311inline void\n+312PatchParam::UnnormalizeTriangle( REAL & u, REAL & v ) const {\n+ 313\n+ 314 if (IsTriangleRotated()) {\n+ 315 REAL frac = GetParamFraction();\n+ 316\n+ 317 int depthFactor = 1 << GetDepth();\n+ 318 u = ((REAL)(depthFactor - GetU()) - u) * frac;\n+ 319 v = ((REAL)(depthFactor - GetV()) - v) * frac;\n+ 320 } else {\n+ 321 Unnormalize(u, v);\n+ 322 }\n+ 323}\n+ 324\n+ 325} // end namespace Far\n+ 326\n+ 327} // end namespace OPENSUBDIV_VERSION\n+ 328using namespace OPENSUBDIV_VERSION;\n+ 329\n+ 330} // end namespace OpenSubdiv\n+ 331\n+ 332#endif /* OPENSUBDIV3_FAR_PATCH_PARAM */\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstPatchDescriptorArray\n-Vtr::ConstArray< PatchDescriptor > ConstPatchDescriptorArray\n-Definition: patchDescriptor.h:146\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SchemeType\n-SchemeType\n-Enumerated type for all subdivision schemes supported by OpenSubdiv.\n-Definition: types.h:37\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor\n-Describes the type of a patch.\n-Definition: patchDescriptor.h:44\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n-Type\n-Definition: patchDescriptor.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::LINES\n-@ LINES\n-lines (useful for cage drawing)\n-Definition: patchDescriptor.h:52\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::QUADS\n-@ QUADS\n-4-sided quadrilateral (bilinear)\n-Definition: patchDescriptor.h:54\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::NON_PATCH\n-@ NON_PATCH\n-undefined\n-Definition: patchDescriptor.h:49\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::LOOP\n-@ LOOP\n-regular triangular patch for the Loop scheme\n-Definition: patchDescriptor.h:57\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_TRIANGLE\n-@ GREGORY_TRIANGLE\n-Definition: patchDescriptor.h:63\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::TRIANGLES\n-@ TRIANGLES\n-3-sided triangle\n-Definition: patchDescriptor.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_BOUNDARY\n-@ GREGORY_BOUNDARY\n-Definition: patchDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_BASIS\n-@ GREGORY_BASIS\n-Definition: patchDescriptor.h:62\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::POINTS\n-@ POINTS\n-points (useful for cage drawing)\n-Definition: patchDescriptor.h:51\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY\n-@ GREGORY\n-Definition: patchDescriptor.h:60\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::REGULAR\n-@ REGULAR\n-regular B-Spline patch for the Catmark scheme\n-Definition: patchDescriptor.h:59\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::IsAdaptive\n-bool IsAdaptive() const\n-Returns true if the type is an adaptive patch.\n-Definition: patchDescriptor.h:97\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetType\n-Type GetType() const\n-Returns the type of the patch.\n-Definition: patchDescriptor.h:87\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetGregoryBasisPatchSize\n-static short GetGregoryBasisPatchSize()\n-Number of control vertices of Gregory patch basis (20)\n-Definition: patchDescriptor.h:126\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetNumControlVertices\n-short GetNumControlVertices() const\n-Returns the number of control vertices expected for a patch of the type\n-described.\n-Definition: patchDescriptor.h:110\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::PatchDescriptor\n-PatchDescriptor(PatchDescriptor const &d)\n-Copy Constructor.\n-Definition: patchDescriptor.h:77\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator==\n-bool operator==(PatchDescriptor const other) const\n-True if the descriptors are identical.\n-Definition: patchDescriptor.h:180\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetRegularPatchSize\n-static short GetRegularPatchSize()\n-Number of control vertices of Regular Patches in table.\n-Definition: patchDescriptor.h:120\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetNumFVarControlVertices\n-short GetNumFVarControlVertices() const\n-Deprecated.\n-Definition: patchDescriptor.h:115\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::PatchDescriptor\n-PatchDescriptor(int type)\n-Constructor.\n-Definition: patchDescriptor.h:73\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator=\n-PatchDescriptor & operator=(PatchDescriptor const &d)\n-Assignment operator.\n-Definition: patchDescriptor.h:81\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::IsAdaptive\n-static bool IsAdaptive(Type type)\n-Returns true if the type is an adaptive (non-linear) patch.\n-Definition: patchDescriptor.h:92\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::operator<\n-bool operator<(PatchDescriptor const other) const\n-Allows ordering of patches by type.\n-Definition: patchDescriptor.h:174\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetGregoryPatchSize\n-static short GetGregoryPatchSize()\n-Number of control vertices of Gregory (and Gregory Boundary) Patches in table.\n-Definition: patchDescriptor.h:123\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::print\n-void print() const\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::\n-GetAdaptivePatchDescriptors\n-static Vtr::ConstArray< PatchDescriptor > GetAdaptivePatchDescriptors(Sdc::\n-SchemeType type)\n-Returns a vector of all the legal patch descriptors for the given adaptive\n-subdivision scheme.\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::PatchDescriptor\n-PatchDescriptor()\n-Default constructor.\n-Definition: patchDescriptor.h:69\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::ConstPatchParamArray\n+Vtr::ConstArray< PatchParam > ConstPatchParamArray\n+Definition: patchParam.h:246\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParamTable\n+std::vector< PatchParam > PatchParamTable\n+Definition: patchParam.h:243\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::Index\n+Vtr::Index Index\n+Definition: types.h:41\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParamArray\n+Vtr::Array< PatchParam > PatchParamArray\n+Definition: patchParam.h:245\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam\n+Patch parameterization.\n+Definition: patchParam.h:152\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::IsTriangleRotated\n+bool IsTriangleRotated() const\n+Returns if a triangular patch is parametrically rotated 180 degrees.\n+Definition: patchParam.h:290\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::IsRegular\n+bool IsRegular() const\n+Returns whether the patch is regular.\n+Definition: patchParam.h:228\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::NormalizeTriangle\n+void NormalizeTriangle(REAL &u, REAL &v) const\n+Definition: patchParam.h:297\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetBoundary\n+unsigned short GetBoundary() const\n+Returns the boundary edge encoding for the patch.\n+Definition: patchParam.h:191\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetTransition\n+unsigned short GetTransition() const\n+Returns the transition edge encoding for the patch.\n+Definition: patchParam.h:188\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetParamFraction\n+float GetParamFraction() const\n+Returns the fraction of unit parametric space covered by this face.\n+Definition: patchParam.h:265\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetU\n+unsigned short GetU() const\n+Returns the log2 value of the u parameter at the first corner of the patch.\n+Definition: patchParam.h:181\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetFaceId\n+Index GetFaceId() const\n+Returns the faceid.\n+Definition: patchParam.h:177\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetDepth\n+unsigned short GetDepth() const\n+Returns the level of subdivision of the patch.\n+Definition: patchParam.h:197\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::NonQuadRoot\n+bool NonQuadRoot() const\n+True if the parent base face is a non-quad.\n+Definition: patchParam.h:194\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Set\n+void Set(Index faceid, short u, short v, unsigned short depth, bool nonquad,\n+unsigned short boundary, unsigned short transition, bool regular=false)\n+Sets the values of the bit fields.\n+Definition: patchParam.h:249\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Clear\n+void Clear()\n+Resets everything to 0.\n+Definition: patchParam.h:174\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::field1\n+unsigned int field1\n+Definition: patchParam.h:231\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Unnormalize\n+void Unnormalize(REAL &u, REAL &v) const\n+A (u,v) pair in a normalized parametric space is mapped back into the fraction\n+of parametric space co...\n+Definition: patchParam.h:281\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::UnnormalizeTriangle\n+void UnnormalizeTriangle(REAL &u, REAL &v) const\n+Definition: patchParam.h:312\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::GetV\n+unsigned short GetV() const\n+Returns the log2 value of the v parameter at the first corner of the patch.\n+Definition: patchParam.h:185\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::field0\n+unsigned int field0\n+Definition: patchParam.h:230\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Normalize\n+void Normalize(REAL &u, REAL &v) const\n+A (u,v) pair in the fraction of parametric space covered by this face is mapped\n+into a normalized par...\n+Definition: patchParam.h:271\n OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstArray\n Definition: array.h:53\n+OpenSubdiv::OPENSUBDIV_VERSION::Vtr::Array\n+Definition: array.h:105\n * opensubdiv\n * far\n- * patchDescriptor.h\n+ * patchParam.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00749.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00749.html", "unified_diff": "@@ -87,15 +87,15 @@\n
\n Namespaces
\n
sparseSelector.h File Reference
\n \n
\n
#include "../version.h"
\n #include "../vtr/types.h"
\n-#include "../vtr/refinement.h"
\n+#include "../vtr/refinement.h"
\n #include <vector>
\n
\n

Go to the source code of this file.

\n \n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00752.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00752.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/vtr/level.h File Reference\n+OpenSubdiv: opensubdiv/vtr/refinement.h File Reference\n \n \n \n \n \n \n \n@@ -82,26 +82,23 @@\n name=\"MSearchResults\" id=\"MSearchResults\">\n \n \n \n
\n
\n Namespaces
\n-
level.h File Reference
\n+
refinement.h File Reference
\n
\n
\n
#include "../version.h"
\n #include "../sdc/types.h"
\n-#include "../sdc/crease.h"
\n #include "../sdc/options.h"
\n #include "../vtr/types.h"
\n-#include <algorithm>
\n+#include "../vtr/level.h"
\n #include <vector>
\n-#include <cassert>
\n-#include <cstring>
\n
\n

Go to the source code of this file.

\n

\n Namespaces

namespace  OpenSubdiv
\n \n \n \n@@ -111,13 +108,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,29 +5,26 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Namespaces\n-level.h File Reference\n+refinement.h File Reference\n #include \"../version.h\"\n #include \"../sdc/types.h\"\n-#include \"../sdc/crease.h\"\n #include \"../sdc/options.h\"\n #include \"../vtr/types.h\"\n-#include \n+#include \"../vtr/level.h\"\n #include \n-#include \n-#include \n Go_to_the_source_code_of_this_file.\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Vtr\n \u00a0\n * opensubdiv\n * vtr\n- * level.h\n+ * refinement.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00752_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00752_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/vtr/level.h Source File\n+OpenSubdiv: opensubdiv/vtr/refinement.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
level.h
\n+
refinement.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2014 DreamWorks Animation LLC.
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -106,882 +106,457 @@\n
17//
\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n-
24#ifndef OPENSUBDIV3_VTR_LEVEL_H
\n-
25#define OPENSUBDIV3_VTR_LEVEL_H
\n+
24#ifndef OPENSUBDIV3_VTR_REFINEMENT_H
\n+
25#define OPENSUBDIV3_VTR_REFINEMENT_H
\n
26
\n
27#include "../version.h"
\n
28
\n
29#include "../sdc/types.h"
\n-
30#include "../sdc/crease.h"
\n-
31#include "../sdc/options.h"
\n-
32#include "../vtr/types.h"
\n+
30#include "../sdc/options.h"
\n+
31#include "../vtr/types.h"
\n+
32#include "../vtr/level.h"
\n
33
\n-
34#include <algorithm>
\n-
35#include <vector>
\n-
36#include <cassert>
\n-
37#include <cstring>
\n-
38
\n-
39
\n-
40namespace OpenSubdiv {
\n-
41namespace OPENSUBDIV_VERSION {
\n-
42
\n-
43namespace Vtr {
\n-
44namespace internal {
\n-
45
\n-
46class Refinement;
\n-
47class TriRefinement;
\n-
48class QuadRefinement;
\n-
49class FVarRefinement;
\n-
50class FVarLevel;
\n-
51
\n-
52//
\n-
53// Level:
\n-
54// A refinement level includes a vectorized representation of the topology
\n-
55// for a particular subdivision level. The topology is "complete" in that any
\n-
56// level can be used as the base level of another subdivision hierarchy and can
\n-
57// be considered a complete mesh independent of its ancestors. It currently
\n-
58// does contain a "depth" member -- as some inferences can then be made about
\n-
59// the topology (i.e. all quads or all tris if not level 0).
\n+
34#include <vector>
\n+
35
\n+
36//
\n+
37// Declaration for the main refinement class (Refinement) and its pre-requisites:
\n+
38//
\n+
39namespace OpenSubdiv {
\n+
40namespace OPENSUBDIV_VERSION {
\n+
41
\n+
42namespace Vtr {
\n+
43namespace internal {
\n+
44
\n+
45class FVarRefinement;
\n+
46
\n+
47//
\n+
48// Refinement:
\n+
49// A refinement is a mapping between two levels -- relating the components in the original
\n+
50// (parent) level to the one refined (child). The refinement may be complete (uniform) or sparse
\n+
51// (adaptive or otherwise selective), so not all components in the parent level will spawn
\n+
52// components in the child level.
\n+
53//
\n+
54// Refinement is an abstract class and expects subclasses corresponding to the different types
\n+
55// of topological splits that the supported subdivision schemes collectively require, i.e. those
\n+
56// listed in Sdc::SplitType. Note the virtual requirements expected of the subclasses in the list
\n+
57// of protected methods -- they differ mainly in the topology that is created in the child Level
\n+
58// and not the propagation of tags through refinement, subdivision of sharpness values or the
\n+
59// treatment of face-varying data. The primary subclasses are QuadRefinement and TriRefinement.
\n
60//
\n-
61// This class is intended for private use within the library. There are still
\n-
62// opportunities to specialize levels -- e.g. those supporting N-sided faces vs
\n-
63// those that are purely quads or tris -- so we prefer to insulate it from public
\n-
64// access.
\n-
65//
\n-
66// The representation of topology here is to store six topological relationships
\n-
67// in tables of integers. Each is stored in its own array(s) so the result is
\n-
68// a SOA representation of the topology. The six relations are:
\n-
69//
\n-
70// - face-verts: vertices incident/comprising a face
\n-
71// - face-edges: edges incident a face
\n-
72// - edge-verts: vertices incident/comprising an edge
\n-
73// - edge-faces: faces incident an edge
\n-
74// - vert-faces: faces incident a vertex
\n-
75// - vert-edges: edges incident a vertex
\n-
76//
\n-
77// There is some redundancy here but the intent is not that this be a minimal
\n-
78// representation, the intent is that it be amenable to refinement. Classes in
\n-
79// the Far layer essentially store 5 of these 6 in a permuted form -- we add
\n-
80// the face-edges here to simplify refinement.
\n-
81//
\n+
61// At a high level, all that is necessary in terms of interface is to construct, initialize
\n+
62// (linking the two levels), optionally select components for sparse refinement (via use of the
\n+
63// SparseSelector) and call the refine() method. This usage is expected of Far::TopologyRefiner.
\n+
64//
\n+
65// Since we really want this class to be restricted from public access eventually, all methods
\n+
66// begin with lower case (as is the convention for protected methods) and the list of friends
\n+
67// will be maintained more strictly.
\n+
68//
\n+
69class Refinement {
\n+
70
\n+
71public:
\n+
72 Refinement(Level const & parent, Level & child, Sdc::Options const& schemeOptions);
\n+
73 virtual ~Refinement();
\n+
74
\n+
75 Level const& parent() const { return *_parent; }
\n+
76 Level const& child() const { return *_child; }
\n+
77 Level& child() { return *_child; }
\n+
78
\n+
79 Sdc::Split getSplitType() const { return _splitType; }
\n+
80 int getRegularFaceSize() const { return _regFaceSize; }
\n+
81 Sdc::Options getOptions() const { return _options; }
\n
82
\n-
83class Level {
\n-
84
\n-
85public:
\n-
86 //
\n-
87 // Simple nested types to hold the tags for each component type -- some of
\n-
88 // which are user-specified features (e.g. whether a face is a hole or not)
\n-
89 // while others indicate the topological nature of the component, how it
\n-
90 // is affected by creasing in its neighborhood, etc.
\n-
91 //
\n-
92 // Most of these properties are passed down to child components during
\n-
93 // refinement, but some -- notably the designation of a component as semi-
\n-
94 // sharp -- require re-determination as sharpness values are reduced at each
\n-
95 // level.
\n-
96 //
\n-
97 struct VTag {
\n-
98 VTag() { }
\n-
99
\n-
100 // When cleared, the VTag ALMOST represents a smooth, regular, interior
\n-
101 // vertex -- the Type enum requires a bit be explicitly set for Smooth,
\n-
102 // so that must be done explicitly if desired on initialization.
\n-
103 void clear() { std::memset((void*) this, 0, sizeof(VTag)); }
\n-
104
\n-
105 typedef unsigned short VTagSize;
\n-
106
\n-
107 VTagSize _nonManifold : 1; // fixed
\n-
108 VTagSize _xordinary : 1; // fixed
\n-
109 VTagSize _boundary : 1; // fixed
\n-
110 VTagSize _corner : 1; // fixed
\n-
111 VTagSize _infSharp : 1; // fixed
\n-
112 VTagSize _semiSharp : 1; // variable
\n-
113 VTagSize _semiSharpEdges : 1; // variable
\n-
114 VTagSize _rule : 4; // variable when _semiSharp
\n-
115
\n-
116 // These next to tags are complementary -- the "incomplete" tag is only
\n-
117 // relevant for refined levels while the "incident an irregular face" tag
\n-
118 // is only relevant for the base level. They could be combined as both
\n-
119 // indicate "no full regular ring" around a vertex
\n-
120 VTagSize _incomplete : 1; // variable only set in refined levels
\n-
121 VTagSize _incidIrregFace : 1; // variable only set in base level
\n+
83 // Face-varying:
\n+
84 int getNumFVarChannels() const { return (int) _fvarChannels.size(); }
\n+
85
\n+
86 FVarRefinement const & getFVarRefinement(int c) const { return *_fvarChannels[c]; }
\n+
87
\n+
88 //
\n+
89 // Options associated with the actual refinement operation, which may end up
\n+
90 // quite involved if we want to allow for the refinement of data that is not
\n+
91 // of interest to be suppressed. For now we have:
\n+
92 //
\n+
93 // "sparse": the alternative to uniform refinement, which requires that
\n+
94 // components be previously selected/marked to be included.
\n+
95 //
\n+
96 // "minimal topology": this is one that may get broken down into a finer
\n+
97 // set of options. It suppresses "full topology" in the child level
\n+
98 // and only generates what is minimally necessary for interpolation --
\n+
99 // which requires at least the face-vertices for faces, but also the
\n+
100 // vertex-faces for any face-varying channels present. So it will
\n+
101 // generate one or two of the six possible topological relations.
\n+
102 //
\n+
103 // These are strictly controlled right now, e.g. for sparse refinement, we
\n+
104 // currently enforce full topology at the finest level to allow for subsequent
\n+
105 // patch construction.
\n+
106 //
\n+
107 struct Options {
\n+
108 Options() : _sparse(false),
\n+
109 _faceVertsFirst(false),
\n+
110 _minimalTopology(false)
\n+
111 { }
\n+
112
\n+
113 unsigned int _sparse : 1;
\n+
114 unsigned int _faceVertsFirst : 1;
\n+
115 unsigned int _minimalTopology : 1;
\n+
116
\n+
117 // Still under consideration:
\n+
118 //unsigned int _childToParentMap : 1;
\n+
119 };
\n+
120
\n+
121 void refine(Options options = Options());
\n
122
\n-
123 // Tags indicating incident infinitely-sharp (permanent) features
\n-
124 VTagSize _infSharpEdges : 1; // fixed
\n-
125 VTagSize _infSharpCrease : 1; // fixed
\n-
126 VTagSize _infIrregular : 1; // fixed
\n-
127
\n-
128 // Alternate constructor and accessor for dealing with integer bits directly:
\n-
129 explicit VTag(VTagSize bits) {
\n-
130 std::memcpy(this, &bits, sizeof(bits));
\n-
131 }
\n-
132 VTagSize getBits() const {
\n-
133 VTagSize bits;
\n-
134 std::memcpy(&bits, this, sizeof(bits));
\n-
135 return bits;
\n-
136 }
\n-
137
\n-
138 static VTag BitwiseOr(VTag const vTags[], int size = 4);
\n-
139 };
\n-
140 struct ETag {
\n-
141 ETag() { }
\n-
142
\n-
143 // When cleared, the ETag represents a smooth, manifold, interior edge
\n-
144 void clear() { std::memset((void*) this, 0, sizeof(ETag)); }
\n-
145
\n-
146 typedef unsigned char ETagSize;
\n+
123 bool hasFaceVerticesFirst() const { return _faceVertsFirst; }
\n+
124
\n+
125public:
\n+
126 //
\n+
127 // Access to members -- some testing classes (involving vertex interpolation)
\n+
128 // currently make use of these:
\n+
129 //
\n+
130 int getNumChildFacesFromFaces() const { return _childFaceFromFaceCount; }
\n+
131 int getNumChildEdgesFromFaces() const { return _childEdgeFromFaceCount; }
\n+
132 int getNumChildEdgesFromEdges() const { return _childEdgeFromEdgeCount; }
\n+
133 int getNumChildVerticesFromFaces() const { return _childVertFromFaceCount; }
\n+
134 int getNumChildVerticesFromEdges() const { return _childVertFromEdgeCount; }
\n+
135 int getNumChildVerticesFromVertices() const { return _childVertFromVertCount; }
\n+
136
\n+
137 Index getFirstChildFaceFromFaces() const { return _firstChildFaceFromFace; }
\n+
138 Index getFirstChildEdgeFromFaces() const { return _firstChildEdgeFromFace; }
\n+
139 Index getFirstChildEdgeFromEdges() const { return _firstChildEdgeFromEdge; }
\n+
140 Index getFirstChildVertexFromFaces() const { return _firstChildVertFromFace; }
\n+
141 Index getFirstChildVertexFromEdges() const { return _firstChildVertFromEdge; }
\n+
142 Index getFirstChildVertexFromVertices() const { return _firstChildVertFromVert; }
\n+
143
\n+
144 Index getFaceChildVertex(Index f) const { return _faceChildVertIndex[f]; }
\n+
145 Index getEdgeChildVertex(Index e) const { return _edgeChildVertIndex[e]; }
\n+
146 Index getVertexChildVertex(Index v) const { return _vertChildVertIndex[v]; }
\n
147
\n-
148 ETagSize _nonManifold : 1; // fixed
\n-
149 ETagSize _boundary : 1; // fixed
\n-
150 ETagSize _infSharp : 1; // fixed
\n-
151 ETagSize _semiSharp : 1; // variable
\n-
152
\n-
153 // Alternate constructor and accessor for dealing with integer bits directly:
\n-
154 explicit ETag(ETagSize bits) {
\n-
155 std::memcpy(this, &bits, sizeof(bits));
\n-
156 }
\n-
157 ETagSize getBits() const {
\n-
158 ETagSize bits;
\n-
159 std::memcpy(&bits, this, sizeof(bits));
\n-
160 return bits;
\n-
161 }
\n-
162
\n-
163 static ETag BitwiseOr(ETag const eTags[], int size = 4);
\n-
164 };
\n-
165 struct FTag {
\n-
166 FTag() { }
\n-
167
\n-
168 void clear() { std::memset((void*) this, 0, sizeof(FTag)); }
\n-
169
\n-
170 typedef unsigned char FTagSize;
\n-
171
\n-
172 FTagSize _hole : 1; // fixed
\n-
173
\n-
174 // On deck -- coming soon...
\n-
175 //FTagSize _hasEdits : 1; // variable
\n-
176 };
\n-
177
\n-
178 // Additional simple struct to identify a "span" around a vertex, i.e. a
\n-
179 // subset of the faces around a vertex delimited by some property (e.g. a
\n-
180 // face-varying discontinuity, an inf-sharp edge, etc.)
\n-
181 //
\n-
182 // The span requires an "origin" and a "size" to fully define its extent.
\n-
183 // Use of the size is required over a leading/trailing pair as the valence
\n-
184 // around a non-manifold vertex cannot be trivially determined from two
\n-
185 // extremeties. Similarly a start face is chosen over an edge as starting
\n-
186 // with a manifold edge is ambiguous. Additional tags also support
\n-
187 // non-manifold cases, e.g. periodic spans at the apex of a double cone.
\n+
148 ConstIndexArray getFaceChildFaces(Index parentFace) const;
\n+
149 ConstIndexArray getFaceChildEdges(Index parentFace) const;
\n+
150 ConstIndexArray getEdgeChildEdges(Index parentEdge) const;
\n+
151
\n+
152 // Child-to-parent relationships
\n+
153 bool isChildVertexComplete(Index v) const { return ! _childVertexTag[v]._incomplete; }
\n+
154
\n+
155 Index getChildFaceParentFace(Index f) const { return _childFaceParentIndex[f]; }
\n+
156 int getChildFaceInParentFace(Index f) const { return _childFaceTag[f]._indexInParent; }
\n+
157
\n+
158 Index getChildEdgeParentIndex(Index e) const { return _childEdgeParentIndex[e]; }
\n+
159
\n+
160 Index getChildVertexParentIndex(Index v) const { return _childVertexParentIndex[v]; }
\n+
161
\n+
162//
\n+
163// Modifiers intended for internal/protected use:
\n+
164//
\n+
165public:
\n+
166
\n+
167 IndexArray getFaceChildFaces(Index parentFace);
\n+
168 IndexArray getFaceChildEdges(Index parentFace);
\n+
169 IndexArray getEdgeChildEdges(Index parentEdge);
\n+
170
\n+
171public:
\n+
172 //
\n+
173 // Tags have now been added per-component in Level, but there is additional need to tag
\n+
174 // components within Refinement -- we can't tag the parent level components for any
\n+
175 // refinement (in order to keep it const) and tags associated with children that are
\n+
176 // specific to the child-to-parent mapping may not be warranted in the child level.
\n+
177 //
\n+
178 // Parent tags are only required for sparse refinement. The main property to tag is
\n+
179 // whether a component was selected, and so a single SparseTag is used for all three
\n+
180 // component types. Tagging if a component is "transitional" is also useful. This may
\n+
181 // only be necessary for edges but is currently packed into a mask per-edge for faces,
\n+
182 // which could be deferred, in which case "transitional" could be a single bit.
\n+
183 //
\n+
184 // Child tags are part of the child-to-parent mapping, which consists of the parent
\n+
185 // component index for each child component, plus a tag for the child indicating more
\n+
186 // about its relationship to its parent, e.g. is it completely defined, what the parent
\n+
187 // component type is, what is the index of the child within its parent, etc.
\n
188 //
\n-
189 // Currently setting the size to 0 or leaving the span "unassigned" is an
\n-
190 // indication to use the full neighborhood rather than a subset -- prefer
\n-
191 // use of the const method here to direct inspection of the member.
\n-
192 //
\n-
193 struct VSpan {
\n-
194 VSpan() { std::memset((void*) this, 0, sizeof(VSpan)); }
\n+
189 struct SparseTag {
\n+
190 SparseTag() : _selected(0), _transitional(0) { }
\n+
191
\n+
192 unsigned char _selected : 1; // component specifically selected for refinement
\n+
193 unsigned char _transitional : 4; // adjacent to a refined component (4-bits for face)
\n+
194 };
\n
195
\n-
196 void clear() { std::memset((void*) this, 0, sizeof(VSpan)); }
\n-
197 bool isAssigned() const { return _numFaces > 0; }
\n+
196 struct ChildTag {
\n+
197 ChildTag() { }
\n
198
\n-
199 LocalIndex _numFaces;
\n-
200 LocalIndex _startFace;
\n-
201 LocalIndex _cornerInSpan;
\n-
202
\n-
203 unsigned short _periodic : 1;
\n-
204 unsigned short _sharp : 1;
\n-
205 };
\n-
206
\n-
207public:
\n-
208 Level();
\n-
209 ~Level();
\n-
210
\n-
211 // Simple accessors:
\n-
212 int getDepth() const { return _depth; }
\n-
213
\n-
214 int getNumVertices() const { return _vertCount; }
\n-
215 int getNumFaces() const { return _faceCount; }
\n-
216 int getNumEdges() const { return _edgeCount; }
\n-
217
\n-
218 // More global sizes may prove useful...
\n-
219 int getNumFaceVerticesTotal() const { return (int) _faceVertIndices.size(); }
\n-
220 int getNumFaceEdgesTotal() const { return (int) _faceEdgeIndices.size(); }
\n-
221 int getNumEdgeVerticesTotal() const { return (int) _edgeVertIndices.size(); }
\n-
222 int getNumEdgeFacesTotal() const { return (int) _edgeFaceIndices.size(); }
\n-
223 int getNumVertexFacesTotal() const { return (int) _vertFaceIndices.size(); }
\n-
224 int getNumVertexEdgesTotal() const { return (int) _vertEdgeIndices.size(); }
\n-
225
\n-
226 int getMaxValence() const { return _maxValence; }
\n-
227 int getMaxEdgeFaces() const { return _maxEdgeFaces; }
\n-
228
\n-
229 // Methods to access the relation tables/indices -- note that for some relations
\n-
230 // (i.e. those where a component is "contained by" a neighbor, or more generally
\n-
231 // when the neighbor is a simplex of higher dimension) we store an additional
\n-
232 // "local index", e.g. for the case of vert-faces if one of the faces F[i] is
\n-
233 // incident a vertex V, then L[i] is the "local index" in F[i] of vertex V.
\n-
234 // Once have only quads (or tris), this local index need only occupy two bits
\n-
235 // and could conceivably be packed into the same integer as the face index, but
\n-
236 // for now, given the need to support faces of potentially high valence we'll
\n-
237 // use an 8- or 16-bit integer.
\n-
238 //
\n-
239 // Methods to access the six topological relations:
\n-
240 ConstIndexArray getFaceVertices(Index faceIndex) const;
\n-
241 ConstIndexArray getFaceEdges(Index faceIndex) const;
\n-
242 ConstIndexArray getEdgeVertices(Index edgeIndex) const;
\n-
243 ConstIndexArray getEdgeFaces(Index edgeIndex) const;
\n-
244 ConstIndexArray getVertexFaces(Index vertIndex) const;
\n-
245 ConstIndexArray getVertexEdges(Index vertIndex) const;
\n-
246
\n-
247 ConstLocalIndexArray getEdgeFaceLocalIndices(Index edgeIndex) const;
\n-
248 ConstLocalIndexArray getVertexFaceLocalIndices(Index vertIndex) const;
\n-
249 ConstLocalIndexArray getVertexEdgeLocalIndices(Index vertIndex) const;
\n+
199 unsigned char _incomplete : 1; // incomplete neighborhood to represent limit of parent
\n+
200 unsigned char _parentType : 2; // type of parent component: vertex, edge or face
\n+
201 unsigned char _indexInParent : 2; // index of child wrt parent: 0-3, or iterative if N > 4
\n+
202 };
\n+
203
\n+
204 // Methods to access and modify tags:
\n+
205 SparseTag const & getParentFaceSparseTag( Index f) const { return _parentFaceTag[f]; }
\n+
206 SparseTag const & getParentEdgeSparseTag( Index e) const { return _parentEdgeTag[e]; }
\n+
207 SparseTag const & getParentVertexSparseTag(Index v) const { return _parentVertexTag[v]; }
\n+
208
\n+
209 SparseTag & getParentFaceSparseTag( Index f) { return _parentFaceTag[f]; }
\n+
210 SparseTag & getParentEdgeSparseTag( Index e) { return _parentEdgeTag[e]; }
\n+
211 SparseTag & getParentVertexSparseTag(Index v) { return _parentVertexTag[v]; }
\n+
212
\n+
213 ChildTag const & getChildFaceTag( Index f) const { return _childFaceTag[f]; }
\n+
214 ChildTag const & getChildEdgeTag( Index e) const { return _childEdgeTag[e]; }
\n+
215 ChildTag const & getChildVertexTag(Index v) const { return _childVertexTag[v]; }
\n+
216
\n+
217 ChildTag & getChildFaceTag( Index f) { return _childFaceTag[f]; }
\n+
218 ChildTag & getChildEdgeTag( Index e) { return _childEdgeTag[e]; }
\n+
219 ChildTag & getChildVertexTag(Index v) { return _childVertexTag[v]; }
\n+
220
\n+
221// Remaining methods should really be protected -- for use by subclasses...
\n+
222public:
\n+
223 //
\n+
224 // Methods involved in constructing the parent-to-child mapping -- when the
\n+
225 // refinement is sparse, additional methods are needed to identify the selection:
\n+
226 //
\n+
227 void populateParentToChildMapping();
\n+
228 void populateParentChildIndices();
\n+
229 void printParentToChildMapping() const;
\n+
230
\n+
231 virtual void allocateParentChildIndices() = 0;
\n+
232
\n+
233 // Supporting method for sparse refinement:
\n+
234 void initializeSparseSelectionTags();
\n+
235 void markSparseChildComponentIndices();
\n+
236 void markSparseVertexChildren();
\n+
237 void markSparseEdgeChildren();
\n+
238
\n+
239 virtual void markSparseFaceChildren() = 0;
\n+
240
\n+
241 void initializeChildComponentCounts();
\n+
242
\n+
243 //
\n+
244 // Methods involved in constructing the child-to-parent mapping:
\n+
245 //
\n+
246 void populateChildToParentMapping();
\n+
247
\n+
248 void populateFaceParentVectors(ChildTag const initialChildTags[2][4]);
\n+
249 void populateFaceParentFromParentFaces(ChildTag const initialChildTags[2][4]);
\n
250
\n-
251 // Replace these with access to sharpness buffers/arrays rather than elements:
\n-
252 float getEdgeSharpness(Index edgeIndex) const;
\n-
253 float getVertexSharpness(Index vertIndex) const;
\n-
254 Sdc::Crease::Rule getVertexRule(Index vertIndex) const;
\n-
255
\n-
256 Index findEdge(Index v0Index, Index v1Index) const;
\n-
257
\n-
258 // Holes
\n-
259 void setFaceHole(Index faceIndex, bool b);
\n-
260 bool isFaceHole(Index faceIndex) const;
\n-
261
\n-
262 // Face-varying
\n-
263 Sdc::Options getFVarOptions(int channel) const;
\n-
264 int getNumFVarChannels() const { return (int) _fvarChannels.size(); }
\n-
265 int getNumFVarValues(int channel) const;
\n-
266 ConstIndexArray getFaceFVarValues(Index faceIndex, int channel) const;
\n+
251 void populateEdgeParentVectors(ChildTag const initialChildTags[2][4]);
\n+
252 void populateEdgeParentFromParentFaces(ChildTag const initialChildTags[2][4]);
\n+
253 void populateEdgeParentFromParentEdges(ChildTag const initialChildTags[2][4]);
\n+
254
\n+
255 void populateVertexParentVectors(ChildTag const initialChildTags[2][4]);
\n+
256 void populateVertexParentFromParentFaces(ChildTag const initialChildTags[2][4]);
\n+
257 void populateVertexParentFromParentEdges(ChildTag const initialChildTags[2][4]);
\n+
258 void populateVertexParentFromParentVertices(ChildTag const initialChildTags[2][4]);
\n+
259
\n+
260 //
\n+
261 // Methods involved in propagating component tags from parent to child:
\n+
262 //
\n+
263 void propagateComponentTags();
\n+
264
\n+
265 void populateFaceTagVectors();
\n+
266 void populateFaceTagsFromParentFaces();
\n
267
\n-
268 FVarLevel & getFVarLevel(int channel) { return *_fvarChannels[channel]; }
\n-
269 FVarLevel const & getFVarLevel(int channel) const { return *_fvarChannels[channel]; }
\n-
270
\n-
271 // Manifold/non-manifold tags:
\n-
272 void setEdgeNonManifold(Index edgeIndex, bool b);
\n-
273 bool isEdgeNonManifold(Index edgeIndex) const;
\n-
274
\n-
275 void setVertexNonManifold(Index vertIndex, bool b);
\n-
276 bool isVertexNonManifold(Index vertIndex) const;
\n-
277
\n-
278 // General access to all component tags:
\n-
279 VTag const & getVertexTag(Index vertIndex) const { return _vertTags[vertIndex]; }
\n-
280 ETag const & getEdgeTag(Index edgeIndex) const { return _edgeTags[edgeIndex]; }
\n-
281 FTag const & getFaceTag(Index faceIndex) const { return _faceTags[faceIndex]; }
\n-
282
\n-
283 VTag & getVertexTag(Index vertIndex) { return _vertTags[vertIndex]; }
\n-
284 ETag & getEdgeTag(Index edgeIndex) { return _edgeTags[edgeIndex]; }
\n-
285 FTag & getFaceTag(Index faceIndex) { return _faceTags[faceIndex]; }
\n-
286
\n-
287public:
\n+
268 void populateEdgeTagVectors();
\n+
269 void populateEdgeTagsFromParentFaces();
\n+
270 void populateEdgeTagsFromParentEdges();
\n+
271
\n+
272 void populateVertexTagVectors();
\n+
273 void populateVertexTagsFromParentFaces();
\n+
274 void populateVertexTagsFromParentEdges();
\n+
275 void populateVertexTagsFromParentVertices();
\n+
276
\n+
277 //
\n+
278 // Methods (and types) involved in subdividing the topology -- though not
\n+
279 // fully exploited, any subset of the 6 relations can be generated:
\n+
280 //
\n+
281 struct Relations {
\n+
282 unsigned int _faceVertices : 1;
\n+
283 unsigned int _faceEdges : 1;
\n+
284 unsigned int _edgeVertices : 1;
\n+
285 unsigned int _edgeFaces : 1;
\n+
286 unsigned int _vertexFaces : 1;
\n+
287 unsigned int _vertexEdges : 1;
\n
288
\n-
289 // Debugging aides:
\n-
290 enum TopologyError {
\n-
291 TOPOLOGY_MISSING_EDGE_FACES=0,
\n-
292 TOPOLOGY_MISSING_EDGE_VERTS,
\n-
293 TOPOLOGY_MISSING_FACE_EDGES,
\n-
294 TOPOLOGY_MISSING_FACE_VERTS,
\n-
295 TOPOLOGY_MISSING_VERT_FACES,
\n-
296 TOPOLOGY_MISSING_VERT_EDGES,
\n-
297
\n-
298 TOPOLOGY_FAILED_CORRELATION_EDGE_FACE,
\n-
299 TOPOLOGY_FAILED_CORRELATION_FACE_VERT,
\n-
300 TOPOLOGY_FAILED_CORRELATION_FACE_EDGE,
\n-
301
\n-
302 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_EDGE,
\n-
303 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_FACE,
\n-
304 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_FACES_EDGES,
\n-
305
\n-
306 TOPOLOGY_DEGENERATE_EDGE,
\n-
307 TOPOLOGY_NON_MANIFOLD_EDGE,
\n-
308
\n-
309 TOPOLOGY_INVALID_CREASE_EDGE,
\n-
310 TOPOLOGY_INVALID_CREASE_VERT
\n-
311 };
\n+
289 void setAll(bool enable) {
\n+
290 _faceVertices = enable;
\n+
291 _faceEdges = enable;
\n+
292 _edgeVertices = enable;
\n+
293 _edgeFaces = enable;
\n+
294 _vertexFaces = enable;
\n+
295 _vertexEdges = enable;
\n+
296 }
\n+
297 };
\n+
298
\n+
299 void subdivideTopology(Relations const& relationsToSubdivide);
\n+
300
\n+
301 virtual void populateFaceVertexRelation() = 0;
\n+
302 virtual void populateFaceEdgeRelation() = 0;
\n+
303 virtual void populateEdgeVertexRelation() = 0;
\n+
304 virtual void populateEdgeFaceRelation() = 0;
\n+
305 virtual void populateVertexFaceRelation() = 0;
\n+
306 virtual void populateVertexEdgeRelation() = 0;
\n+
307
\n+
308 //
\n+
309 // Methods involved in subdividing and inspecting sharpness values:
\n+
310 //
\n+
311 void subdivideSharpnessValues();
\n
312
\n-
313 static char const * getTopologyErrorString(TopologyError errCode);
\n-
314
\n-
315 typedef void (* ValidationCallback)(TopologyError errCode, char const * msg, void const * clientData);
\n+
313 void subdivideVertexSharpness();
\n+
314 void subdivideEdgeSharpness();
\n+
315 void reclassifySemisharpVertices();
\n
316
\n-
317 bool validateTopology(ValidationCallback callback=0, void const * clientData=0) const;
\n-
318
\n-
319 void print(const Refinement* parentRefinement = 0) const;
\n-
320
\n-
321public:
\n-
322 // High-level topology queries -- these may be moved elsewhere:
\n-
323
\n-
324 bool isSingleCreasePatch(Index face, float* sharpnessOut=NULL, int* rotationOut=NULL) const;
\n+
317 //
\n+
318 // Methods involved in subdividing face-varying topology:
\n+
319 //
\n+
320 void subdivideFVarChannels();
\n+
321
\n+
322protected:
\n+
323 // A debug method of Level prints a Refinement (should really change this)
\n+
324 friend void Level::print(const Refinement *) const;
\n
325
\n
326 //
\n-
327 // When inspecting topology, the component tags -- particularly VTag and ETag -- are most
\n-
328 // often inspected in groups for the face to which they belong. They are designed to be
\n-
329 // bitwise OR'd (the result then referred to as a "composite" tag) to make quick decisions
\n-
330 // about the face as a whole to avoid tedious topological inspection.
\n-
331 //
\n-
332 // The same logic can be applied to topology in a FVar channel when tags specific to that
\n-
333 // channel are used. Note that the VTags apply to the FVar values assigned to the corners
\n-
334 // of the face and not the vertex as a whole. The "composite" face-varying VTag for a
\n-
335 // vertex is the union of VTags of all distinct FVar values for that vertex.
\n-
336 //
\n-
337 bool doesVertexFVarTopologyMatch(Index vIndex, int fvarChannel) const;
\n-
338 bool doesFaceFVarTopologyMatch( Index fIndex, int fvarChannel) const;
\n-
339 bool doesEdgeFVarTopologyMatch( Index eIndex, int fvarChannel) const;
\n-
340
\n-
341 void getFaceVTags(Index fIndex, VTag vTags[], int fvarChannel = -1) const;
\n-
342 void getFaceETags(Index fIndex, ETag eTags[], int fvarChannel = -1) const;
\n+
327 // Data members -- the logical grouping of some of these (and methods that make use
\n+
328 // of them) may lead to grouping them into a few utility classes or structs...
\n+
329 //
\n+
330
\n+
331 // Defined on construction:
\n+
332 Level const * _parent;
\n+
333 Level * _child;
\n+
334 Sdc::Options _options;
\n+
335
\n+
336 // Defined by the subclass:
\n+
337 Sdc::Split _splitType;
\n+
338 int _regFaceSize;
\n+
339
\n+
340 // Determined by the refinement options:
\n+
341 bool _uniform;
\n+
342 bool _faceVertsFirst;
\n
343
\n-
344 VTag getFaceCompositeVTag(Index fIndex, int fvarChannel = -1) const;
\n-
345 VTag getFaceCompositeVTag(ConstIndexArray & fVerts) const;
\n-
346
\n-
347 VTag getVertexCompositeFVarVTag(Index vIndex, int fvarChannel) const;
\n-
348
\n-
349 //
\n-
350 // When gathering "patch points" we may want the indices of the vertices or the corresponding
\n-
351 // FVar values for a particular channel. Both are represented and equally accessible within
\n-
352 // the faces, so we allow all to be returned through these methods. Setting the optional FVar
\n-
353 // channel to -1 will retrieve indices of vertices instead of FVar values:
\n-
354 //
\n-
355 int gatherQuadLinearPatchPoints(Index fIndex, Index patchPoints[], int rotation = 0,
\n-
356 int fvarChannel = -1) const;
\n-
357
\n-
358 int gatherQuadRegularInteriorPatchPoints(Index fIndex, Index patchPoints[], int rotation = 0,
\n-
359 int fvarChannel = -1) const;
\n-
360 int gatherQuadRegularBoundaryPatchPoints(Index fIndex, Index patchPoints[], int boundaryEdgeInFace,
\n-
361 int fvarChannel = -1) const;
\n-
362 int gatherQuadRegularCornerPatchPoints( Index fIndex, Index patchPoints[], int cornerVertInFace,
\n-
363 int fvarChannel = -1) const;
\n-
364
\n-
365 int gatherQuadRegularRingAroundVertex(Index vIndex, Index ringPoints[],
\n-
366 int fvarChannel = -1) const;
\n-
367 int gatherQuadRegularPartialRingAroundVertex(Index vIndex, VSpan const & span, Index ringPoints[],
\n-
368 int fvarChannel = -1) const;
\n-
369
\n-
370 // WIP -- for future use, need to extend for face-varying...
\n-
371 int gatherTriRegularInteriorPatchPoints( Index fIndex, Index patchVerts[], int rotation = 0) const;
\n-
372 int gatherTriRegularBoundaryVertexPatchPoints(Index fIndex, Index patchVerts[], int boundaryVertInFace) const;
\n-
373 int gatherTriRegularBoundaryEdgePatchPoints( Index fIndex, Index patchVerts[], int boundaryEdgeInFace) const;
\n-
374 int gatherTriRegularCornerVertexPatchPoints( Index fIndex, Index patchVerts[], int cornerVertInFace) const;
\n-
375 int gatherTriRegularCornerEdgePatchPoints( Index fIndex, Index patchVerts[], int cornerEdgeInFace) const;
\n+
344 //
\n+
345 // Inventory and ordering of the types of child components:
\n+
346 //
\n+
347 int _childFaceFromFaceCount; // arguably redundant (all faces originate from faces)
\n+
348 int _childEdgeFromFaceCount;
\n+
349 int _childEdgeFromEdgeCount;
\n+
350 int _childVertFromFaceCount;
\n+
351 int _childVertFromEdgeCount;
\n+
352 int _childVertFromVertCount;
\n+
353
\n+
354 int _firstChildFaceFromFace; // arguably redundant (all faces originate from faces)
\n+
355 int _firstChildEdgeFromFace;
\n+
356 int _firstChildEdgeFromEdge;
\n+
357 int _firstChildVertFromFace;
\n+
358 int _firstChildVertFromEdge;
\n+
359 int _firstChildVertFromVert;
\n+
360
\n+
361 //
\n+
362 // The parent-to-child mapping:
\n+
363 // These are vectors sized according to the number of parent components (and
\n+
364 // their topology) that contain references/indices to the child components that
\n+
365 // result from them by refinement. When refinement is sparse, parent components
\n+
366 // that have not spawned all child components will have their missing children
\n+
367 // marked as invalid.
\n+
368 //
\n+
369 // NOTE the "Array" members here. Often vectors within the Level can be shared
\n+
370 // with the Refinement, and an Array instance is used to do so. If not shared
\n+
371 // the subclass just initializes the Array members after allocating its own local
\n+
372 // vector members.
\n+
373 //
\n+
374 IndexArray _faceChildFaceCountsAndOffsets;
\n+
375 IndexArray _faceChildEdgeCountsAndOffsets;
\n
376
\n-
377public:
\n-
378 // Sizing methods used to construct a level to populate:
\n-
379 void resizeFaces( int numFaces);
\n-
380 void resizeFaceVertices(int numFaceVertsTotal);
\n-
381 void resizeFaceEdges( int numFaceEdgesTotal);
\n-
382
\n-
383 void resizeEdges( int numEdges);
\n-
384 void resizeEdgeVertices(); // always 2*edgeCount
\n-
385 void resizeEdgeFaces(int numEdgeFacesTotal);
\n-
386
\n-
387 void resizeVertices( int numVertices);
\n-
388 void resizeVertexFaces(int numVertexFacesTotal);
\n-
389 void resizeVertexEdges(int numVertexEdgesTotal);
\n-
390
\n-
391 void setMaxValence(int maxValence);
\n+
377 IndexVector _faceChildFaceIndices; // *cannot* always use face-vert counts/offsets
\n+
378 IndexVector _faceChildEdgeIndices; // can use face-vert counts/offsets
\n+
379 IndexVector _faceChildVertIndex;
\n+
380
\n+
381 IndexVector _edgeChildEdgeIndices; // trivial/corresponding pair for each
\n+
382 IndexVector _edgeChildVertIndex;
\n+
383
\n+
384 IndexVector _vertChildVertIndex;
\n+
385
\n+
386 //
\n+
387 // The child-to-parent mapping:
\n+
388 //
\n+
389 IndexVector _childFaceParentIndex;
\n+
390 IndexVector _childEdgeParentIndex;
\n+
391 IndexVector _childVertexParentIndex;
\n
392
\n-
393 // Modifiers to populate the relations for each component:
\n-
394 IndexArray getFaceVertices(Index faceIndex);
\n-
395 IndexArray getFaceEdges(Index faceIndex);
\n-
396 IndexArray getEdgeVertices(Index edgeIndex);
\n-
397 IndexArray getEdgeFaces(Index edgeIndex);
\n-
398 IndexArray getVertexFaces(Index vertIndex);
\n-
399 IndexArray getVertexEdges(Index vertIndex);
\n-
400
\n-
401 LocalIndexArray getEdgeFaceLocalIndices(Index edgeIndex);
\n-
402 LocalIndexArray getVertexFaceLocalIndices(Index vertIndex);
\n-
403 LocalIndexArray getVertexEdgeLocalIndices(Index vertIndex);
\n-
404
\n-
405 // Replace these with access to sharpness buffers/arrays rather than elements:
\n-
406 float& getEdgeSharpness(Index edgeIndex);
\n-
407 float& getVertexSharpness(Index vertIndex);
\n-
408
\n-
409 // Create, destroy and populate face-varying channels:
\n-
410 int createFVarChannel(int fvarValueCount, Sdc::Options const& options);
\n-
411 void destroyFVarChannel(int channel);
\n+
393 std::vector<ChildTag> _childFaceTag;
\n+
394 std::vector<ChildTag> _childEdgeTag;
\n+
395 std::vector<ChildTag> _childVertexTag;
\n+
396
\n+
397 //
\n+
398 // Tags for sparse selection of components:
\n+
399 //
\n+
400 std::vector<SparseTag> _parentFaceTag;
\n+
401 std::vector<SparseTag> _parentEdgeTag;
\n+
402 std::vector<SparseTag> _parentVertexTag;
\n+
403
\n+
404 //
\n+
405 // Refinement data for face-varying channels present in the Levels being refined:
\n+
406 //
\n+
407 std::vector<FVarRefinement*> _fvarChannels;
\n+
408};
\n+
409
\n+
410inline ConstIndexArray
\n+
411Refinement::getFaceChildFaces(Index parentFace) const {
\n
412
\n-
413 IndexArray getFaceFVarValues(Index faceIndex, int channel);
\n-
414
\n-
415 void completeFVarChannelTopology(int channel, int regBoundaryValence);
\n+
413 return ConstIndexArray(&_faceChildFaceIndices[_faceChildFaceCountsAndOffsets[2*parentFace+1]],
\n+
414 _faceChildFaceCountsAndOffsets[2*parentFace]);
\n+
415}
\n
416
\n-
417 // Counts and offsets for all relation types:
\n-
418 // - these may be unwarranted if we let Refinement access members directly...
\n-
419 int getNumFaceVertices( Index faceIndex) const { return _faceVertCountsAndOffsets[2*faceIndex]; }
\n-
420 int getOffsetOfFaceVertices(Index faceIndex) const { return _faceVertCountsAndOffsets[2*faceIndex + 1]; }
\n-
421
\n-
422 int getNumFaceEdges( Index faceIndex) const { return getNumFaceVertices(faceIndex); }
\n-
423 int getOffsetOfFaceEdges(Index faceIndex) const { return getOffsetOfFaceVertices(faceIndex); }
\n-
424
\n-
425 int getNumEdgeVertices( Index ) const { return 2; }
\n-
426 int getOffsetOfEdgeVertices(Index edgeIndex) const { return 2 * edgeIndex; }
\n-
427
\n-
428 int getNumEdgeFaces( Index edgeIndex) const { return _edgeFaceCountsAndOffsets[2*edgeIndex]; }
\n-
429 int getOffsetOfEdgeFaces(Index edgeIndex) const { return _edgeFaceCountsAndOffsets[2*edgeIndex + 1]; }
\n-
430
\n-
431 int getNumVertexFaces( Index vertIndex) const { return _vertFaceCountsAndOffsets[2*vertIndex]; }
\n-
432 int getOffsetOfVertexFaces(Index vertIndex) const { return _vertFaceCountsAndOffsets[2*vertIndex + 1]; }
\n-
433
\n-
434 int getNumVertexEdges( Index vertIndex) const { return _vertEdgeCountsAndOffsets[2*vertIndex]; }
\n-
435 int getOffsetOfVertexEdges(Index vertIndex) const { return _vertEdgeCountsAndOffsets[2*vertIndex + 1]; }
\n+
417inline IndexArray
\n+
418Refinement::getFaceChildFaces(Index parentFace) {
\n+
419
\n+
420 return IndexArray(&_faceChildFaceIndices[_faceChildFaceCountsAndOffsets[2*parentFace+1]],
\n+
421 _faceChildFaceCountsAndOffsets[2*parentFace]);
\n+
422}
\n+
423
\n+
424inline ConstIndexArray
\n+
425Refinement::getFaceChildEdges(Index parentFace) const {
\n+
426
\n+
427 return ConstIndexArray(&_faceChildEdgeIndices[_faceChildEdgeCountsAndOffsets[2*parentFace+1]],
\n+
428 _faceChildEdgeCountsAndOffsets[2*parentFace]);
\n+
429}
\n+
430inline IndexArray
\n+
431Refinement::getFaceChildEdges(Index parentFace) {
\n+
432
\n+
433 return IndexArray(&_faceChildEdgeIndices[_faceChildEdgeCountsAndOffsets[2*parentFace+1]],
\n+
434 _faceChildEdgeCountsAndOffsets[2*parentFace]);
\n+
435}
\n
436
\n-
437 ConstIndexArray getFaceVertices() const;
\n-
438
\n-
439 //
\n-
440 // Note that for some relations, the size of the relations for a child component
\n-
441 // can vary radically from its parent due to the sparsity of the refinement. So
\n-
442 // in these cases a few additional utilities are provided to help define the set
\n-
443 // of incident components. Assuming adequate memory has been allocated, the
\n-
444 // "resize" methods here initialize the set of incident components by setting
\n-
445 // both the size and the appropriate offset, while "trim" is use to quickly lower
\n-
446 // the size from an upper bound and nothing else.
\n-
447 //
\n-
448 void resizeFaceVertices(Index FaceIndex, int count);
\n-
449
\n-
450 void resizeEdgeFaces(Index edgeIndex, int count);
\n-
451 void trimEdgeFaces( Index edgeIndex, int count);
\n-
452
\n-
453 void resizeVertexFaces(Index vertIndex, int count);
\n-
454 void trimVertexFaces( Index vertIndex, int count);
\n+
437inline ConstIndexArray
\n+
438Refinement::getEdgeChildEdges(Index parentEdge) const {
\n+
439
\n+
440 return ConstIndexArray(&_edgeChildEdgeIndices[parentEdge*2], 2);
\n+
441}
\n+
442
\n+
443inline IndexArray
\n+
444Refinement::getEdgeChildEdges(Index parentEdge) {
\n+
445
\n+
446 return IndexArray(&_edgeChildEdgeIndices[parentEdge*2], 2);
\n+
447}
\n+
448
\n+
449} // end namespace internal
\n+
450} // end namespace Vtr
\n+
451
\n+
452} // end namespace OPENSUBDIV_VERSION
\n+
453using namespace OPENSUBDIV_VERSION;
\n+
454} // end namespace OpenSubdiv
\n
455
\n-
456 void resizeVertexEdges(Index vertIndex, int count);
\n-
457 void trimVertexEdges( Index vertIndex, int count);
\n-
458
\n-
459public:
\n-
460 //
\n-
461 // Initial plans were to have a few specific classes properly construct the
\n-
462 // topology from scratch, e.g. the Refinement class and a Factory class for
\n-
463 // the base level, by populating all topological relations. The need to have
\n-
464 // a class construct full topology given only a simple face-vertex list, made
\n-
465 // it necessary to write code to define and orient all relations -- and most
\n-
466 // of that seemed best placed here.
\n-
467 //
\n-
468 bool completeTopologyFromFaceVertices();
\n-
469 Index findEdge(Index v0, Index v1, ConstIndexArray v0Edges) const;
\n-
470
\n-
471 // Methods supporting the above:
\n-
472 void orientIncidentComponents();
\n-
473 bool orderVertexFacesAndEdges(Index vIndex, Index* vFaces, Index* vEdges) const;
\n-
474 bool orderVertexFacesAndEdges(Index vIndex);
\n-
475 void populateLocalIndices();
\n-
476
\n-
477 IndexArray shareFaceVertCountsAndOffsets() const;
\n-
478
\n-
479private:
\n-
480 // Refinement classes (including all subclasses) build a Level:
\n-
481 friend class Refinement;
\n-
482 friend class TriRefinement;
\n-
483 friend class QuadRefinement;
\n-
484
\n-
485 //
\n-
486 // A Level is independent of subdivision scheme or options. While it may have been
\n-
487 // affected by them in its construction, they are not associated with it -- a Level
\n-
488 // is pure topology and any subdivision parameters are external.
\n-
489 //
\n-
490
\n-
491 // Simple members for inventory, etc.
\n-
492 int _faceCount;
\n-
493 int _edgeCount;
\n-
494 int _vertCount;
\n-
495
\n-
496 // The "depth" member is clearly useful in both the topological splitting and the
\n-
497 // stencil queries, but arguably it ties the Level to a hierarchy which counters
\n-
498 // the idea of it being independent.
\n-
499 int _depth;
\n-
500
\n-
501 // Maxima to help clients manage sizing of data buffers. Given "max valence",
\n-
502 // the "max edge faces" is strictly redundant as it will always be less, but
\n-
503 // since it will typically be so much less (i.e. 2) it is kept for now.
\n-
504 int _maxEdgeFaces;
\n-
505 int _maxValence;
\n-
506
\n-
507 //
\n-
508 // Topology vectors:
\n-
509 // Note that of all of these, only data for the face-edge relation is not
\n-
510 // stored in the osd::FarTables in any form. The FarTable vectors combine
\n-
511 // the edge-vert and edge-face relations. The eventual goal is that this
\n-
512 // data be part of the osd::Far classes and be a superset of the FarTable
\n-
513 // vectors, i.e. no data duplication or conversion. The fact that FarTable
\n-
514 // already stores 5 of the 6 possible relations should make the topology
\n-
515 // storage as a whole a non-issue.
\n-
516 //
\n-
517 // The vert-face-child and vert-edge-child indices are also arguably not
\n-
518 // a topology relation but more one for parent/child relations. But it is
\n-
519 // a topological relationship, and if named differently would not likely
\n-
520 // raise this. It has been named with "child" in the name as it does play
\n-
521 // a more significant role during subdivision in mapping between parent
\n-
522 // and child components, and so has been named to reflect that more clearly.
\n-
523 //
\n-
524
\n-
525 // Per-face:
\n-
526 std::vector<Index> _faceVertCountsAndOffsets; // 2 per face, redundant after level 0
\n-
527 std::vector<Index> _faceVertIndices; // 3 or 4 per face, variable at level 0
\n-
528 std::vector<Index> _faceEdgeIndices; // matches face-vert indices
\n-
529 std::vector<FTag> _faceTags; // 1 per face: includes "hole" tag
\n-
530
\n-
531 // Per-edge:
\n-
532 std::vector<Index> _edgeVertIndices; // 2 per edge
\n-
533 std::vector<Index> _edgeFaceCountsAndOffsets; // 2 per edge
\n-
534 std::vector<Index> _edgeFaceIndices; // varies with faces per edge
\n-
535 std::vector<LocalIndex> _edgeFaceLocalIndices; // varies with faces per edge
\n-
536
\n-
537 std::vector<float> _edgeSharpness; // 1 per edge
\n-
538 std::vector<ETag> _edgeTags; // 1 per edge: manifold, boundary, etc.
\n-
539
\n-
540 // Per-vertex:
\n-
541 std::vector<Index> _vertFaceCountsAndOffsets; // 2 per vertex
\n-
542 std::vector<Index> _vertFaceIndices; // varies with valence
\n-
543 std::vector<LocalIndex> _vertFaceLocalIndices; // varies with valence, 8-bit for now
\n-
544
\n-
545 std::vector<Index> _vertEdgeCountsAndOffsets; // 2 per vertex
\n-
546 std::vector<Index> _vertEdgeIndices; // varies with valence
\n-
547 std::vector<LocalIndex> _vertEdgeLocalIndices; // varies with valence, 8-bit for now
\n-
548
\n-
549 std::vector<float> _vertSharpness; // 1 per vertex
\n-
550 std::vector<VTag> _vertTags; // 1 per vertex: manifold, Sdc::Rule, etc.
\n-
551
\n-
552 // Face-varying channels:
\n-
553 std::vector<FVarLevel*> _fvarChannels;
\n-
554};
\n-
555
\n-
556//
\n-
557// Access/modify the vertices incident a given face:
\n-
558//
\n-
559inline ConstIndexArray
\n-
560Level::getFaceVertices(Index faceIndex) const {
\n-
561 return ConstIndexArray(&_faceVertIndices[_faceVertCountsAndOffsets[faceIndex*2+1]],
\n-
562 _faceVertCountsAndOffsets[faceIndex*2]);
\n-
563}
\n-
564inline IndexArray
\n-
565Level::getFaceVertices(Index faceIndex) {
\n-
566 return IndexArray(&_faceVertIndices[_faceVertCountsAndOffsets[faceIndex*2+1]],
\n-
567 _faceVertCountsAndOffsets[faceIndex*2]);
\n-
568}
\n-
569
\n-
570inline void
\n-
571Level::resizeFaceVertices(Index faceIndex, int count) {
\n-
572
\n-
573 int* countOffsetPair = &_faceVertCountsAndOffsets[faceIndex*2];
\n-
574
\n-
575 countOffsetPair[0] = count;
\n-
576 countOffsetPair[1] = (faceIndex == 0) ? 0 : (countOffsetPair[-2] + countOffsetPair[-1]);
\n-
577
\n-
578 _maxValence = std::max(_maxValence, count);
\n-
579}
\n-
580
\n-
581inline ConstIndexArray
\n-
582Level::getFaceVertices() const {
\n-
583 return ConstIndexArray(&_faceVertIndices[0], (int)_faceVertIndices.size());
\n-
584}
\n-
585
\n-
586//
\n-
587// Access/modify the edges incident a given face:
\n-
588//
\n-
589inline ConstIndexArray
\n-
590Level::getFaceEdges(Index faceIndex) const {
\n-
591 return ConstIndexArray(&_faceEdgeIndices[_faceVertCountsAndOffsets[faceIndex*2+1]],
\n-
592 _faceVertCountsAndOffsets[faceIndex*2]);
\n-
593}
\n-
594inline IndexArray
\n-
595Level::getFaceEdges(Index faceIndex) {
\n-
596 return IndexArray(&_faceEdgeIndices[_faceVertCountsAndOffsets[faceIndex*2+1]],
\n-
597 _faceVertCountsAndOffsets[faceIndex*2]);
\n-
598}
\n-
599
\n-
600//
\n-
601// Access/modify the faces incident a given vertex:
\n-
602//
\n-
603inline ConstIndexArray
\n-
604Level::getVertexFaces(Index vertIndex) const {
\n-
605 return ConstIndexArray( (&_vertFaceIndices[0]) + _vertFaceCountsAndOffsets[vertIndex*2+1],
\n-
606 _vertFaceCountsAndOffsets[vertIndex*2]);
\n-
607}
\n-
608inline IndexArray
\n-
609Level::getVertexFaces(Index vertIndex) {
\n-
610 return IndexArray( (&_vertFaceIndices[0]) + _vertFaceCountsAndOffsets[vertIndex*2+1],
\n-
611 _vertFaceCountsAndOffsets[vertIndex*2]);
\n-
612}
\n-
613
\n-
614inline ConstLocalIndexArray
\n-
615Level::getVertexFaceLocalIndices(Index vertIndex) const {
\n-
616 return ConstLocalIndexArray( (&_vertFaceLocalIndices[0]) + _vertFaceCountsAndOffsets[vertIndex*2+1],
\n-
617 _vertFaceCountsAndOffsets[vertIndex*2]);
\n-
618}
\n-
619inline LocalIndexArray
\n-
620Level::getVertexFaceLocalIndices(Index vertIndex) {
\n-
621 return LocalIndexArray( (&_vertFaceLocalIndices[0]) + _vertFaceCountsAndOffsets[vertIndex*2+1],
\n-
622 _vertFaceCountsAndOffsets[vertIndex*2]);
\n-
623}
\n-
624
\n-
625inline void
\n-
626Level::resizeVertexFaces(Index vertIndex, int count) {
\n-
627 int* countOffsetPair = &_vertFaceCountsAndOffsets[vertIndex*2];
\n-
628
\n-
629 countOffsetPair[0] = count;
\n-
630 countOffsetPair[1] = (vertIndex == 0) ? 0 : (countOffsetPair[-2] + countOffsetPair[-1]);
\n-
631}
\n-
632inline void
\n-
633Level::trimVertexFaces(Index vertIndex, int count) {
\n-
634 _vertFaceCountsAndOffsets[vertIndex*2] = count;
\n-
635}
\n-
636
\n-
637//
\n-
638// Access/modify the edges incident a given vertex:
\n-
639//
\n-
640inline ConstIndexArray
\n-
641Level::getVertexEdges(Index vertIndex) const {
\n-
642 return ConstIndexArray( (&_vertEdgeIndices[0]) +_vertEdgeCountsAndOffsets[vertIndex*2+1],
\n-
643 _vertEdgeCountsAndOffsets[vertIndex*2]);
\n-
644}
\n-
645inline IndexArray
\n-
646Level::getVertexEdges(Index vertIndex) {
\n-
647 return IndexArray( (&_vertEdgeIndices[0]) +_vertEdgeCountsAndOffsets[vertIndex*2+1],
\n-
648 _vertEdgeCountsAndOffsets[vertIndex*2]);
\n-
649}
\n-
650
\n-
651inline ConstLocalIndexArray
\n-
652Level::getVertexEdgeLocalIndices(Index vertIndex) const {
\n-
653 return ConstLocalIndexArray( (&_vertEdgeLocalIndices[0]) + _vertEdgeCountsAndOffsets[vertIndex*2+1],
\n-
654 _vertEdgeCountsAndOffsets[vertIndex*2]);
\n-
655}
\n-
656inline LocalIndexArray
\n-
657Level::getVertexEdgeLocalIndices(Index vertIndex) {
\n-
658 return LocalIndexArray( (&_vertEdgeLocalIndices[0]) + _vertEdgeCountsAndOffsets[vertIndex*2+1],
\n-
659 _vertEdgeCountsAndOffsets[vertIndex*2]);
\n-
660}
\n-
661
\n-
662inline void
\n-
663Level::resizeVertexEdges(Index vertIndex, int count) {
\n-
664 int* countOffsetPair = &_vertEdgeCountsAndOffsets[vertIndex*2];
\n-
665
\n-
666 countOffsetPair[0] = count;
\n-
667 countOffsetPair[1] = (vertIndex == 0) ? 0 : (countOffsetPair[-2] + countOffsetPair[-1]);
\n-
668
\n-
669 _maxValence = std::max(_maxValence, count);
\n-
670}
\n-
671inline void
\n-
672Level::trimVertexEdges(Index vertIndex, int count) {
\n-
673 _vertEdgeCountsAndOffsets[vertIndex*2] = count;
\n-
674}
\n-
675
\n-
676inline void
\n-
677Level::setMaxValence(int valence) {
\n-
678 _maxValence = valence;
\n-
679}
\n-
680
\n-
681//
\n-
682// Access/modify the vertices incident a given edge:
\n-
683//
\n-
684inline ConstIndexArray
\n-
685Level::getEdgeVertices(Index edgeIndex) const {
\n-
686 return ConstIndexArray(&_edgeVertIndices[edgeIndex*2], 2);
\n-
687}
\n-
688inline IndexArray
\n-
689Level::getEdgeVertices(Index edgeIndex) {
\n-
690 return IndexArray(&_edgeVertIndices[edgeIndex*2], 2);
\n-
691}
\n-
692
\n-
693//
\n-
694// Access/modify the faces incident a given edge:
\n-
695//
\n-
696inline ConstIndexArray
\n-
697Level::getEdgeFaces(Index edgeIndex) const {
\n-
698 return ConstIndexArray(&_edgeFaceIndices[0] +
\n-
699 _edgeFaceCountsAndOffsets[edgeIndex*2+1],
\n-
700 _edgeFaceCountsAndOffsets[edgeIndex*2]);
\n-
701}
\n-
702inline IndexArray
\n-
703Level::getEdgeFaces(Index edgeIndex) {
\n-
704 return IndexArray(&_edgeFaceIndices[0] +
\n-
705 _edgeFaceCountsAndOffsets[edgeIndex*2+1],
\n-
706 _edgeFaceCountsAndOffsets[edgeIndex*2]);
\n-
707}
\n-
708
\n-
709inline ConstLocalIndexArray
\n-
710Level::getEdgeFaceLocalIndices(Index edgeIndex) const {
\n-
711 return ConstLocalIndexArray(&_edgeFaceLocalIndices[0] +
\n-
712 _edgeFaceCountsAndOffsets[edgeIndex*2+1],
\n-
713 _edgeFaceCountsAndOffsets[edgeIndex*2]);
\n-
714}
\n-
715inline LocalIndexArray
\n-
716Level::getEdgeFaceLocalIndices(Index edgeIndex) {
\n-
717 return LocalIndexArray(&_edgeFaceLocalIndices[0] +
\n-
718 _edgeFaceCountsAndOffsets[edgeIndex*2+1],
\n-
719 _edgeFaceCountsAndOffsets[edgeIndex*2]);
\n-
720}
\n-
721
\n-
722inline void
\n-
723Level::resizeEdgeFaces(Index edgeIndex, int count) {
\n-
724 int* countOffsetPair = &_edgeFaceCountsAndOffsets[edgeIndex*2];
\n-
725
\n-
726 countOffsetPair[0] = count;
\n-
727 countOffsetPair[1] = (edgeIndex == 0) ? 0 : (countOffsetPair[-2] + countOffsetPair[-1]);
\n-
728
\n-
729 _maxEdgeFaces = std::max(_maxEdgeFaces, count);
\n-
730}
\n-
731inline void
\n-
732Level::trimEdgeFaces(Index edgeIndex, int count) {
\n-
733 _edgeFaceCountsAndOffsets[edgeIndex*2] = count;
\n-
734}
\n-
735
\n-
736//
\n-
737// Access/modify sharpness values:
\n-
738//
\n-
739inline float
\n-
740Level::getEdgeSharpness(Index edgeIndex) const {
\n-
741 return _edgeSharpness[edgeIndex];
\n-
742}
\n-
743inline float&
\n-
744Level::getEdgeSharpness(Index edgeIndex) {
\n-
745 return _edgeSharpness[edgeIndex];
\n-
746}
\n-
747
\n-
748inline float
\n-
749Level::getVertexSharpness(Index vertIndex) const {
\n-
750 return _vertSharpness[vertIndex];
\n-
751}
\n-
752inline float&
\n-
753Level::getVertexSharpness(Index vertIndex) {
\n-
754 return _vertSharpness[vertIndex];
\n-
755}
\n-
756
\n-
757inline Sdc::Crease::Rule
\n-
758Level::getVertexRule(Index vertIndex) const {
\n-
759 return (Sdc::Crease::Rule) _vertTags[vertIndex]._rule;
\n-
760}
\n-
761
\n-
762//
\n-
763// Access/modify hole tag:
\n-
764//
\n-
765inline void
\n-
766Level::setFaceHole(Index faceIndex, bool b) {
\n-
767 _faceTags[faceIndex]._hole = b;
\n-
768}
\n-
769inline bool
\n-
770Level::isFaceHole(Index faceIndex) const {
\n-
771 return _faceTags[faceIndex]._hole;
\n-
772}
\n-
773
\n-
774//
\n-
775// Access/modify non-manifold tags:
\n-
776//
\n-
777inline void
\n-
778Level::setEdgeNonManifold(Index edgeIndex, bool b) {
\n-
779 _edgeTags[edgeIndex]._nonManifold = b;
\n-
780}
\n-
781inline bool
\n-
782Level::isEdgeNonManifold(Index edgeIndex) const {
\n-
783 return _edgeTags[edgeIndex]._nonManifold;
\n-
784}
\n-
785
\n-
786inline void
\n-
787Level::setVertexNonManifold(Index vertIndex, bool b) {
\n-
788 _vertTags[vertIndex]._nonManifold = b;
\n-
789}
\n-
790inline bool
\n-
791Level::isVertexNonManifold(Index vertIndex) const {
\n-
792 return _vertTags[vertIndex]._nonManifold;
\n-
793}
\n-
794
\n-
795//
\n-
796// Sizing methods to allocate space:
\n-
797//
\n-
798inline void
\n-
799Level::resizeFaces(int faceCount) {
\n-
800 _faceCount = faceCount;
\n-
801 _faceVertCountsAndOffsets.resize(2 * faceCount);
\n-
802
\n-
803 _faceTags.resize(faceCount);
\n-
804 std::memset((void*) &_faceTags[0], 0, _faceCount * sizeof(FTag));
\n-
805}
\n-
806inline void
\n-
807Level::resizeFaceVertices(int totalFaceVertCount) {
\n-
808 _faceVertIndices.resize(totalFaceVertCount);
\n-
809}
\n-
810inline void
\n-
811Level::resizeFaceEdges(int totalFaceEdgeCount) {
\n-
812 _faceEdgeIndices.resize(totalFaceEdgeCount);
\n-
813}
\n-
814
\n-
815inline void
\n-
816Level::resizeEdges(int edgeCount) {
\n-
817
\n-
818 _edgeCount = edgeCount;
\n-
819 _edgeFaceCountsAndOffsets.resize(2 * edgeCount);
\n-
820
\n-
821 _edgeSharpness.resize(edgeCount);
\n-
822 _edgeTags.resize(edgeCount);
\n-
823
\n-
824 if (edgeCount>0) {
\n-
825 std::memset((void*) &_edgeTags[0], 0, _edgeCount * sizeof(ETag));
\n-
826 }
\n-
827}
\n-
828inline void
\n-
829Level::resizeEdgeVertices() {
\n-
830
\n-
831 _edgeVertIndices.resize(2 * _edgeCount);
\n-
832}
\n-
833inline void
\n-
834Level::resizeEdgeFaces(int totalEdgeFaceCount) {
\n-
835
\n-
836 _edgeFaceIndices.resize(totalEdgeFaceCount);
\n-
837 _edgeFaceLocalIndices.resize(totalEdgeFaceCount);
\n-
838}
\n-
839
\n-
840inline void
\n-
841Level::resizeVertices(int vertCount) {
\n-
842
\n-
843 _vertCount = vertCount;
\n-
844 _vertFaceCountsAndOffsets.resize(2 * vertCount);
\n-
845 _vertEdgeCountsAndOffsets.resize(2 * vertCount);
\n-
846
\n-
847 _vertSharpness.resize(vertCount);
\n-
848 _vertTags.resize(vertCount);
\n-
849 std::memset((void*) &_vertTags[0], 0, _vertCount * sizeof(VTag));
\n-
850}
\n-
851inline void
\n-
852Level::resizeVertexFaces(int totalVertFaceCount) {
\n-
853
\n-
854 _vertFaceIndices.resize(totalVertFaceCount);
\n-
855 _vertFaceLocalIndices.resize(totalVertFaceCount);
\n-
856}
\n-
857inline void
\n-
858Level::resizeVertexEdges(int totalVertEdgeCount) {
\n-
859
\n-
860 _vertEdgeIndices.resize(totalVertEdgeCount);
\n-
861 _vertEdgeLocalIndices.resize(totalVertEdgeCount);
\n-
862}
\n-
863
\n-
864inline IndexArray
\n-
865Level::shareFaceVertCountsAndOffsets() const {
\n-
866 // XXXX manuelk we have to force const casting here (classes don't 'share'
\n-
867 // members usually...)
\n-
868 return IndexArray(const_cast<Index *>(&_faceVertCountsAndOffsets[0]),
\n-
869 (int)_faceVertCountsAndOffsets.size());
\n-
870}
\n-
871
\n-
872} // end namespace internal
\n-
873} // end namespace Vtr
\n-
874
\n-
875} // end namespace OPENSUBDIV_VERSION
\n-
876using namespace OPENSUBDIV_VERSION;
\n-
877} // end namespace OpenSubdiv
\n-
878
\n-
879#endif /* OPENSUBDIV3_VTR_LEVEL_H */
\n+
456#endif /* OPENSUBDIV3_VTR_REFINEMENT_H */
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstLocalIndexArray
ConstArray< LocalIndex > ConstLocalIndexArray
Definition: types.h:83
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Vtr::LocalIndexArray
Array< LocalIndex > LocalIndexArray
Definition: types.h:82
\n \n+
OpenSubdiv::OPENSUBDIV_VERSION::Vtr::IndexVector
std::vector< Index > IndexVector
Definition: types.h:77
\n
OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstIndexArray
ConstArray< Index > ConstIndexArray
Definition: types.h:80
\n-\n \n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Split
Split
Enumerated type for all face splitting schemes.
Definition: types.h:47
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-level.h\n+refinement.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2014 DreamWorks Animation LLC.\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -29,997 +29,548 @@\n 17//\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n- 24#ifndef OPENSUBDIV3_VTR_LEVEL_H\n- 25#define OPENSUBDIV3_VTR_LEVEL_H\n+ 24#ifndef OPENSUBDIV3_VTR_REFINEMENT_H\n+ 25#define OPENSUBDIV3_VTR_REFINEMENT_H\n 26\n 27#include \"../version.h\"\n 28\n 29#include \"../sdc/types.h\"\n- 30#include \"../sdc/crease.h\"\n- 31#include \"../sdc/options.h\"\n- 32#include \"../vtr/types.h\"\n+ 30#include \"../sdc/options.h\"\n+ 31#include \"../vtr/types.h\"\n+ 32#include \"../vtr/level.h\"\n 33\n- 34#include \n- 35#include \n- 36#include \n- 37#include \n- 38\n- 39\n- 40namespace OpenSubdiv {\n- 41namespace OPENSUBDIV_VERSION {\n- 42\n- 43namespace Vtr {\n- 44namespace internal {\n- 45\n- 46class Refinement;\n- 47class TriRefinement;\n- 48class QuadRefinement;\n- 49class FVarRefinement;\n- 50class FVarLevel;\n- 51\n- 52//\n- 53// Level:\n- 54// A refinement level includes a vectorized representation of the topology\n- 55// for a particular subdivision level. The topology is \"complete\" in that\n-any\n- 56// level can be used as the base level of another subdivision hierarchy and\n-can\n- 57// be considered a complete mesh independent of its ancestors. It currently\n- 58// does contain a \"depth\" member -- as some inferences can then be made\n-about\n- 59// the topology (i.e. all quads or all tris if not level 0).\n+ 34#include \n+ 35\n+ 36//\n+ 37// Declaration for the main refinement class (Refinement) and its pre-\n+requisites:\n+ 38//\n+ 39namespace OpenSubdiv {\n+ 40namespace OPENSUBDIV_VERSION {\n+ 41\n+ 42namespace Vtr {\n+ 43namespace internal {\n+ 44\n+ 45class FVarRefinement;\n+ 46\n+ 47//\n+ 48// Refinement:\n+ 49// A refinement is a mapping between two levels -- relating the components\n+in the original\n+ 50// (parent) level to the one refined (child). The refinement may be complete\n+(uniform) or sparse\n+ 51// (adaptive or otherwise selective), so not all components in the parent\n+level will spawn\n+ 52// components in the child level.\n+ 53//\n+ 54// Refinement is an abstract class and expects subclasses corresponding to\n+the different types\n+ 55// of topological splits that the supported subdivision schemes collectively\n+require, i.e. those\n+ 56// listed in Sdc::SplitType. Note the virtual requirements expected of the\n+subclasses in the list\n+ 57// of protected methods -- they differ mainly in the topology that is\n+created in the child Level\n+ 58// and not the propagation of tags through refinement, subdivision of\n+sharpness values or the\n+ 59// treatment of face-varying data. The primary subclasses are QuadRefinement\n+and TriRefinement.\n 60//\n- 61// This class is intended for private use within the library. There are\n-still\n- 62// opportunities to specialize levels -- e.g. those supporting N-sided faces\n-vs\n- 63// those that are purely quads or tris -- so we prefer to insulate it from\n-public\n- 64// access.\n- 65//\n- 66// The representation of topology here is to store six topological\n-relationships\n- 67// in tables of integers. Each is stored in its own array(s) so the result\n-is\n- 68// a SOA representation of the topology. The six relations are:\n- 69//\n- 70// - face-verts: vertices incident/comprising a face\n- 71// - face-edges: edges incident a face\n- 72// - edge-verts: vertices incident/comprising an edge\n- 73// - edge-faces: faces incident an edge\n- 74// - vert-faces: faces incident a vertex\n- 75// - vert-edges: edges incident a vertex\n- 76//\n- 77// There is some redundancy here but the intent is not that this be a\n-minimal\n- 78// representation, the intent is that it be amenable to refinement. Classes\n-in\n- 79// the Far layer essentially store 5 of these 6 in a permuted form -- we add\n- 80// the face-edges here to simplify refinement.\n- 81//\n+ 61// At a high level, all that is necessary in terms of interface is to\n+construct, initialize\n+ 62// (linking the two levels), optionally select components for sparse\n+refinement (via use of the\n+ 63// SparseSelector) and call the refine() method. This usage is expected of\n+Far::TopologyRefiner.\n+ 64//\n+ 65// Since we really want this class to be restricted from public access\n+eventually, all methods\n+ 66// begin with lower case (as is the convention for protected methods) and\n+the list of friends\n+ 67// will be maintained more strictly.\n+ 68//\n+ 69class Refinement {\n+ 70\n+ 71public:\n+ 72 Refinement(Level const & parent, Level & child, Sdc::Options const&\n+schemeOptions);\n+ 73 virtual ~Refinement();\n+ 74\n+ 75 Level const& parent() const { return *_parent; }\n+ 76 Level const& child() const { return *_child; }\n+ 77 Level& child() { return *_child; }\n+ 78\n+ 79 Sdc::Split getSplitType() const { return _splitType; }\n+ 80 int getRegularFaceSize() const { return _regFaceSize; }\n+ 81 Sdc::Options getOptions() const { return _options; }\n 82\n- 83class Level {\n- 84\n- 85public:\n- 86 //\n- 87 // Simple nested types to hold the tags for each component type -- some of\n- 88 // which are user-specified features (e.g. whether a face is a hole or not)\n- 89 // while others indicate the topological nature of the component, how it\n- 90 // is affected by creasing in its neighborhood, etc.\n- 91 //\n- 92 // Most of these properties are passed down to child components during\n- 93 // refinement, but some -- notably the designation of a component as semi-\n- 94 // sharp -- require re-determination as sharpness values are reduced at\n-each\n- 95 // level.\n- 96 //\n- 97 struct VTag {\n- 98 VTag() { }\n- 99\n- 100 // When cleared, the VTag ALMOST represents a smooth, regular, interior\n- 101 // vertex -- the Type enum requires a bit be explicitly set for Smooth,\n- 102 // so that must be done explicitly if desired on initialization.\n- 103 void clear() { std::memset((void*) this, 0, sizeof(VTag)); }\n- 104\n- 105 typedef unsigned short VTagSize;\n- 106\n- 107 VTagSize _nonManifold : 1; // fixed\n- 108 VTagSize _xordinary : 1; // fixed\n- 109 VTagSize _boundary : 1; // fixed\n- 110 VTagSize _corner : 1; // fixed\n- 111 VTagSize _infSharp : 1; // fixed\n- 112 VTagSize _semiSharp : 1; // variable\n- 113 VTagSize _semiSharpEdges : 1; // variable\n- 114 VTagSize _rule : 4; // variable when _semiSharp\n- 115\n- 116 // These next to tags are complementary -- the \"incomplete\" tag is only\n- 117 // relevant for refined levels while the \"incident an irregular face\" tag\n- 118 // is only relevant for the base level. They could be combined as both\n- 119 // indicate \"no full regular ring\" around a vertex\n- 120 VTagSize _incomplete : 1; // variable only set in refined levels\n- 121 VTagSize _incidIrregFace : 1; // variable only set in base level\n+ 83 // Face-varying:\n+ 84 int getNumFVarChannels() const { return (int) _fvarChannels.size(); }\n+ 85\n+ 86 FVarRefinement const & getFVarRefinement(int c) const { return\n+*_fvarChannels[c]; }\n+ 87\n+ 88 //\n+ 89 // Options associated with the actual refinement operation, which may end\n+up\n+ 90 // quite involved if we want to allow for the refinement of data that is\n+not\n+ 91 // of interest to be suppressed. For now we have:\n+ 92 //\n+ 93 // \"sparse\": the alternative to uniform refinement, which requires that\n+ 94 // components be previously selected/marked to be included.\n+ 95 //\n+ 96 // \"minimal topology\": this is one that may get broken down into a finer\n+ 97 // set of options. It suppresses \"full topology\" in the child level\n+ 98 // and only generates what is minimally necessary for interpolation --\n+ 99 // which requires at least the face-vertices for faces, but also the\n+ 100 // vertex-faces for any face-varying channels present. So it will\n+ 101 // generate one or two of the six possible topological relations.\n+ 102 //\n+ 103 // These are strictly controlled right now, e.g. for sparse refinement, we\n+ 104 // currently enforce full topology at the finest level to allow for\n+subsequent\n+ 105 // patch construction.\n+ 106 //\n+ 107 struct Options {\n+ 108 Options() : _sparse(false),\n+ 109 _faceVertsFirst(false),\n+ 110 _minimalTopology(false)\n+ 111 { }\n+ 112\n+ 113 unsigned int _sparse : 1;\n+ 114 unsigned int _faceVertsFirst : 1;\n+ 115 unsigned int _minimalTopology : 1;\n+ 116\n+ 117 // Still under consideration:\n+ 118 //unsigned int _childToParentMap : 1;\n+ 119 };\n+ 120\n+ 121 void refine(Options options = Options());\n 122\n- 123 // Tags indicating incident infinitely-sharp (permanent) features\n- 124 VTagSize _infSharpEdges : 1; // fixed\n- 125 VTagSize _infSharpCrease : 1; // fixed\n- 126 VTagSize _infIrregular : 1; // fixed\n- 127\n- 128 // Alternate constructor and accessor for dealing with integer bits\n-directly:\n- 129 explicit VTag(VTagSize bits) {\n- 130 std::memcpy(this, &bits, sizeof(bits));\n- 131 }\n- 132 VTagSize getBits() const {\n- 133 VTagSize bits;\n- 134 std::memcpy(&bits, this, sizeof(bits));\n- 135 return bits;\n- 136 }\n- 137\n- 138 static VTag BitwiseOr(VTag const vTags[], int size = 4);\n- 139 };\n- 140 struct ETag {\n- 141 ETag() { }\n- 142\n- 143 // When cleared, the ETag represents a smooth, manifold, interior edge\n- 144 void clear() { std::memset((void*) this, 0, sizeof(ETag)); }\n- 145\n- 146 typedef unsigned char ETagSize;\n+ 123 bool hasFaceVerticesFirst() const { return _faceVertsFirst; }\n+ 124\n+ 125public:\n+ 126 //\n+ 127 // Access to members -- some testing classes (involving vertex\n+interpolation)\n+ 128 // currently make use of these:\n+ 129 //\n+ 130 int getNumChildFacesFromFaces() const { return _childFaceFromFaceCount; }\n+ 131 int getNumChildEdgesFromFaces() const { return _childEdgeFromFaceCount; }\n+ 132 int getNumChildEdgesFromEdges() const { return _childEdgeFromEdgeCount; }\n+ 133 int getNumChildVerticesFromFaces() const { return _childVertFromFaceCount;\n+}\n+ 134 int getNumChildVerticesFromEdges() const { return _childVertFromEdgeCount;\n+}\n+ 135 int getNumChildVerticesFromVertices() const { return\n+_childVertFromVertCount; }\n+ 136\n+ 137 Index getFirstChildFaceFromFaces() const { return _firstChildFaceFromFace;\n+}\n+ 138 Index getFirstChildEdgeFromFaces() const { return _firstChildEdgeFromFace;\n+}\n+ 139 Index getFirstChildEdgeFromEdges() const { return _firstChildEdgeFromEdge;\n+}\n+ 140 Index getFirstChildVertexFromFaces() const { return\n+_firstChildVertFromFace; }\n+ 141 Index getFirstChildVertexFromEdges() const { return\n+_firstChildVertFromEdge; }\n+ 142 Index getFirstChildVertexFromVertices() const { return\n+_firstChildVertFromVert; }\n+ 143\n+ 144 Index getFaceChildVertex(Index f) const { return _faceChildVertIndex[f]; }\n+ 145 Index getEdgeChildVertex(Index e) const { return _edgeChildVertIndex[e]; }\n+ 146 Index getVertexChildVertex(Index v) const { return _vertChildVertIndex[v];\n+}\n 147\n- 148 ETagSize _nonManifold : 1; // fixed\n- 149 ETagSize _boundary : 1; // fixed\n- 150 ETagSize _infSharp : 1; // fixed\n- 151 ETagSize _semiSharp : 1; // variable\n- 152\n- 153 // Alternate constructor and accessor for dealing with integer bits\n-directly:\n- 154 explicit ETag(ETagSize bits) {\n- 155 std::memcpy(this, &bits, sizeof(bits));\n- 156 }\n- 157 ETagSize getBits() const {\n- 158 ETagSize bits;\n- 159 std::memcpy(&bits, this, sizeof(bits));\n- 160 return bits;\n- 161 }\n- 162\n- 163 static ETag BitwiseOr(ETag const eTags[], int size = 4);\n- 164 };\n- 165 struct FTag {\n- 166 FTag() { }\n- 167\n- 168 void clear() { std::memset((void*) this, 0, sizeof(FTag)); }\n- 169\n- 170 typedef unsigned char FTagSize;\n- 171\n- 172 FTagSize _hole : 1; // fixed\n- 173\n- 174 // On deck -- coming soon...\n- 175 //FTagSize _hasEdits : 1; // variable\n- 176 };\n- 177\n- 178 // Additional simple struct to identify a \"span\" around a vertex, i.e. a\n- 179 // subset of the faces around a vertex delimited by some property (e.g. a\n- 180 // face-varying discontinuity, an inf-sharp edge, etc.)\n- 181 //\n- 182 // The span requires an \"origin\" and a \"size\" to fully define its extent.\n- 183 // Use of the size is required over a leading/trailing pair as the valence\n- 184 // around a non-manifold vertex cannot be trivially determined from two\n- 185 // extremeties. Similarly a start face is chosen over an edge as starting\n- 186 // with a manifold edge is ambiguous. Additional tags also support\n- 187 // non-manifold cases, e.g. periodic spans at the apex of a double cone.\n+ 148 ConstIndexArray getFaceChildFaces(Index parentFace) const;\n+ 149 ConstIndexArray getFaceChildEdges(Index parentFace) const;\n+ 150 ConstIndexArray getEdgeChildEdges(Index parentEdge) const;\n+ 151\n+ 152 // Child-to-parent relationships\n+ 153 bool isChildVertexComplete(Index v) const { return ! _childVertexTag\n+[v]._incomplete; }\n+ 154\n+ 155 Index getChildFaceParentFace(Index f) const { return _childFaceParentIndex\n+[f]; }\n+ 156 int getChildFaceInParentFace(Index f) const { return _childFaceTag\n+[f]._indexInParent; }\n+ 157\n+ 158 Index getChildEdgeParentIndex(Index e) const { return\n+_childEdgeParentIndex[e]; }\n+ 159\n+ 160 Index getChildVertexParentIndex(Index v) const { return\n+_childVertexParentIndex[v]; }\n+ 161\n+ 162//\n+ 163// Modifiers intended for internal/protected use:\n+ 164//\n+ 165public:\n+ 166\n+ 167 IndexArray getFaceChildFaces(Index parentFace);\n+ 168 IndexArray getFaceChildEdges(Index parentFace);\n+ 169 IndexArray getEdgeChildEdges(Index parentEdge);\n+ 170\n+ 171public:\n+ 172 //\n+ 173 // Tags have now been added per-component in Level, but there is\n+additional need to tag\n+ 174 // components within Refinement -- we can't tag the parent level\n+components for any\n+ 175 // refinement (in order to keep it const) and tags associated with\n+children that are\n+ 176 // specific to the child-to-parent mapping may not be warranted in the\n+child level.\n+ 177 //\n+ 178 // Parent tags are only required for sparse refinement. The main property\n+to tag is\n+ 179 // whether a component was selected, and so a single SparseTag is used for\n+all three\n+ 180 // component types. Tagging if a component is \"transitional\" is also\n+useful. This may\n+ 181 // only be necessary for edges but is currently packed into a mask per-\n+edge for faces,\n+ 182 // which could be deferred, in which case \"transitional\" could be a single\n+bit.\n+ 183 //\n+ 184 // Child tags are part of the child-to-parent mapping, which consists of\n+the parent\n+ 185 // component index for each child component, plus a tag for the child\n+indicating more\n+ 186 // about its relationship to its parent, e.g. is it completely defined,\n+what the parent\n+ 187 // component type is, what is the index of the child within its parent,\n+etc.\n 188 //\n- 189 // Currently setting the size to 0 or leaving the span \"unassigned\" is an\n- 190 // indication to use the full neighborhood rather than a subset -- prefer\n- 191 // use of the const method here to direct inspection of the member.\n- 192 //\n- 193 struct VSpan {\n- 194 VSpan() { std::memset((void*) this, 0, sizeof(VSpan)); }\n+ 189 struct SparseTag {\n+ 190 SparseTag() : _selected(0), _transitional(0) { }\n+ 191\n+ 192 unsigned char _selected : 1; // component specifically selected for\n+refinement\n+ 193 unsigned char _transitional : 4; // adjacent to a refined component (4-\n+bits for face)\n+ 194 };\n 195\n- 196 void clear() { std::memset((void*) this, 0, sizeof(VSpan)); }\n- 197 bool isAssigned() const { return _numFaces > 0; }\n+ 196 struct ChildTag {\n+ 197 ChildTag() { }\n 198\n- 199 LocalIndex _numFaces;\n- 200 LocalIndex _startFace;\n- 201 LocalIndex _cornerInSpan;\n- 202\n- 203 unsigned short _periodic : 1;\n- 204 unsigned short _sharp : 1;\n- 205 };\n- 206\n- 207public:\n- 208 Level();\n- 209 ~Level();\n- 210\n- 211 // Simple accessors:\n- 212 int getDepth() const { return _depth; }\n- 213\n- 214 int getNumVertices() const { return _vertCount; }\n- 215 int getNumFaces() const { return _faceCount; }\n- 216 int getNumEdges() const { return _edgeCount; }\n- 217\n- 218 // More global sizes may prove useful...\n- 219 int getNumFaceVerticesTotal() const { return (int) _faceVertIndices.size\n-(); }\n- 220 int getNumFaceEdgesTotal() const { return (int) _faceEdgeIndices.size(); }\n- 221 int getNumEdgeVerticesTotal() const { return (int) _edgeVertIndices.size\n-(); }\n- 222 int getNumEdgeFacesTotal() const { return (int) _edgeFaceIndices.size(); }\n- 223 int getNumVertexFacesTotal() const { return (int) _vertFaceIndices.size();\n-}\n- 224 int getNumVertexEdgesTotal() const { return (int) _vertEdgeIndices.size();\n-}\n- 225\n- 226 int getMaxValence() const { return _maxValence; }\n- 227 int getMaxEdgeFaces() const { return _maxEdgeFaces; }\n- 228\n- 229 // Methods to access the relation tables/indices -- note that for some\n-relations\n- 230 // (i.e. those where a component is \"contained by\" a neighbor, or more\n-generally\n- 231 // when the neighbor is a simplex of higher dimension) we store an\n-additional\n- 232 // \"local index\", e.g. for the case of vert-faces if one of the faces F[i]\n-is\n- 233 // incident a vertex V, then L[i] is the \"local index\" in F[i] of vertex\n-V.\n- 234 // Once have only quads (or tris), this local index need only occupy two\n-bits\n- 235 // and could conceivably be packed into the same integer as the face\n-index, but\n- 236 // for now, given the need to support faces of potentially high valence\n-we'll\n- 237 // use an 8- or 16-bit integer.\n- 238 //\n- 239 // Methods to access the six topological relations:\n- 240 ConstIndexArray getFaceVertices(Index faceIndex) const;\n- 241 ConstIndexArray getFaceEdges(Index faceIndex) const;\n- 242 ConstIndexArray getEdgeVertices(Index edgeIndex) const;\n- 243 ConstIndexArray getEdgeFaces(Index edgeIndex) const;\n- 244 ConstIndexArray getVertexFaces(Index vertIndex) const;\n- 245 ConstIndexArray getVertexEdges(Index vertIndex) const;\n- 246\n- 247 ConstLocalIndexArray getEdgeFaceLocalIndices(Index edgeIndex) const;\n- 248 ConstLocalIndexArray getVertexFaceLocalIndices(Index vertIndex) const;\n- 249 ConstLocalIndexArray getVertexEdgeLocalIndices(Index vertIndex) const;\n+ 199 unsigned char _incomplete : 1; // incomplete neighborhood to represent\n+limit of parent\n+ 200 unsigned char _parentType : 2; // type of parent component: vertex, edge\n+or face\n+ 201 unsigned char _indexInParent : 2; // index of child wrt parent: 0-3, or\n+iterative if N > 4\n+ 202 };\n+ 203\n+ 204 // Methods to access and modify tags:\n+ 205 SparseTag const & getParentFaceSparseTag( Index f) const { return\n+_parentFaceTag[f]; }\n+ 206 SparseTag const & getParentEdgeSparseTag( Index e) const { return\n+_parentEdgeTag[e]; }\n+ 207 SparseTag const & getParentVertexSparseTag(Index v) const { return\n+_parentVertexTag[v]; }\n+ 208\n+ 209 SparseTag & getParentFaceSparseTag( Index f) { return _parentFaceTag[f]; }\n+ 210 SparseTag & getParentEdgeSparseTag( Index e) { return _parentEdgeTag[e]; }\n+ 211 SparseTag & getParentVertexSparseTag(Index v) { return _parentVertexTag\n+[v]; }\n+ 212\n+ 213 ChildTag const & getChildFaceTag( Index f) const { return _childFaceTag\n+[f]; }\n+ 214 ChildTag const & getChildEdgeTag( Index e) const { return _childEdgeTag\n+[e]; }\n+ 215 ChildTag const & getChildVertexTag(Index v) const { return _childVertexTag\n+[v]; }\n+ 216\n+ 217 ChildTag & getChildFaceTag( Index f) { return _childFaceTag[f]; }\n+ 218 ChildTag & getChildEdgeTag( Index e) { return _childEdgeTag[e]; }\n+ 219 ChildTag & getChildVertexTag(Index v) { return _childVertexTag[v]; }\n+ 220\n+ 221// Remaining methods should really be protected -- for use by subclasses...\n+ 222public:\n+ 223 //\n+ 224 // Methods involved in constructing the parent-to-child mapping -- when\n+the\n+ 225 // refinement is sparse, additional methods are needed to identify the\n+selection:\n+ 226 //\n+ 227 void populateParentToChildMapping();\n+ 228 void populateParentChildIndices();\n+ 229 void printParentToChildMapping() const;\n+ 230\n+ 231 virtual void allocateParentChildIndices() = 0;\n+ 232\n+ 233 // Supporting method for sparse refinement:\n+ 234 void initializeSparseSelectionTags();\n+ 235 void markSparseChildComponentIndices();\n+ 236 void markSparseVertexChildren();\n+ 237 void markSparseEdgeChildren();\n+ 238\n+ 239 virtual void markSparseFaceChildren() = 0;\n+ 240\n+ 241 void initializeChildComponentCounts();\n+ 242\n+ 243 //\n+ 244 // Methods involved in constructing the child-to-parent mapping:\n+ 245 //\n+ 246 void populateChildToParentMapping();\n+ 247\n+ 248 void populateFaceParentVectors(ChildTag const initialChildTags[2][4]);\n+ 249 void populateFaceParentFromParentFaces(ChildTag const initialChildTags[2]\n+[4]);\n 250\n- 251 // Replace these with access to sharpness buffers/arrays rather than\n-elements:\n- 252 float getEdgeSharpness(Index edgeIndex) const;\n- 253 float getVertexSharpness(Index vertIndex) const;\n- 254 Sdc::Crease::Rule getVertexRule(Index vertIndex) const;\n- 255\n- 256 Index findEdge(Index v0Index, Index v1Index) const;\n- 257\n- 258 // Holes\n- 259 void setFaceHole(Index faceIndex, bool b);\n- 260 bool isFaceHole(Index faceIndex) const;\n- 261\n- 262 // Face-varying\n- 263 Sdc::Options getFVarOptions(int channel) const;\n- 264 int getNumFVarChannels() const { return (int) _fvarChannels.size(); }\n- 265 int getNumFVarValues(int channel) const;\n- 266 ConstIndexArray getFaceFVarValues(Index faceIndex, int channel) const;\n+ 251 void populateEdgeParentVectors(ChildTag const initialChildTags[2][4]);\n+ 252 void populateEdgeParentFromParentFaces(ChildTag const initialChildTags[2]\n+[4]);\n+ 253 void populateEdgeParentFromParentEdges(ChildTag const initialChildTags[2]\n+[4]);\n+ 254\n+ 255 void populateVertexParentVectors(ChildTag const initialChildTags[2][4]);\n+ 256 void populateVertexParentFromParentFaces(ChildTag const initialChildTags\n+[2][4]);\n+ 257 void populateVertexParentFromParentEdges(ChildTag const initialChildTags\n+[2][4]);\n+ 258 void populateVertexParentFromParentVertices(ChildTag const\n+initialChildTags[2][4]);\n+ 259\n+ 260 //\n+ 261 // Methods involved in propagating component tags from parent to child:\n+ 262 //\n+ 263 void propagateComponentTags();\n+ 264\n+ 265 void populateFaceTagVectors();\n+ 266 void populateFaceTagsFromParentFaces();\n 267\n- 268 FVarLevel & getFVarLevel(int channel) { return *_fvarChannels[channel]; }\n- 269 FVarLevel const & getFVarLevel(int channel) const { return *_fvarChannels\n-[channel]; }\n- 270\n- 271 // Manifold/non-manifold tags:\n- 272 void setEdgeNonManifold(Index edgeIndex, bool b);\n- 273 bool isEdgeNonManifold(Index edgeIndex) const;\n- 274\n- 275 void setVertexNonManifold(Index vertIndex, bool b);\n- 276 bool isVertexNonManifold(Index vertIndex) const;\n- 277\n- 278 // General access to all component tags:\n- 279 VTag const & getVertexTag(Index vertIndex) const { return _vertTags\n-[vertIndex]; }\n- 280 ETag const & getEdgeTag(Index edgeIndex) const { return _edgeTags\n-[edgeIndex]; }\n- 281 FTag const & getFaceTag(Index faceIndex) const { return _faceTags\n-[faceIndex]; }\n- 282\n- 283 VTag & getVertexTag(Index vertIndex) { return _vertTags[vertIndex]; }\n- 284 ETag & getEdgeTag(Index edgeIndex) { return _edgeTags[edgeIndex]; }\n- 285 FTag & getFaceTag(Index faceIndex) { return _faceTags[faceIndex]; }\n- 286\n- 287public:\n+ 268 void populateEdgeTagVectors();\n+ 269 void populateEdgeTagsFromParentFaces();\n+ 270 void populateEdgeTagsFromParentEdges();\n+ 271\n+ 272 void populateVertexTagVectors();\n+ 273 void populateVertexTagsFromParentFaces();\n+ 274 void populateVertexTagsFromParentEdges();\n+ 275 void populateVertexTagsFromParentVertices();\n+ 276\n+ 277 //\n+ 278 // Methods (and types) involved in subdividing the topology -- though not\n+ 279 // fully exploited, any subset of the 6 relations can be generated:\n+ 280 //\n+ 281 struct Relations {\n+ 282 unsigned int _faceVertices : 1;\n+ 283 unsigned int _faceEdges : 1;\n+ 284 unsigned int _edgeVertices : 1;\n+ 285 unsigned int _edgeFaces : 1;\n+ 286 unsigned int _vertexFaces : 1;\n+ 287 unsigned int _vertexEdges : 1;\n 288\n- 289 // Debugging aides:\n- 290 enum TopologyError {\n- 291 TOPOLOGY_MISSING_EDGE_FACES=0,\n- 292 TOPOLOGY_MISSING_EDGE_VERTS,\n- 293 TOPOLOGY_MISSING_FACE_EDGES,\n- 294 TOPOLOGY_MISSING_FACE_VERTS,\n- 295 TOPOLOGY_MISSING_VERT_FACES,\n- 296 TOPOLOGY_MISSING_VERT_EDGES,\n- 297\n- 298 TOPOLOGY_FAILED_CORRELATION_EDGE_FACE,\n- 299 TOPOLOGY_FAILED_CORRELATION_FACE_VERT,\n- 300 TOPOLOGY_FAILED_CORRELATION_FACE_EDGE,\n- 301\n- 302 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_EDGE,\n- 303 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_FACE,\n- 304 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_FACES_EDGES,\n- 305\n- 306 TOPOLOGY_DEGENERATE_EDGE,\n- 307 TOPOLOGY_NON_MANIFOLD_EDGE,\n- 308\n- 309 TOPOLOGY_INVALID_CREASE_EDGE,\n- 310 TOPOLOGY_INVALID_CREASE_VERT\n- 311 };\n+ 289 void setAll(bool enable) {\n+ 290 _faceVertices = enable;\n+ 291 _faceEdges = enable;\n+ 292 _edgeVertices = enable;\n+ 293 _edgeFaces = enable;\n+ 294 _vertexFaces = enable;\n+ 295 _vertexEdges = enable;\n+ 296 }\n+ 297 };\n+ 298\n+ 299 void subdivideTopology(Relations const& relationsToSubdivide);\n+ 300\n+ 301 virtual void populateFaceVertexRelation() = 0;\n+ 302 virtual void populateFaceEdgeRelation() = 0;\n+ 303 virtual void populateEdgeVertexRelation() = 0;\n+ 304 virtual void populateEdgeFaceRelation() = 0;\n+ 305 virtual void populateVertexFaceRelation() = 0;\n+ 306 virtual void populateVertexEdgeRelation() = 0;\n+ 307\n+ 308 //\n+ 309 // Methods involved in subdividing and inspecting sharpness values:\n+ 310 //\n+ 311 void subdivideSharpnessValues();\n 312\n- 313 static char const * getTopologyErrorString(TopologyError errCode);\n- 314\n- 315 typedef void (* ValidationCallback)(TopologyError errCode, char const *\n-msg, void const * clientData);\n+ 313 void subdivideVertexSharpness();\n+ 314 void subdivideEdgeSharpness();\n+ 315 void reclassifySemisharpVertices();\n 316\n- 317 bool validateTopology(ValidationCallback callback=0, void const *\n-clientData=0) const;\n- 318\n- 319 void print(const Refinement* parentRefinement = 0) const;\n- 320\n- 321public:\n- 322 // High-level topology queries -- these may be moved elsewhere:\n- 323\n- 324 bool isSingleCreasePatch(Index face, float* sharpnessOut=NULL, int*\n-rotationOut=NULL) const;\n+ 317 //\n+ 318 // Methods involved in subdividing face-varying topology:\n+ 319 //\n+ 320 void subdivideFVarChannels();\n+ 321\n+ 322protected:\n+ 323 // A debug method of Level prints a Refinement (should really change this)\n+ 324 friend void Level::print(const Refinement *) const;\n 325\n 326 //\n- 327 // When inspecting topology, the component tags -- particularly VTag and\n-ETag -- are most\n- 328 // often inspected in groups for the face to which they belong. They are\n-designed to be\n- 329 // bitwise OR'd (the result then referred to as a \"composite\" tag) to make\n-quick decisions\n- 330 // about the face as a whole to avoid tedious topological inspection.\n- 331 //\n- 332 // The same logic can be applied to topology in a FVar channel when tags\n-specific to that\n- 333 // channel are used. Note that the VTags apply to the FVar values assigned\n-to the corners\n- 334 // of the face and not the vertex as a whole. The \"composite\" face-varying\n-VTag for a\n- 335 // vertex is the union of VTags of all distinct FVar values for that\n-vertex.\n- 336 //\n- 337 bool doesVertexFVarTopologyMatch(Index vIndex, int fvarChannel) const;\n- 338 bool doesFaceFVarTopologyMatch( Index fIndex, int fvarChannel) const;\n- 339 bool doesEdgeFVarTopologyMatch( Index eIndex, int fvarChannel) const;\n- 340\n- 341 void getFaceVTags(Index fIndex, VTag vTags[], int fvarChannel = -1) const;\n- 342 void getFaceETags(Index fIndex, ETag eTags[], int fvarChannel = -1) const;\n+ 327 // Data members -- the logical grouping of some of these (and methods that\n+make use\n+ 328 // of them) may lead to grouping them into a few utility classes or\n+structs...\n+ 329 //\n+ 330\n+ 331 // Defined on construction:\n+ 332 Level const * _parent;\n+ 333 Level * _child;\n+ 334 Sdc::Options _options;\n+ 335\n+ 336 // Defined by the subclass:\n+ 337 Sdc::Split _splitType;\n+ 338 int _regFaceSize;\n+ 339\n+ 340 // Determined by the refinement options:\n+ 341 bool _uniform;\n+ 342 bool _faceVertsFirst;\n 343\n- 344 VTag getFaceCompositeVTag(Index fIndex, int fvarChannel = -1) const;\n- 345 VTag getFaceCompositeVTag(ConstIndexArray & fVerts) const;\n- 346\n- 347 VTag getVertexCompositeFVarVTag(Index vIndex, int fvarChannel) const;\n- 348\n- 349 //\n- 350 // When gathering \"patch points\" we may want the indices of the vertices\n-or the corresponding\n- 351 // FVar values for a particular channel. Both are represented and equally\n-accessible within\n- 352 // the faces, so we allow all to be returned through these methods.\n-Setting the optional FVar\n- 353 // channel to -1 will retrieve indices of vertices instead of FVar values:\n- 354 //\n- 355 int gatherQuadLinearPatchPoints(Index fIndex, Index patchPoints[], int\n-rotation = 0,\n- 356 int fvarChannel = -1) const;\n- 357\n- 358 int gatherQuadRegularInteriorPatchPoints(Index fIndex, Index patchPoints\n-[], int rotation = 0,\n- 359 int fvarChannel = -1) const;\n- 360 int gatherQuadRegularBoundaryPatchPoints(Index fIndex, Index patchPoints\n-[], int boundaryEdgeInFace,\n- 361 int fvarChannel = -1) const;\n- 362 int gatherQuadRegularCornerPatchPoints( Index fIndex, Index patchPoints[],\n-int cornerVertInFace,\n- 363 int fvarChannel = -1) const;\n- 364\n- 365 int gatherQuadRegularRingAroundVertex(Index vIndex, Index ringPoints[],\n- 366 int fvarChannel = -1) const;\n- 367 int gatherQuadRegularPartialRingAroundVertex(Index vIndex, VSpan const &\n-span, Index ringPoints[],\n- 368 int fvarChannel = -1) const;\n- 369\n- 370 // WIP -- for future use, need to extend for face-varying...\n- 371 int gatherTriRegularInteriorPatchPoints( Index fIndex, Index patchVerts[],\n-int rotation = 0) const;\n- 372 int gatherTriRegularBoundaryVertexPatchPoints(Index fIndex, Index\n-patchVerts[], int boundaryVertInFace) const;\n- 373 int gatherTriRegularBoundaryEdgePatchPoints( Index fIndex, Index\n-patchVerts[], int boundaryEdgeInFace) const;\n- 374 int gatherTriRegularCornerVertexPatchPoints( Index fIndex, Index\n-patchVerts[], int cornerVertInFace) const;\n- 375 int gatherTriRegularCornerEdgePatchPoints( Index fIndex, Index patchVerts\n-[], int cornerEdgeInFace) const;\n+ 344 //\n+ 345 // Inventory and ordering of the types of child components:\n+ 346 //\n+ 347 int _childFaceFromFaceCount; // arguably redundant (all faces originate\n+from faces)\n+ 348 int _childEdgeFromFaceCount;\n+ 349 int _childEdgeFromEdgeCount;\n+ 350 int _childVertFromFaceCount;\n+ 351 int _childVertFromEdgeCount;\n+ 352 int _childVertFromVertCount;\n+ 353\n+ 354 int _firstChildFaceFromFace; // arguably redundant (all faces originate\n+from faces)\n+ 355 int _firstChildEdgeFromFace;\n+ 356 int _firstChildEdgeFromEdge;\n+ 357 int _firstChildVertFromFace;\n+ 358 int _firstChildVertFromEdge;\n+ 359 int _firstChildVertFromVert;\n+ 360\n+ 361 //\n+ 362 // The parent-to-child mapping:\n+ 363 // These are vectors sized according to the number of parent components\n+(and\n+ 364 // their topology) that contain references/indices to the child components\n+that\n+ 365 // result from them by refinement. When refinement is sparse, parent\n+components\n+ 366 // that have not spawned all child components will have their missing\n+children\n+ 367 // marked as invalid.\n+ 368 //\n+ 369 // NOTE the \"Array\" members here. Often vectors within the Level can be\n+shared\n+ 370 // with the Refinement, and an Array instance is used to do so. If not\n+shared\n+ 371 // the subclass just initializes the Array members after allocating its\n+own local\n+ 372 // vector members.\n+ 373 //\n+ 374 IndexArray _faceChildFaceCountsAndOffsets;\n+ 375 IndexArray _faceChildEdgeCountsAndOffsets;\n 376\n- 377public:\n- 378 // Sizing methods used to construct a level to populate:\n- 379 void resizeFaces( int numFaces);\n- 380 void resizeFaceVertices(int numFaceVertsTotal);\n- 381 void resizeFaceEdges( int numFaceEdgesTotal);\n- 382\n- 383 void resizeEdges( int numEdges);\n- 384 void resizeEdgeVertices(); // always 2*edgeCount\n- 385 void resizeEdgeFaces(int numEdgeFacesTotal);\n- 386\n- 387 void resizeVertices( int numVertices);\n- 388 void resizeVertexFaces(int numVertexFacesTotal);\n- 389 void resizeVertexEdges(int numVertexEdgesTotal);\n- 390\n- 391 void setMaxValence(int maxValence);\n+ 377 IndexVector _faceChildFaceIndices; // *cannot* always use face-vert\n+counts/offsets\n+ 378 IndexVector _faceChildEdgeIndices; // can use face-vert counts/offsets\n+ 379 IndexVector _faceChildVertIndex;\n+ 380\n+ 381 IndexVector _edgeChildEdgeIndices; // trivial/corresponding pair for each\n+ 382 IndexVector _edgeChildVertIndex;\n+ 383\n+ 384 IndexVector _vertChildVertIndex;\n+ 385\n+ 386 //\n+ 387 // The child-to-parent mapping:\n+ 388 //\n+ 389 IndexVector _childFaceParentIndex;\n+ 390 IndexVector _childEdgeParentIndex;\n+ 391 IndexVector _childVertexParentIndex;\n 392\n- 393 // Modifiers to populate the relations for each component:\n- 394 IndexArray getFaceVertices(Index faceIndex);\n- 395 IndexArray getFaceEdges(Index faceIndex);\n- 396 IndexArray getEdgeVertices(Index edgeIndex);\n- 397 IndexArray getEdgeFaces(Index edgeIndex);\n- 398 IndexArray getVertexFaces(Index vertIndex);\n- 399 IndexArray getVertexEdges(Index vertIndex);\n- 400\n- 401 LocalIndexArray getEdgeFaceLocalIndices(Index edgeIndex);\n- 402 LocalIndexArray getVertexFaceLocalIndices(Index vertIndex);\n- 403 LocalIndexArray getVertexEdgeLocalIndices(Index vertIndex);\n- 404\n- 405 // Replace these with access to sharpness buffers/arrays rather than\n-elements:\n- 406 float& getEdgeSharpness(Index edgeIndex);\n- 407 float& getVertexSharpness(Index vertIndex);\n- 408\n- 409 // Create, destroy and populate face-varying channels:\n- 410 int createFVarChannel(int fvarValueCount, Sdc::Options const& options);\n- 411 void destroyFVarChannel(int channel);\n+ 393 std::vector _childFaceTag;\n+ 394 std::vector _childEdgeTag;\n+ 395 std::vector _childVertexTag;\n+ 396\n+ 397 //\n+ 398 // Tags for sparse selection of components:\n+ 399 //\n+ 400 std::vector _parentFaceTag;\n+ 401 std::vector _parentEdgeTag;\n+ 402 std::vector _parentVertexTag;\n+ 403\n+ 404 //\n+ 405 // Refinement data for face-varying channels present in the Levels being\n+refined:\n+ 406 //\n+ 407 std::vector _fvarChannels;\n+ 408};\n+ 409\n+ 410inline ConstIndexArray\n+ 411Refinement::getFaceChildFaces(Index parentFace) const {\n 412\n- 413 IndexArray getFaceFVarValues(Index faceIndex, int channel);\n- 414\n- 415 void completeFVarChannelTopology(int channel, int regBoundaryValence);\n+ 413 return ConstIndexArray(&_faceChildFaceIndices\n+[_faceChildFaceCountsAndOffsets[2*parentFace+1]],\n+ 414 _faceChildFaceCountsAndOffsets[2*parentFace]);\n+ 415}\n 416\n- 417 // Counts and offsets for all relation types:\n- 418 // - these may be unwarranted if we let Refinement access members\n-directly...\n- 419 int getNumFaceVertices( Index faceIndex) const { return\n-_faceVertCountsAndOffsets[2*faceIndex]; }\n- 420 int getOffsetOfFaceVertices(Index faceIndex) const { return\n-_faceVertCountsAndOffsets[2*faceIndex + 1]; }\n- 421\n- 422 int getNumFaceEdges( Index faceIndex) const { return getNumFaceVertices\n-(faceIndex); }\n- 423 int getOffsetOfFaceEdges(Index faceIndex) const { return\n-getOffsetOfFaceVertices(faceIndex); }\n- 424\n- 425 int getNumEdgeVertices( Index ) const { return 2; }\n- 426 int getOffsetOfEdgeVertices(Index edgeIndex) const { return 2 * edgeIndex;\n-}\n- 427\n- 428 int getNumEdgeFaces( Index edgeIndex) const { return\n-_edgeFaceCountsAndOffsets[2*edgeIndex]; }\n- 429 int getOffsetOfEdgeFaces(Index edgeIndex) const { return\n-_edgeFaceCountsAndOffsets[2*edgeIndex + 1]; }\n- 430\n- 431 int getNumVertexFaces( Index vertIndex) const { return\n-_vertFaceCountsAndOffsets[2*vertIndex]; }\n- 432 int getOffsetOfVertexFaces(Index vertIndex) const { return\n-_vertFaceCountsAndOffsets[2*vertIndex + 1]; }\n- 433\n- 434 int getNumVertexEdges( Index vertIndex) const { return\n-_vertEdgeCountsAndOffsets[2*vertIndex]; }\n- 435 int getOffsetOfVertexEdges(Index vertIndex) const { return\n-_vertEdgeCountsAndOffsets[2*vertIndex + 1]; }\n+ 417inline IndexArray\n+ 418Refinement::getFaceChildFaces(Index parentFace) {\n+ 419\n+ 420 return IndexArray(&_faceChildFaceIndices[_faceChildFaceCountsAndOffsets\n+[2*parentFace+1]],\n+ 421 _faceChildFaceCountsAndOffsets[2*parentFace]);\n+ 422}\n+ 423\n+ 424inline ConstIndexArray\n+ 425Refinement::getFaceChildEdges(Index parentFace) const {\n+ 426\n+ 427 return ConstIndexArray(&_faceChildEdgeIndices\n+[_faceChildEdgeCountsAndOffsets[2*parentFace+1]],\n+ 428 _faceChildEdgeCountsAndOffsets[2*parentFace]);\n+ 429}\n+ 430inline IndexArray\n+ 431Refinement::getFaceChildEdges(Index parentFace) {\n+ 432\n+ 433 return IndexArray(&_faceChildEdgeIndices[_faceChildEdgeCountsAndOffsets\n+[2*parentFace+1]],\n+ 434 _faceChildEdgeCountsAndOffsets[2*parentFace]);\n+ 435}\n 436\n- 437 ConstIndexArray getFaceVertices() const;\n- 438\n- 439 //\n- 440 // Note that for some relations, the size of the relations for a child\n-component\n- 441 // can vary radically from its parent due to the sparsity of the\n-refinement. So\n- 442 // in these cases a few additional utilities are provided to help define\n-the set\n- 443 // of incident components. Assuming adequate memory has been allocated,\n-the\n- 444 // \"resize\" methods here initialize the set of incident components by\n-setting\n- 445 // both the size and the appropriate offset, while \"trim\" is use to\n-quickly lower\n- 446 // the size from an upper bound and nothing else.\n- 447 //\n- 448 void resizeFaceVertices(Index FaceIndex, int count);\n- 449\n- 450 void resizeEdgeFaces(Index edgeIndex, int count);\n- 451 void trimEdgeFaces( Index edgeIndex, int count);\n- 452\n- 453 void resizeVertexFaces(Index vertIndex, int count);\n- 454 void trimVertexFaces( Index vertIndex, int count);\n+ 437inline ConstIndexArray\n+ 438Refinement::getEdgeChildEdges(Index parentEdge) const {\n+ 439\n+ 440 return ConstIndexArray(&_edgeChildEdgeIndices[parentEdge*2], 2);\n+ 441}\n+ 442\n+ 443inline IndexArray\n+ 444Refinement::getEdgeChildEdges(Index parentEdge) {\n+ 445\n+ 446 return IndexArray(&_edgeChildEdgeIndices[parentEdge*2], 2);\n+ 447}\n+ 448\n+ 449} // end namespace internal\n+ 450} // end namespace Vtr\n+ 451\n+ 452} // end namespace OPENSUBDIV_VERSION\n+ 453using namespace OPENSUBDIV_VERSION;\n+ 454} // end namespace OpenSubdiv\n 455\n- 456 void resizeVertexEdges(Index vertIndex, int count);\n- 457 void trimVertexEdges( Index vertIndex, int count);\n- 458\n- 459public:\n- 460 //\n- 461 // Initial plans were to have a few specific classes properly construct\n-the\n- 462 // topology from scratch, e.g. the Refinement class and a Factory class\n-for\n- 463 // the base level, by populating all topological relations. The need to\n-have\n- 464 // a class construct full topology given only a simple face-vertex list,\n-made\n- 465 // it necessary to write code to define and orient all relations -- and\n-most\n- 466 // of that seemed best placed here.\n- 467 //\n- 468 bool completeTopologyFromFaceVertices();\n- 469 Index findEdge(Index v0, Index v1, ConstIndexArray v0Edges) const;\n- 470\n- 471 // Methods supporting the above:\n- 472 void orientIncidentComponents();\n- 473 bool orderVertexFacesAndEdges(Index vIndex, Index* vFaces, Index* vEdges)\n-const;\n- 474 bool orderVertexFacesAndEdges(Index vIndex);\n- 475 void populateLocalIndices();\n- 476\n- 477 IndexArray shareFaceVertCountsAndOffsets() const;\n- 478\n- 479private:\n- 480 // Refinement classes (including all subclasses) build a Level:\n- 481 friend class Refinement;\n- 482 friend class TriRefinement;\n- 483 friend class QuadRefinement;\n- 484\n- 485 //\n- 486 // A Level is independent of subdivision scheme or options. While it may\n-have been\n- 487 // affected by them in its construction, they are not associated with it -\n-- a Level\n- 488 // is pure topology and any subdivision parameters are external.\n- 489 //\n- 490\n- 491 // Simple members for inventory, etc.\n- 492 int _faceCount;\n- 493 int _edgeCount;\n- 494 int _vertCount;\n- 495\n- 496 // The \"depth\" member is clearly useful in both the topological splitting\n-and the\n- 497 // stencil queries, but arguably it ties the Level to a hierarchy which\n-counters\n- 498 // the idea of it being independent.\n- 499 int _depth;\n- 500\n- 501 // Maxima to help clients manage sizing of data buffers. Given \"max\n-valence\",\n- 502 // the \"max edge faces\" is strictly redundant as it will always be less,\n-but\n- 503 // since it will typically be so much less (i.e. 2) it is kept for now.\n- 504 int _maxEdgeFaces;\n- 505 int _maxValence;\n- 506\n- 507 //\n- 508 // Topology vectors:\n- 509 // Note that of all of these, only data for the face-edge relation is not\n- 510 // stored in the osd::FarTables in any form. The FarTable vectors combine\n- 511 // the edge-vert and edge-face relations. The eventual goal is that this\n- 512 // data be part of the osd::Far classes and be a superset of the FarTable\n- 513 // vectors, i.e. no data duplication or conversion. The fact that FarTable\n- 514 // already stores 5 of the 6 possible relations should make the topology\n- 515 // storage as a whole a non-issue.\n- 516 //\n- 517 // The vert-face-child and vert-edge-child indices are also arguably not\n- 518 // a topology relation but more one for parent/child relations. But it is\n- 519 // a topological relationship, and if named differently would not likely\n- 520 // raise this. It has been named with \"child\" in the name as it does play\n- 521 // a more significant role during subdivision in mapping between parent\n- 522 // and child components, and so has been named to reflect that more\n-clearly.\n- 523 //\n- 524\n- 525 // Per-face:\n- 526 std::vector _faceVertCountsAndOffsets; // 2 per face, redundant\n-after level 0\n- 527 std::vector _faceVertIndices; // 3 or 4 per face, variable at level\n-0\n- 528 std::vector _faceEdgeIndices; // matches face-vert indices\n- 529 std::vector _faceTags; // 1 per face: includes \"hole\" tag\n- 530\n- 531 // Per-edge:\n- 532 std::vector _edgeVertIndices; // 2 per edge\n- 533 std::vector _edgeFaceCountsAndOffsets; // 2 per edge\n- 534 std::vector _edgeFaceIndices; // varies with faces per edge\n- 535 std::vector _edgeFaceLocalIndices; // varies with faces per\n-edge\n- 536\n- 537 std::vector _edgeSharpness; // 1 per edge\n- 538 std::vector _edgeTags; // 1 per edge: manifold, boundary, etc.\n- 539\n- 540 // Per-vertex:\n- 541 std::vector _vertFaceCountsAndOffsets; // 2 per vertex\n- 542 std::vector _vertFaceIndices; // varies with valence\n- 543 std::vector _vertFaceLocalIndices; // varies with valence, 8-\n-bit for now\n- 544\n- 545 std::vector _vertEdgeCountsAndOffsets; // 2 per vertex\n- 546 std::vector _vertEdgeIndices; // varies with valence\n- 547 std::vector _vertEdgeLocalIndices; // varies with valence, 8-\n-bit for now\n- 548\n- 549 std::vector _vertSharpness; // 1 per vertex\n- 550 std::vector _vertTags; // 1 per vertex: manifold, Sdc::Rule, etc.\n- 551\n- 552 // Face-varying channels:\n- 553 std::vector _fvarChannels;\n- 554};\n- 555\n- 556//\n- 557// Access/modify the vertices incident a given face:\n- 558//\n- 559inline ConstIndexArray\n- 560Level::getFaceVertices(Index faceIndex) const {\n- 561 return ConstIndexArray(&_faceVertIndices[_faceVertCountsAndOffsets\n-[faceIndex*2+1]],\n- 562 _faceVertCountsAndOffsets[faceIndex*2]);\n- 563}\n- 564inline IndexArray\n- 565Level::getFaceVertices(Index faceIndex) {\n- 566 return IndexArray(&_faceVertIndices[_faceVertCountsAndOffsets\n-[faceIndex*2+1]],\n- 567 _faceVertCountsAndOffsets[faceIndex*2]);\n- 568}\n- 569\n- 570inline void\n- 571Level::resizeFaceVertices(Index faceIndex, int count) {\n- 572\n- 573 int* countOffsetPair = &_faceVertCountsAndOffsets[faceIndex*2];\n- 574\n- 575 countOffsetPair[0] = count;\n- 576 countOffsetPair[1] = (faceIndex == 0) ? 0 : (countOffsetPair[-2] +\n-countOffsetPair[-1]);\n- 577\n- 578 _maxValence = std::max(_maxValence, count);\n- 579}\n- 580\n- 581inline ConstIndexArray\n- 582Level::getFaceVertices() const {\n- 583 return ConstIndexArray(&_faceVertIndices[0], (int)_faceVertIndices.size\n-());\n- 584}\n- 585\n- 586//\n- 587// Access/modify the edges incident a given face:\n- 588//\n- 589inline ConstIndexArray\n- 590Level::getFaceEdges(Index faceIndex) const {\n- 591 return ConstIndexArray(&_faceEdgeIndices[_faceVertCountsAndOffsets\n-[faceIndex*2+1]],\n- 592 _faceVertCountsAndOffsets[faceIndex*2]);\n- 593}\n- 594inline IndexArray\n- 595Level::getFaceEdges(Index faceIndex) {\n- 596 return IndexArray(&_faceEdgeIndices[_faceVertCountsAndOffsets\n-[faceIndex*2+1]],\n- 597 _faceVertCountsAndOffsets[faceIndex*2]);\n- 598}\n- 599\n- 600//\n- 601// Access/modify the faces incident a given vertex:\n- 602//\n- 603inline ConstIndexArray\n- 604Level::getVertexFaces(Index vertIndex) const {\n- 605 return ConstIndexArray( (&_vertFaceIndices[0]) + _vertFaceCountsAndOffsets\n-[vertIndex*2+1],\n- 606 _vertFaceCountsAndOffsets[vertIndex*2]);\n- 607}\n- 608inline IndexArray\n- 609Level::getVertexFaces(Index vertIndex) {\n- 610 return IndexArray( (&_vertFaceIndices[0]) + _vertFaceCountsAndOffsets\n-[vertIndex*2+1],\n- 611 _vertFaceCountsAndOffsets[vertIndex*2]);\n- 612}\n- 613\n- 614inline ConstLocalIndexArray\n- 615Level::getVertexFaceLocalIndices(Index vertIndex) const {\n- 616 return ConstLocalIndexArray( (&_vertFaceLocalIndices[0]) +\n-_vertFaceCountsAndOffsets[vertIndex*2+1],\n- 617 _vertFaceCountsAndOffsets[vertIndex*2]);\n- 618}\n- 619inline LocalIndexArray\n- 620Level::getVertexFaceLocalIndices(Index vertIndex) {\n- 621 return LocalIndexArray( (&_vertFaceLocalIndices[0]) +\n-_vertFaceCountsAndOffsets[vertIndex*2+1],\n- 622 _vertFaceCountsAndOffsets[vertIndex*2]);\n- 623}\n- 624\n- 625inline void\n- 626Level::resizeVertexFaces(Index vertIndex, int count) {\n- 627 int* countOffsetPair = &_vertFaceCountsAndOffsets[vertIndex*2];\n- 628\n- 629 countOffsetPair[0] = count;\n- 630 countOffsetPair[1] = (vertIndex == 0) ? 0 : (countOffsetPair[-2] +\n-countOffsetPair[-1]);\n- 631}\n- 632inline void\n- 633Level::trimVertexFaces(Index vertIndex, int count) {\n- 634 _vertFaceCountsAndOffsets[vertIndex*2] = count;\n- 635}\n- 636\n- 637//\n- 638// Access/modify the edges incident a given vertex:\n- 639//\n- 640inline ConstIndexArray\n- 641Level::getVertexEdges(Index vertIndex) const {\n- 642 return ConstIndexArray( (&_vertEdgeIndices[0]) +_vertEdgeCountsAndOffsets\n-[vertIndex*2+1],\n- 643 _vertEdgeCountsAndOffsets[vertIndex*2]);\n- 644}\n- 645inline IndexArray\n- 646Level::getVertexEdges(Index vertIndex) {\n- 647 return IndexArray( (&_vertEdgeIndices[0]) +_vertEdgeCountsAndOffsets\n-[vertIndex*2+1],\n- 648 _vertEdgeCountsAndOffsets[vertIndex*2]);\n- 649}\n- 650\n- 651inline ConstLocalIndexArray\n- 652Level::getVertexEdgeLocalIndices(Index vertIndex) const {\n- 653 return ConstLocalIndexArray( (&_vertEdgeLocalIndices[0]) +\n-_vertEdgeCountsAndOffsets[vertIndex*2+1],\n- 654 _vertEdgeCountsAndOffsets[vertIndex*2]);\n- 655}\n- 656inline LocalIndexArray\n- 657Level::getVertexEdgeLocalIndices(Index vertIndex) {\n- 658 return LocalIndexArray( (&_vertEdgeLocalIndices[0]) +\n-_vertEdgeCountsAndOffsets[vertIndex*2+1],\n- 659 _vertEdgeCountsAndOffsets[vertIndex*2]);\n- 660}\n- 661\n- 662inline void\n- 663Level::resizeVertexEdges(Index vertIndex, int count) {\n- 664 int* countOffsetPair = &_vertEdgeCountsAndOffsets[vertIndex*2];\n- 665\n- 666 countOffsetPair[0] = count;\n- 667 countOffsetPair[1] = (vertIndex == 0) ? 0 : (countOffsetPair[-2] +\n-countOffsetPair[-1]);\n- 668\n- 669 _maxValence = std::max(_maxValence, count);\n- 670}\n- 671inline void\n- 672Level::trimVertexEdges(Index vertIndex, int count) {\n- 673 _vertEdgeCountsAndOffsets[vertIndex*2] = count;\n- 674}\n- 675\n- 676inline void\n- 677Level::setMaxValence(int valence) {\n- 678 _maxValence = valence;\n- 679}\n- 680\n- 681//\n- 682// Access/modify the vertices incident a given edge:\n- 683//\n- 684inline ConstIndexArray\n- 685Level::getEdgeVertices(Index edgeIndex) const {\n- 686 return ConstIndexArray(&_edgeVertIndices[edgeIndex*2], 2);\n- 687}\n- 688inline IndexArray\n- 689Level::getEdgeVertices(Index edgeIndex) {\n- 690 return IndexArray(&_edgeVertIndices[edgeIndex*2], 2);\n- 691}\n- 692\n- 693//\n- 694// Access/modify the faces incident a given edge:\n- 695//\n- 696inline ConstIndexArray\n- 697Level::getEdgeFaces(Index edgeIndex) const {\n- 698 return ConstIndexArray(&_edgeFaceIndices[0] +\n- 699 _edgeFaceCountsAndOffsets[edgeIndex*2+1],\n- 700 _edgeFaceCountsAndOffsets[edgeIndex*2]);\n- 701}\n- 702inline IndexArray\n- 703Level::getEdgeFaces(Index edgeIndex) {\n- 704 return IndexArray(&_edgeFaceIndices[0] +\n- 705 _edgeFaceCountsAndOffsets[edgeIndex*2+1],\n- 706 _edgeFaceCountsAndOffsets[edgeIndex*2]);\n- 707}\n- 708\n- 709inline ConstLocalIndexArray\n- 710Level::getEdgeFaceLocalIndices(Index edgeIndex) const {\n- 711 return ConstLocalIndexArray(&_edgeFaceLocalIndices[0] +\n- 712 _edgeFaceCountsAndOffsets[edgeIndex*2+1],\n- 713 _edgeFaceCountsAndOffsets[edgeIndex*2]);\n- 714}\n- 715inline LocalIndexArray\n- 716Level::getEdgeFaceLocalIndices(Index edgeIndex) {\n- 717 return LocalIndexArray(&_edgeFaceLocalIndices[0] +\n- 718 _edgeFaceCountsAndOffsets[edgeIndex*2+1],\n- 719 _edgeFaceCountsAndOffsets[edgeIndex*2]);\n- 720}\n- 721\n- 722inline void\n- 723Level::resizeEdgeFaces(Index edgeIndex, int count) {\n- 724 int* countOffsetPair = &_edgeFaceCountsAndOffsets[edgeIndex*2];\n- 725\n- 726 countOffsetPair[0] = count;\n- 727 countOffsetPair[1] = (edgeIndex == 0) ? 0 : (countOffsetPair[-2] +\n-countOffsetPair[-1]);\n- 728\n- 729 _maxEdgeFaces = std::max(_maxEdgeFaces, count);\n- 730}\n- 731inline void\n- 732Level::trimEdgeFaces(Index edgeIndex, int count) {\n- 733 _edgeFaceCountsAndOffsets[edgeIndex*2] = count;\n- 734}\n- 735\n- 736//\n- 737// Access/modify sharpness values:\n- 738//\n- 739inline float\n- 740Level::getEdgeSharpness(Index edgeIndex) const {\n- 741 return _edgeSharpness[edgeIndex];\n- 742}\n- 743inline float&\n- 744Level::getEdgeSharpness(Index edgeIndex) {\n- 745 return _edgeSharpness[edgeIndex];\n- 746}\n- 747\n- 748inline float\n- 749Level::getVertexSharpness(Index vertIndex) const {\n- 750 return _vertSharpness[vertIndex];\n- 751}\n- 752inline float&\n- 753Level::getVertexSharpness(Index vertIndex) {\n- 754 return _vertSharpness[vertIndex];\n- 755}\n- 756\n- 757inline Sdc::Crease::Rule\n- 758Level::getVertexRule(Index vertIndex) const {\n- 759 return (Sdc::Crease::Rule) _vertTags[vertIndex]._rule;\n- 760}\n- 761\n- 762//\n- 763// Access/modify hole tag:\n- 764//\n- 765inline void\n- 766Level::setFaceHole(Index faceIndex, bool b) {\n- 767 _faceTags[faceIndex]._hole = b;\n- 768}\n- 769inline bool\n- 770Level::isFaceHole(Index faceIndex) const {\n- 771 return _faceTags[faceIndex]._hole;\n- 772}\n- 773\n- 774//\n- 775// Access/modify non-manifold tags:\n- 776//\n- 777inline void\n- 778Level::setEdgeNonManifold(Index edgeIndex, bool b) {\n- 779 _edgeTags[edgeIndex]._nonManifold = b;\n- 780}\n- 781inline bool\n- 782Level::isEdgeNonManifold(Index edgeIndex) const {\n- 783 return _edgeTags[edgeIndex]._nonManifold;\n- 784}\n- 785\n- 786inline void\n- 787Level::setVertexNonManifold(Index vertIndex, bool b) {\n- 788 _vertTags[vertIndex]._nonManifold = b;\n- 789}\n- 790inline bool\n- 791Level::isVertexNonManifold(Index vertIndex) const {\n- 792 return _vertTags[vertIndex]._nonManifold;\n- 793}\n- 794\n- 795//\n- 796// Sizing methods to allocate space:\n- 797//\n- 798inline void\n- 799Level::resizeFaces(int faceCount) {\n- 800 _faceCount = faceCount;\n- 801 _faceVertCountsAndOffsets.resize(2 * faceCount);\n- 802\n- 803 _faceTags.resize(faceCount);\n- 804 std::memset((void*) &_faceTags[0], 0, _faceCount * sizeof(FTag));\n- 805}\n- 806inline void\n- 807Level::resizeFaceVertices(int totalFaceVertCount) {\n- 808 _faceVertIndices.resize(totalFaceVertCount);\n- 809}\n- 810inline void\n- 811Level::resizeFaceEdges(int totalFaceEdgeCount) {\n- 812 _faceEdgeIndices.resize(totalFaceEdgeCount);\n- 813}\n- 814\n- 815inline void\n- 816Level::resizeEdges(int edgeCount) {\n- 817\n- 818 _edgeCount = edgeCount;\n- 819 _edgeFaceCountsAndOffsets.resize(2 * edgeCount);\n- 820\n- 821 _edgeSharpness.resize(edgeCount);\n- 822 _edgeTags.resize(edgeCount);\n- 823\n- 824 if (edgeCount>0) {\n- 825 std::memset((void*) &_edgeTags[0], 0, _edgeCount * sizeof(ETag));\n- 826 }\n- 827}\n- 828inline void\n- 829Level::resizeEdgeVertices() {\n- 830\n- 831 _edgeVertIndices.resize(2 * _edgeCount);\n- 832}\n- 833inline void\n- 834Level::resizeEdgeFaces(int totalEdgeFaceCount) {\n- 835\n- 836 _edgeFaceIndices.resize(totalEdgeFaceCount);\n- 837 _edgeFaceLocalIndices.resize(totalEdgeFaceCount);\n- 838}\n- 839\n- 840inline void\n- 841Level::resizeVertices(int vertCount) {\n- 842\n- 843 _vertCount = vertCount;\n- 844 _vertFaceCountsAndOffsets.resize(2 * vertCount);\n- 845 _vertEdgeCountsAndOffsets.resize(2 * vertCount);\n- 846\n- 847 _vertSharpness.resize(vertCount);\n- 848 _vertTags.resize(vertCount);\n- 849 std::memset((void*) &_vertTags[0], 0, _vertCount * sizeof(VTag));\n- 850}\n- 851inline void\n- 852Level::resizeVertexFaces(int totalVertFaceCount) {\n- 853\n- 854 _vertFaceIndices.resize(totalVertFaceCount);\n- 855 _vertFaceLocalIndices.resize(totalVertFaceCount);\n- 856}\n- 857inline void\n- 858Level::resizeVertexEdges(int totalVertEdgeCount) {\n- 859\n- 860 _vertEdgeIndices.resize(totalVertEdgeCount);\n- 861 _vertEdgeLocalIndices.resize(totalVertEdgeCount);\n- 862}\n- 863\n- 864inline IndexArray\n- 865Level::shareFaceVertCountsAndOffsets() const {\n- 866 // XXXX manuelk we have to force const casting here (classes don't 'share'\n- 867 // members usually...)\n- 868 return IndexArray(const_cast(&_faceVertCountsAndOffsets[0]),\n- 869 (int)_faceVertCountsAndOffsets.size());\n- 870}\n- 871\n- 872} // end namespace internal\n- 873} // end namespace Vtr\n- 874\n- 875} // end namespace OPENSUBDIV_VERSION\n- 876using namespace OPENSUBDIV_VERSION;\n- 877} // end namespace OpenSubdiv\n- 878\n- 879#endif /* OPENSUBDIV3_VTR_LEVEL_H */\n+ 456#endif /* OPENSUBDIV3_VTR_REFINEMENT_H */\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstLocalIndexArray\n-ConstArray< LocalIndex > ConstLocalIndexArray\n-Definition: types.h:83\n-OpenSubdiv::OPENSUBDIV_VERSION::Vtr::LocalIndexArray\n-Array< LocalIndex > LocalIndexArray\n-Definition: types.h:82\n OpenSubdiv::OPENSUBDIV_VERSION::Vtr::Index\n int Index\n Definition: types.h:54\n+OpenSubdiv::OPENSUBDIV_VERSION::Vtr::IndexVector\n+std::vector< Index > IndexVector\n+Definition: types.h:77\n OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstIndexArray\n ConstArray< Index > ConstIndexArray\n Definition: types.h:80\n-OpenSubdiv::OPENSUBDIV_VERSION::Vtr::LocalIndex\n-unsigned short LocalIndex\n-Definition: types.h:66\n OpenSubdiv::OPENSUBDIV_VERSION::Vtr::IndexArray\n Array< Index > IndexArray\n Definition: types.h:79\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::Rule\n-Rule\n-Definition: crease.h:82\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Split\n+Split\n+Enumerated type for all face splitting schemes.\n+Definition: types.h:47\n * opensubdiv\n * vtr\n- * level.h\n+ * refinement.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00755.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00755.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/vtr/refinement.h File Reference\n+OpenSubdiv: opensubdiv/vtr/level.h File Reference\n \n \n \n \n \n \n \n@@ -82,23 +82,26 @@\n name=\"MSearchResults\" id=\"MSearchResults\">\n \n \n \n
\n
\n Namespaces
\n-
refinement.h File Reference
\n+
level.h File Reference
\n
\n
\n
#include "../version.h"
\n #include "../sdc/types.h"
\n+#include "../sdc/crease.h"
\n #include "../sdc/options.h"
\n #include "../vtr/types.h"
\n-#include "../vtr/level.h"
\n+#include <algorithm>
\n #include <vector>
\n+#include <cassert>
\n+#include <cstring>
\n
\n

Go to the source code of this file.

\n \n \n \n \n@@ -108,13 +111,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,26 +5,29 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Namespaces\n-refinement.h File Reference\n+level.h File Reference\n #include \"../version.h\"\n #include \"../sdc/types.h\"\n+#include \"../sdc/crease.h\"\n #include \"../sdc/options.h\"\n #include \"../vtr/types.h\"\n-#include \"../vtr/level.h\"\n+#include \n #include \n+#include \n+#include \n Go_to_the_source_code_of_this_file.\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Vtr\n \u00a0\n * opensubdiv\n * vtr\n- * refinement.h\n+ * level.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00755_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00755_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/vtr/refinement.h Source File\n+OpenSubdiv: opensubdiv/vtr/level.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
refinement.h
\n+
level.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2014 DreamWorks Animation LLC.
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -106,457 +106,882 @@\n
17//
\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n-
24#ifndef OPENSUBDIV3_VTR_REFINEMENT_H
\n-
25#define OPENSUBDIV3_VTR_REFINEMENT_H
\n+
24#ifndef OPENSUBDIV3_VTR_LEVEL_H
\n+
25#define OPENSUBDIV3_VTR_LEVEL_H
\n
26
\n
27#include "../version.h"
\n
28
\n
29#include "../sdc/types.h"
\n-
30#include "../sdc/options.h"
\n-
31#include "../vtr/types.h"
\n-
32#include "../vtr/level.h"
\n+
30#include "../sdc/crease.h"
\n+
31#include "../sdc/options.h"
\n+
32#include "../vtr/types.h"
\n
33
\n-
34#include <vector>
\n-
35
\n-
36//
\n-
37// Declaration for the main refinement class (Refinement) and its pre-requisites:
\n-
38//
\n-
39namespace OpenSubdiv {
\n-
40namespace OPENSUBDIV_VERSION {
\n-
41
\n-
42namespace Vtr {
\n-
43namespace internal {
\n-
44
\n-
45class FVarRefinement;
\n-
46
\n-
47//
\n-
48// Refinement:
\n-
49// A refinement is a mapping between two levels -- relating the components in the original
\n-
50// (parent) level to the one refined (child). The refinement may be complete (uniform) or sparse
\n-
51// (adaptive or otherwise selective), so not all components in the parent level will spawn
\n-
52// components in the child level.
\n-
53//
\n-
54// Refinement is an abstract class and expects subclasses corresponding to the different types
\n-
55// of topological splits that the supported subdivision schemes collectively require, i.e. those
\n-
56// listed in Sdc::SplitType. Note the virtual requirements expected of the subclasses in the list
\n-
57// of protected methods -- they differ mainly in the topology that is created in the child Level
\n-
58// and not the propagation of tags through refinement, subdivision of sharpness values or the
\n-
59// treatment of face-varying data. The primary subclasses are QuadRefinement and TriRefinement.
\n+
34#include <algorithm>
\n+
35#include <vector>
\n+
36#include <cassert>
\n+
37#include <cstring>
\n+
38
\n+
39
\n+
40namespace OpenSubdiv {
\n+
41namespace OPENSUBDIV_VERSION {
\n+
42
\n+
43namespace Vtr {
\n+
44namespace internal {
\n+
45
\n+
46class Refinement;
\n+
47class TriRefinement;
\n+
48class QuadRefinement;
\n+
49class FVarRefinement;
\n+
50class FVarLevel;
\n+
51
\n+
52//
\n+
53// Level:
\n+
54// A refinement level includes a vectorized representation of the topology
\n+
55// for a particular subdivision level. The topology is "complete" in that any
\n+
56// level can be used as the base level of another subdivision hierarchy and can
\n+
57// be considered a complete mesh independent of its ancestors. It currently
\n+
58// does contain a "depth" member -- as some inferences can then be made about
\n+
59// the topology (i.e. all quads or all tris if not level 0).
\n
60//
\n-
61// At a high level, all that is necessary in terms of interface is to construct, initialize
\n-
62// (linking the two levels), optionally select components for sparse refinement (via use of the
\n-
63// SparseSelector) and call the refine() method. This usage is expected of Far::TopologyRefiner.
\n-
64//
\n-
65// Since we really want this class to be restricted from public access eventually, all methods
\n-
66// begin with lower case (as is the convention for protected methods) and the list of friends
\n-
67// will be maintained more strictly.
\n-
68//
\n-
69class Refinement {
\n-
70
\n-
71public:
\n-
72 Refinement(Level const & parent, Level & child, Sdc::Options const& schemeOptions);
\n-
73 virtual ~Refinement();
\n-
74
\n-
75 Level const& parent() const { return *_parent; }
\n-
76 Level const& child() const { return *_child; }
\n-
77 Level& child() { return *_child; }
\n-
78
\n-
79 Sdc::Split getSplitType() const { return _splitType; }
\n-
80 int getRegularFaceSize() const { return _regFaceSize; }
\n-
81 Sdc::Options getOptions() const { return _options; }
\n+
61// This class is intended for private use within the library. There are still
\n+
62// opportunities to specialize levels -- e.g. those supporting N-sided faces vs
\n+
63// those that are purely quads or tris -- so we prefer to insulate it from public
\n+
64// access.
\n+
65//
\n+
66// The representation of topology here is to store six topological relationships
\n+
67// in tables of integers. Each is stored in its own array(s) so the result is
\n+
68// a SOA representation of the topology. The six relations are:
\n+
69//
\n+
70// - face-verts: vertices incident/comprising a face
\n+
71// - face-edges: edges incident a face
\n+
72// - edge-verts: vertices incident/comprising an edge
\n+
73// - edge-faces: faces incident an edge
\n+
74// - vert-faces: faces incident a vertex
\n+
75// - vert-edges: edges incident a vertex
\n+
76//
\n+
77// There is some redundancy here but the intent is not that this be a minimal
\n+
78// representation, the intent is that it be amenable to refinement. Classes in
\n+
79// the Far layer essentially store 5 of these 6 in a permuted form -- we add
\n+
80// the face-edges here to simplify refinement.
\n+
81//
\n
82
\n-
83 // Face-varying:
\n-
84 int getNumFVarChannels() const { return (int) _fvarChannels.size(); }
\n-
85
\n-
86 FVarRefinement const & getFVarRefinement(int c) const { return *_fvarChannels[c]; }
\n-
87
\n-
88 //
\n-
89 // Options associated with the actual refinement operation, which may end up
\n-
90 // quite involved if we want to allow for the refinement of data that is not
\n-
91 // of interest to be suppressed. For now we have:
\n-
92 //
\n-
93 // "sparse": the alternative to uniform refinement, which requires that
\n-
94 // components be previously selected/marked to be included.
\n-
95 //
\n-
96 // "minimal topology": this is one that may get broken down into a finer
\n-
97 // set of options. It suppresses "full topology" in the child level
\n-
98 // and only generates what is minimally necessary for interpolation --
\n-
99 // which requires at least the face-vertices for faces, but also the
\n-
100 // vertex-faces for any face-varying channels present. So it will
\n-
101 // generate one or two of the six possible topological relations.
\n-
102 //
\n-
103 // These are strictly controlled right now, e.g. for sparse refinement, we
\n-
104 // currently enforce full topology at the finest level to allow for subsequent
\n-
105 // patch construction.
\n-
106 //
\n-
107 struct Options {
\n-
108 Options() : _sparse(false),
\n-
109 _faceVertsFirst(false),
\n-
110 _minimalTopology(false)
\n-
111 { }
\n-
112
\n-
113 unsigned int _sparse : 1;
\n-
114 unsigned int _faceVertsFirst : 1;
\n-
115 unsigned int _minimalTopology : 1;
\n-
116
\n-
117 // Still under consideration:
\n-
118 //unsigned int _childToParentMap : 1;
\n-
119 };
\n-
120
\n-
121 void refine(Options options = Options());
\n+
83class Level {
\n+
84
\n+
85public:
\n+
86 //
\n+
87 // Simple nested types to hold the tags for each component type -- some of
\n+
88 // which are user-specified features (e.g. whether a face is a hole or not)
\n+
89 // while others indicate the topological nature of the component, how it
\n+
90 // is affected by creasing in its neighborhood, etc.
\n+
91 //
\n+
92 // Most of these properties are passed down to child components during
\n+
93 // refinement, but some -- notably the designation of a component as semi-
\n+
94 // sharp -- require re-determination as sharpness values are reduced at each
\n+
95 // level.
\n+
96 //
\n+
97 struct VTag {
\n+
98 VTag() { }
\n+
99
\n+
100 // When cleared, the VTag ALMOST represents a smooth, regular, interior
\n+
101 // vertex -- the Type enum requires a bit be explicitly set for Smooth,
\n+
102 // so that must be done explicitly if desired on initialization.
\n+
103 void clear() { std::memset((void*) this, 0, sizeof(VTag)); }
\n+
104
\n+
105 typedef unsigned short VTagSize;
\n+
106
\n+
107 VTagSize _nonManifold : 1; // fixed
\n+
108 VTagSize _xordinary : 1; // fixed
\n+
109 VTagSize _boundary : 1; // fixed
\n+
110 VTagSize _corner : 1; // fixed
\n+
111 VTagSize _infSharp : 1; // fixed
\n+
112 VTagSize _semiSharp : 1; // variable
\n+
113 VTagSize _semiSharpEdges : 1; // variable
\n+
114 VTagSize _rule : 4; // variable when _semiSharp
\n+
115
\n+
116 // These next to tags are complementary -- the "incomplete" tag is only
\n+
117 // relevant for refined levels while the "incident an irregular face" tag
\n+
118 // is only relevant for the base level. They could be combined as both
\n+
119 // indicate "no full regular ring" around a vertex
\n+
120 VTagSize _incomplete : 1; // variable only set in refined levels
\n+
121 VTagSize _incidIrregFace : 1; // variable only set in base level
\n
122
\n-
123 bool hasFaceVerticesFirst() const { return _faceVertsFirst; }
\n-
124
\n-
125public:
\n-
126 //
\n-
127 // Access to members -- some testing classes (involving vertex interpolation)
\n-
128 // currently make use of these:
\n-
129 //
\n-
130 int getNumChildFacesFromFaces() const { return _childFaceFromFaceCount; }
\n-
131 int getNumChildEdgesFromFaces() const { return _childEdgeFromFaceCount; }
\n-
132 int getNumChildEdgesFromEdges() const { return _childEdgeFromEdgeCount; }
\n-
133 int getNumChildVerticesFromFaces() const { return _childVertFromFaceCount; }
\n-
134 int getNumChildVerticesFromEdges() const { return _childVertFromEdgeCount; }
\n-
135 int getNumChildVerticesFromVertices() const { return _childVertFromVertCount; }
\n-
136
\n-
137 Index getFirstChildFaceFromFaces() const { return _firstChildFaceFromFace; }
\n-
138 Index getFirstChildEdgeFromFaces() const { return _firstChildEdgeFromFace; }
\n-
139 Index getFirstChildEdgeFromEdges() const { return _firstChildEdgeFromEdge; }
\n-
140 Index getFirstChildVertexFromFaces() const { return _firstChildVertFromFace; }
\n-
141 Index getFirstChildVertexFromEdges() const { return _firstChildVertFromEdge; }
\n-
142 Index getFirstChildVertexFromVertices() const { return _firstChildVertFromVert; }
\n-
143
\n-
144 Index getFaceChildVertex(Index f) const { return _faceChildVertIndex[f]; }
\n-
145 Index getEdgeChildVertex(Index e) const { return _edgeChildVertIndex[e]; }
\n-
146 Index getVertexChildVertex(Index v) const { return _vertChildVertIndex[v]; }
\n+
123 // Tags indicating incident infinitely-sharp (permanent) features
\n+
124 VTagSize _infSharpEdges : 1; // fixed
\n+
125 VTagSize _infSharpCrease : 1; // fixed
\n+
126 VTagSize _infIrregular : 1; // fixed
\n+
127
\n+
128 // Alternate constructor and accessor for dealing with integer bits directly:
\n+
129 explicit VTag(VTagSize bits) {
\n+
130 std::memcpy(this, &bits, sizeof(bits));
\n+
131 }
\n+
132 VTagSize getBits() const {
\n+
133 VTagSize bits;
\n+
134 std::memcpy(&bits, this, sizeof(bits));
\n+
135 return bits;
\n+
136 }
\n+
137
\n+
138 static VTag BitwiseOr(VTag const vTags[], int size = 4);
\n+
139 };
\n+
140 struct ETag {
\n+
141 ETag() { }
\n+
142
\n+
143 // When cleared, the ETag represents a smooth, manifold, interior edge
\n+
144 void clear() { std::memset((void*) this, 0, sizeof(ETag)); }
\n+
145
\n+
146 typedef unsigned char ETagSize;
\n
147
\n-
148 ConstIndexArray getFaceChildFaces(Index parentFace) const;
\n-
149 ConstIndexArray getFaceChildEdges(Index parentFace) const;
\n-
150 ConstIndexArray getEdgeChildEdges(Index parentEdge) const;
\n-
151
\n-
152 // Child-to-parent relationships
\n-
153 bool isChildVertexComplete(Index v) const { return ! _childVertexTag[v]._incomplete; }
\n-
154
\n-
155 Index getChildFaceParentFace(Index f) const { return _childFaceParentIndex[f]; }
\n-
156 int getChildFaceInParentFace(Index f) const { return _childFaceTag[f]._indexInParent; }
\n-
157
\n-
158 Index getChildEdgeParentIndex(Index e) const { return _childEdgeParentIndex[e]; }
\n-
159
\n-
160 Index getChildVertexParentIndex(Index v) const { return _childVertexParentIndex[v]; }
\n-
161
\n-
162//
\n-
163// Modifiers intended for internal/protected use:
\n-
164//
\n-
165public:
\n-
166
\n-
167 IndexArray getFaceChildFaces(Index parentFace);
\n-
168 IndexArray getFaceChildEdges(Index parentFace);
\n-
169 IndexArray getEdgeChildEdges(Index parentEdge);
\n-
170
\n-
171public:
\n-
172 //
\n-
173 // Tags have now been added per-component in Level, but there is additional need to tag
\n-
174 // components within Refinement -- we can't tag the parent level components for any
\n-
175 // refinement (in order to keep it const) and tags associated with children that are
\n-
176 // specific to the child-to-parent mapping may not be warranted in the child level.
\n-
177 //
\n-
178 // Parent tags are only required for sparse refinement. The main property to tag is
\n-
179 // whether a component was selected, and so a single SparseTag is used for all three
\n-
180 // component types. Tagging if a component is "transitional" is also useful. This may
\n-
181 // only be necessary for edges but is currently packed into a mask per-edge for faces,
\n-
182 // which could be deferred, in which case "transitional" could be a single bit.
\n-
183 //
\n-
184 // Child tags are part of the child-to-parent mapping, which consists of the parent
\n-
185 // component index for each child component, plus a tag for the child indicating more
\n-
186 // about its relationship to its parent, e.g. is it completely defined, what the parent
\n-
187 // component type is, what is the index of the child within its parent, etc.
\n+
148 ETagSize _nonManifold : 1; // fixed
\n+
149 ETagSize _boundary : 1; // fixed
\n+
150 ETagSize _infSharp : 1; // fixed
\n+
151 ETagSize _semiSharp : 1; // variable
\n+
152
\n+
153 // Alternate constructor and accessor for dealing with integer bits directly:
\n+
154 explicit ETag(ETagSize bits) {
\n+
155 std::memcpy(this, &bits, sizeof(bits));
\n+
156 }
\n+
157 ETagSize getBits() const {
\n+
158 ETagSize bits;
\n+
159 std::memcpy(&bits, this, sizeof(bits));
\n+
160 return bits;
\n+
161 }
\n+
162
\n+
163 static ETag BitwiseOr(ETag const eTags[], int size = 4);
\n+
164 };
\n+
165 struct FTag {
\n+
166 FTag() { }
\n+
167
\n+
168 void clear() { std::memset((void*) this, 0, sizeof(FTag)); }
\n+
169
\n+
170 typedef unsigned char FTagSize;
\n+
171
\n+
172 FTagSize _hole : 1; // fixed
\n+
173
\n+
174 // On deck -- coming soon...
\n+
175 //FTagSize _hasEdits : 1; // variable
\n+
176 };
\n+
177
\n+
178 // Additional simple struct to identify a "span" around a vertex, i.e. a
\n+
179 // subset of the faces around a vertex delimited by some property (e.g. a
\n+
180 // face-varying discontinuity, an inf-sharp edge, etc.)
\n+
181 //
\n+
182 // The span requires an "origin" and a "size" to fully define its extent.
\n+
183 // Use of the size is required over a leading/trailing pair as the valence
\n+
184 // around a non-manifold vertex cannot be trivially determined from two
\n+
185 // extremeties. Similarly a start face is chosen over an edge as starting
\n+
186 // with a manifold edge is ambiguous. Additional tags also support
\n+
187 // non-manifold cases, e.g. periodic spans at the apex of a double cone.
\n
188 //
\n-
189 struct SparseTag {
\n-
190 SparseTag() : _selected(0), _transitional(0) { }
\n-
191
\n-
192 unsigned char _selected : 1; // component specifically selected for refinement
\n-
193 unsigned char _transitional : 4; // adjacent to a refined component (4-bits for face)
\n-
194 };
\n+
189 // Currently setting the size to 0 or leaving the span "unassigned" is an
\n+
190 // indication to use the full neighborhood rather than a subset -- prefer
\n+
191 // use of the const method here to direct inspection of the member.
\n+
192 //
\n+
193 struct VSpan {
\n+
194 VSpan() { std::memset((void*) this, 0, sizeof(VSpan)); }
\n
195
\n-
196 struct ChildTag {
\n-
197 ChildTag() { }
\n+
196 void clear() { std::memset((void*) this, 0, sizeof(VSpan)); }
\n+
197 bool isAssigned() const { return _numFaces > 0; }
\n
198
\n-
199 unsigned char _incomplete : 1; // incomplete neighborhood to represent limit of parent
\n-
200 unsigned char _parentType : 2; // type of parent component: vertex, edge or face
\n-
201 unsigned char _indexInParent : 2; // index of child wrt parent: 0-3, or iterative if N > 4
\n-
202 };
\n-
203
\n-
204 // Methods to access and modify tags:
\n-
205 SparseTag const & getParentFaceSparseTag( Index f) const { return _parentFaceTag[f]; }
\n-
206 SparseTag const & getParentEdgeSparseTag( Index e) const { return _parentEdgeTag[e]; }
\n-
207 SparseTag const & getParentVertexSparseTag(Index v) const { return _parentVertexTag[v]; }
\n-
208
\n-
209 SparseTag & getParentFaceSparseTag( Index f) { return _parentFaceTag[f]; }
\n-
210 SparseTag & getParentEdgeSparseTag( Index e) { return _parentEdgeTag[e]; }
\n-
211 SparseTag & getParentVertexSparseTag(Index v) { return _parentVertexTag[v]; }
\n-
212
\n-
213 ChildTag const & getChildFaceTag( Index f) const { return _childFaceTag[f]; }
\n-
214 ChildTag const & getChildEdgeTag( Index e) const { return _childEdgeTag[e]; }
\n-
215 ChildTag const & getChildVertexTag(Index v) const { return _childVertexTag[v]; }
\n-
216
\n-
217 ChildTag & getChildFaceTag( Index f) { return _childFaceTag[f]; }
\n-
218 ChildTag & getChildEdgeTag( Index e) { return _childEdgeTag[e]; }
\n-
219 ChildTag & getChildVertexTag(Index v) { return _childVertexTag[v]; }
\n-
220
\n-
221// Remaining methods should really be protected -- for use by subclasses...
\n-
222public:
\n-
223 //
\n-
224 // Methods involved in constructing the parent-to-child mapping -- when the
\n-
225 // refinement is sparse, additional methods are needed to identify the selection:
\n-
226 //
\n-
227 void populateParentToChildMapping();
\n-
228 void populateParentChildIndices();
\n-
229 void printParentToChildMapping() const;
\n-
230
\n-
231 virtual void allocateParentChildIndices() = 0;
\n-
232
\n-
233 // Supporting method for sparse refinement:
\n-
234 void initializeSparseSelectionTags();
\n-
235 void markSparseChildComponentIndices();
\n-
236 void markSparseVertexChildren();
\n-
237 void markSparseEdgeChildren();
\n-
238
\n-
239 virtual void markSparseFaceChildren() = 0;
\n-
240
\n-
241 void initializeChildComponentCounts();
\n-
242
\n-
243 //
\n-
244 // Methods involved in constructing the child-to-parent mapping:
\n-
245 //
\n-
246 void populateChildToParentMapping();
\n-
247
\n-
248 void populateFaceParentVectors(ChildTag const initialChildTags[2][4]);
\n-
249 void populateFaceParentFromParentFaces(ChildTag const initialChildTags[2][4]);
\n+
199 LocalIndex _numFaces;
\n+
200 LocalIndex _startFace;
\n+
201 LocalIndex _cornerInSpan;
\n+
202
\n+
203 unsigned short _periodic : 1;
\n+
204 unsigned short _sharp : 1;
\n+
205 };
\n+
206
\n+
207public:
\n+
208 Level();
\n+
209 ~Level();
\n+
210
\n+
211 // Simple accessors:
\n+
212 int getDepth() const { return _depth; }
\n+
213
\n+
214 int getNumVertices() const { return _vertCount; }
\n+
215 int getNumFaces() const { return _faceCount; }
\n+
216 int getNumEdges() const { return _edgeCount; }
\n+
217
\n+
218 // More global sizes may prove useful...
\n+
219 int getNumFaceVerticesTotal() const { return (int) _faceVertIndices.size(); }
\n+
220 int getNumFaceEdgesTotal() const { return (int) _faceEdgeIndices.size(); }
\n+
221 int getNumEdgeVerticesTotal() const { return (int) _edgeVertIndices.size(); }
\n+
222 int getNumEdgeFacesTotal() const { return (int) _edgeFaceIndices.size(); }
\n+
223 int getNumVertexFacesTotal() const { return (int) _vertFaceIndices.size(); }
\n+
224 int getNumVertexEdgesTotal() const { return (int) _vertEdgeIndices.size(); }
\n+
225
\n+
226 int getMaxValence() const { return _maxValence; }
\n+
227 int getMaxEdgeFaces() const { return _maxEdgeFaces; }
\n+
228
\n+
229 // Methods to access the relation tables/indices -- note that for some relations
\n+
230 // (i.e. those where a component is "contained by" a neighbor, or more generally
\n+
231 // when the neighbor is a simplex of higher dimension) we store an additional
\n+
232 // "local index", e.g. for the case of vert-faces if one of the faces F[i] is
\n+
233 // incident a vertex V, then L[i] is the "local index" in F[i] of vertex V.
\n+
234 // Once have only quads (or tris), this local index need only occupy two bits
\n+
235 // and could conceivably be packed into the same integer as the face index, but
\n+
236 // for now, given the need to support faces of potentially high valence we'll
\n+
237 // use an 8- or 16-bit integer.
\n+
238 //
\n+
239 // Methods to access the six topological relations:
\n+
240 ConstIndexArray getFaceVertices(Index faceIndex) const;
\n+
241 ConstIndexArray getFaceEdges(Index faceIndex) const;
\n+
242 ConstIndexArray getEdgeVertices(Index edgeIndex) const;
\n+
243 ConstIndexArray getEdgeFaces(Index edgeIndex) const;
\n+
244 ConstIndexArray getVertexFaces(Index vertIndex) const;
\n+
245 ConstIndexArray getVertexEdges(Index vertIndex) const;
\n+
246
\n+
247 ConstLocalIndexArray getEdgeFaceLocalIndices(Index edgeIndex) const;
\n+
248 ConstLocalIndexArray getVertexFaceLocalIndices(Index vertIndex) const;
\n+
249 ConstLocalIndexArray getVertexEdgeLocalIndices(Index vertIndex) const;
\n
250
\n-
251 void populateEdgeParentVectors(ChildTag const initialChildTags[2][4]);
\n-
252 void populateEdgeParentFromParentFaces(ChildTag const initialChildTags[2][4]);
\n-
253 void populateEdgeParentFromParentEdges(ChildTag const initialChildTags[2][4]);
\n-
254
\n-
255 void populateVertexParentVectors(ChildTag const initialChildTags[2][4]);
\n-
256 void populateVertexParentFromParentFaces(ChildTag const initialChildTags[2][4]);
\n-
257 void populateVertexParentFromParentEdges(ChildTag const initialChildTags[2][4]);
\n-
258 void populateVertexParentFromParentVertices(ChildTag const initialChildTags[2][4]);
\n-
259
\n-
260 //
\n-
261 // Methods involved in propagating component tags from parent to child:
\n-
262 //
\n-
263 void propagateComponentTags();
\n-
264
\n-
265 void populateFaceTagVectors();
\n-
266 void populateFaceTagsFromParentFaces();
\n+
251 // Replace these with access to sharpness buffers/arrays rather than elements:
\n+
252 float getEdgeSharpness(Index edgeIndex) const;
\n+
253 float getVertexSharpness(Index vertIndex) const;
\n+
254 Sdc::Crease::Rule getVertexRule(Index vertIndex) const;
\n+
255
\n+
256 Index findEdge(Index v0Index, Index v1Index) const;
\n+
257
\n+
258 // Holes
\n+
259 void setFaceHole(Index faceIndex, bool b);
\n+
260 bool isFaceHole(Index faceIndex) const;
\n+
261
\n+
262 // Face-varying
\n+
263 Sdc::Options getFVarOptions(int channel) const;
\n+
264 int getNumFVarChannels() const { return (int) _fvarChannels.size(); }
\n+
265 int getNumFVarValues(int channel) const;
\n+
266 ConstIndexArray getFaceFVarValues(Index faceIndex, int channel) const;
\n
267
\n-
268 void populateEdgeTagVectors();
\n-
269 void populateEdgeTagsFromParentFaces();
\n-
270 void populateEdgeTagsFromParentEdges();
\n-
271
\n-
272 void populateVertexTagVectors();
\n-
273 void populateVertexTagsFromParentFaces();
\n-
274 void populateVertexTagsFromParentEdges();
\n-
275 void populateVertexTagsFromParentVertices();
\n-
276
\n-
277 //
\n-
278 // Methods (and types) involved in subdividing the topology -- though not
\n-
279 // fully exploited, any subset of the 6 relations can be generated:
\n-
280 //
\n-
281 struct Relations {
\n-
282 unsigned int _faceVertices : 1;
\n-
283 unsigned int _faceEdges : 1;
\n-
284 unsigned int _edgeVertices : 1;
\n-
285 unsigned int _edgeFaces : 1;
\n-
286 unsigned int _vertexFaces : 1;
\n-
287 unsigned int _vertexEdges : 1;
\n+
268 FVarLevel & getFVarLevel(int channel) { return *_fvarChannels[channel]; }
\n+
269 FVarLevel const & getFVarLevel(int channel) const { return *_fvarChannels[channel]; }
\n+
270
\n+
271 // Manifold/non-manifold tags:
\n+
272 void setEdgeNonManifold(Index edgeIndex, bool b);
\n+
273 bool isEdgeNonManifold(Index edgeIndex) const;
\n+
274
\n+
275 void setVertexNonManifold(Index vertIndex, bool b);
\n+
276 bool isVertexNonManifold(Index vertIndex) const;
\n+
277
\n+
278 // General access to all component tags:
\n+
279 VTag const & getVertexTag(Index vertIndex) const { return _vertTags[vertIndex]; }
\n+
280 ETag const & getEdgeTag(Index edgeIndex) const { return _edgeTags[edgeIndex]; }
\n+
281 FTag const & getFaceTag(Index faceIndex) const { return _faceTags[faceIndex]; }
\n+
282
\n+
283 VTag & getVertexTag(Index vertIndex) { return _vertTags[vertIndex]; }
\n+
284 ETag & getEdgeTag(Index edgeIndex) { return _edgeTags[edgeIndex]; }
\n+
285 FTag & getFaceTag(Index faceIndex) { return _faceTags[faceIndex]; }
\n+
286
\n+
287public:
\n
288
\n-
289 void setAll(bool enable) {
\n-
290 _faceVertices = enable;
\n-
291 _faceEdges = enable;
\n-
292 _edgeVertices = enable;
\n-
293 _edgeFaces = enable;
\n-
294 _vertexFaces = enable;
\n-
295 _vertexEdges = enable;
\n-
296 }
\n-
297 };
\n-
298
\n-
299 void subdivideTopology(Relations const& relationsToSubdivide);
\n-
300
\n-
301 virtual void populateFaceVertexRelation() = 0;
\n-
302 virtual void populateFaceEdgeRelation() = 0;
\n-
303 virtual void populateEdgeVertexRelation() = 0;
\n-
304 virtual void populateEdgeFaceRelation() = 0;
\n-
305 virtual void populateVertexFaceRelation() = 0;
\n-
306 virtual void populateVertexEdgeRelation() = 0;
\n-
307
\n-
308 //
\n-
309 // Methods involved in subdividing and inspecting sharpness values:
\n-
310 //
\n-
311 void subdivideSharpnessValues();
\n+
289 // Debugging aides:
\n+
290 enum TopologyError {
\n+
291 TOPOLOGY_MISSING_EDGE_FACES=0,
\n+
292 TOPOLOGY_MISSING_EDGE_VERTS,
\n+
293 TOPOLOGY_MISSING_FACE_EDGES,
\n+
294 TOPOLOGY_MISSING_FACE_VERTS,
\n+
295 TOPOLOGY_MISSING_VERT_FACES,
\n+
296 TOPOLOGY_MISSING_VERT_EDGES,
\n+
297
\n+
298 TOPOLOGY_FAILED_CORRELATION_EDGE_FACE,
\n+
299 TOPOLOGY_FAILED_CORRELATION_FACE_VERT,
\n+
300 TOPOLOGY_FAILED_CORRELATION_FACE_EDGE,
\n+
301
\n+
302 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_EDGE,
\n+
303 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_FACE,
\n+
304 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_FACES_EDGES,
\n+
305
\n+
306 TOPOLOGY_DEGENERATE_EDGE,
\n+
307 TOPOLOGY_NON_MANIFOLD_EDGE,
\n+
308
\n+
309 TOPOLOGY_INVALID_CREASE_EDGE,
\n+
310 TOPOLOGY_INVALID_CREASE_VERT
\n+
311 };
\n
312
\n-
313 void subdivideVertexSharpness();
\n-
314 void subdivideEdgeSharpness();
\n-
315 void reclassifySemisharpVertices();
\n+
313 static char const * getTopologyErrorString(TopologyError errCode);
\n+
314
\n+
315 typedef void (* ValidationCallback)(TopologyError errCode, char const * msg, void const * clientData);
\n
316
\n-
317 //
\n-
318 // Methods involved in subdividing face-varying topology:
\n-
319 //
\n-
320 void subdivideFVarChannels();
\n-
321
\n-
322protected:
\n-
323 // A debug method of Level prints a Refinement (should really change this)
\n-
324 friend void Level::print(const Refinement *) const;
\n+
317 bool validateTopology(ValidationCallback callback=0, void const * clientData=0) const;
\n+
318
\n+
319 void print(const Refinement* parentRefinement = 0) const;
\n+
320
\n+
321public:
\n+
322 // High-level topology queries -- these may be moved elsewhere:
\n+
323
\n+
324 bool isSingleCreasePatch(Index face, float* sharpnessOut=NULL, int* rotationOut=NULL) const;
\n
325
\n
326 //
\n-
327 // Data members -- the logical grouping of some of these (and methods that make use
\n-
328 // of them) may lead to grouping them into a few utility classes or structs...
\n-
329 //
\n-
330
\n-
331 // Defined on construction:
\n-
332 Level const * _parent;
\n-
333 Level * _child;
\n-
334 Sdc::Options _options;
\n-
335
\n-
336 // Defined by the subclass:
\n-
337 Sdc::Split _splitType;
\n-
338 int _regFaceSize;
\n-
339
\n-
340 // Determined by the refinement options:
\n-
341 bool _uniform;
\n-
342 bool _faceVertsFirst;
\n+
327 // When inspecting topology, the component tags -- particularly VTag and ETag -- are most
\n+
328 // often inspected in groups for the face to which they belong. They are designed to be
\n+
329 // bitwise OR'd (the result then referred to as a "composite" tag) to make quick decisions
\n+
330 // about the face as a whole to avoid tedious topological inspection.
\n+
331 //
\n+
332 // The same logic can be applied to topology in a FVar channel when tags specific to that
\n+
333 // channel are used. Note that the VTags apply to the FVar values assigned to the corners
\n+
334 // of the face and not the vertex as a whole. The "composite" face-varying VTag for a
\n+
335 // vertex is the union of VTags of all distinct FVar values for that vertex.
\n+
336 //
\n+
337 bool doesVertexFVarTopologyMatch(Index vIndex, int fvarChannel) const;
\n+
338 bool doesFaceFVarTopologyMatch( Index fIndex, int fvarChannel) const;
\n+
339 bool doesEdgeFVarTopologyMatch( Index eIndex, int fvarChannel) const;
\n+
340
\n+
341 void getFaceVTags(Index fIndex, VTag vTags[], int fvarChannel = -1) const;
\n+
342 void getFaceETags(Index fIndex, ETag eTags[], int fvarChannel = -1) const;
\n
343
\n-
344 //
\n-
345 // Inventory and ordering of the types of child components:
\n-
346 //
\n-
347 int _childFaceFromFaceCount; // arguably redundant (all faces originate from faces)
\n-
348 int _childEdgeFromFaceCount;
\n-
349 int _childEdgeFromEdgeCount;
\n-
350 int _childVertFromFaceCount;
\n-
351 int _childVertFromEdgeCount;
\n-
352 int _childVertFromVertCount;
\n-
353
\n-
354 int _firstChildFaceFromFace; // arguably redundant (all faces originate from faces)
\n-
355 int _firstChildEdgeFromFace;
\n-
356 int _firstChildEdgeFromEdge;
\n-
357 int _firstChildVertFromFace;
\n-
358 int _firstChildVertFromEdge;
\n-
359 int _firstChildVertFromVert;
\n-
360
\n-
361 //
\n-
362 // The parent-to-child mapping:
\n-
363 // These are vectors sized according to the number of parent components (and
\n-
364 // their topology) that contain references/indices to the child components that
\n-
365 // result from them by refinement. When refinement is sparse, parent components
\n-
366 // that have not spawned all child components will have their missing children
\n-
367 // marked as invalid.
\n-
368 //
\n-
369 // NOTE the "Array" members here. Often vectors within the Level can be shared
\n-
370 // with the Refinement, and an Array instance is used to do so. If not shared
\n-
371 // the subclass just initializes the Array members after allocating its own local
\n-
372 // vector members.
\n-
373 //
\n-
374 IndexArray _faceChildFaceCountsAndOffsets;
\n-
375 IndexArray _faceChildEdgeCountsAndOffsets;
\n+
344 VTag getFaceCompositeVTag(Index fIndex, int fvarChannel = -1) const;
\n+
345 VTag getFaceCompositeVTag(ConstIndexArray & fVerts) const;
\n+
346
\n+
347 VTag getVertexCompositeFVarVTag(Index vIndex, int fvarChannel) const;
\n+
348
\n+
349 //
\n+
350 // When gathering "patch points" we may want the indices of the vertices or the corresponding
\n+
351 // FVar values for a particular channel. Both are represented and equally accessible within
\n+
352 // the faces, so we allow all to be returned through these methods. Setting the optional FVar
\n+
353 // channel to -1 will retrieve indices of vertices instead of FVar values:
\n+
354 //
\n+
355 int gatherQuadLinearPatchPoints(Index fIndex, Index patchPoints[], int rotation = 0,
\n+
356 int fvarChannel = -1) const;
\n+
357
\n+
358 int gatherQuadRegularInteriorPatchPoints(Index fIndex, Index patchPoints[], int rotation = 0,
\n+
359 int fvarChannel = -1) const;
\n+
360 int gatherQuadRegularBoundaryPatchPoints(Index fIndex, Index patchPoints[], int boundaryEdgeInFace,
\n+
361 int fvarChannel = -1) const;
\n+
362 int gatherQuadRegularCornerPatchPoints( Index fIndex, Index patchPoints[], int cornerVertInFace,
\n+
363 int fvarChannel = -1) const;
\n+
364
\n+
365 int gatherQuadRegularRingAroundVertex(Index vIndex, Index ringPoints[],
\n+
366 int fvarChannel = -1) const;
\n+
367 int gatherQuadRegularPartialRingAroundVertex(Index vIndex, VSpan const & span, Index ringPoints[],
\n+
368 int fvarChannel = -1) const;
\n+
369
\n+
370 // WIP -- for future use, need to extend for face-varying...
\n+
371 int gatherTriRegularInteriorPatchPoints( Index fIndex, Index patchVerts[], int rotation = 0) const;
\n+
372 int gatherTriRegularBoundaryVertexPatchPoints(Index fIndex, Index patchVerts[], int boundaryVertInFace) const;
\n+
373 int gatherTriRegularBoundaryEdgePatchPoints( Index fIndex, Index patchVerts[], int boundaryEdgeInFace) const;
\n+
374 int gatherTriRegularCornerVertexPatchPoints( Index fIndex, Index patchVerts[], int cornerVertInFace) const;
\n+
375 int gatherTriRegularCornerEdgePatchPoints( Index fIndex, Index patchVerts[], int cornerEdgeInFace) const;
\n
376
\n-
377 IndexVector _faceChildFaceIndices; // *cannot* always use face-vert counts/offsets
\n-
378 IndexVector _faceChildEdgeIndices; // can use face-vert counts/offsets
\n-
379 IndexVector _faceChildVertIndex;
\n-
380
\n-
381 IndexVector _edgeChildEdgeIndices; // trivial/corresponding pair for each
\n-
382 IndexVector _edgeChildVertIndex;
\n-
383
\n-
384 IndexVector _vertChildVertIndex;
\n-
385
\n-
386 //
\n-
387 // The child-to-parent mapping:
\n-
388 //
\n-
389 IndexVector _childFaceParentIndex;
\n-
390 IndexVector _childEdgeParentIndex;
\n-
391 IndexVector _childVertexParentIndex;
\n+
377public:
\n+
378 // Sizing methods used to construct a level to populate:
\n+
379 void resizeFaces( int numFaces);
\n+
380 void resizeFaceVertices(int numFaceVertsTotal);
\n+
381 void resizeFaceEdges( int numFaceEdgesTotal);
\n+
382
\n+
383 void resizeEdges( int numEdges);
\n+
384 void resizeEdgeVertices(); // always 2*edgeCount
\n+
385 void resizeEdgeFaces(int numEdgeFacesTotal);
\n+
386
\n+
387 void resizeVertices( int numVertices);
\n+
388 void resizeVertexFaces(int numVertexFacesTotal);
\n+
389 void resizeVertexEdges(int numVertexEdgesTotal);
\n+
390
\n+
391 void setMaxValence(int maxValence);
\n
392
\n-
393 std::vector<ChildTag> _childFaceTag;
\n-
394 std::vector<ChildTag> _childEdgeTag;
\n-
395 std::vector<ChildTag> _childVertexTag;
\n-
396
\n-
397 //
\n-
398 // Tags for sparse selection of components:
\n-
399 //
\n-
400 std::vector<SparseTag> _parentFaceTag;
\n-
401 std::vector<SparseTag> _parentEdgeTag;
\n-
402 std::vector<SparseTag> _parentVertexTag;
\n-
403
\n-
404 //
\n-
405 // Refinement data for face-varying channels present in the Levels being refined:
\n-
406 //
\n-
407 std::vector<FVarRefinement*> _fvarChannels;
\n-
408};
\n-
409
\n-
410inline ConstIndexArray
\n-
411Refinement::getFaceChildFaces(Index parentFace) const {
\n+
393 // Modifiers to populate the relations for each component:
\n+
394 IndexArray getFaceVertices(Index faceIndex);
\n+
395 IndexArray getFaceEdges(Index faceIndex);
\n+
396 IndexArray getEdgeVertices(Index edgeIndex);
\n+
397 IndexArray getEdgeFaces(Index edgeIndex);
\n+
398 IndexArray getVertexFaces(Index vertIndex);
\n+
399 IndexArray getVertexEdges(Index vertIndex);
\n+
400
\n+
401 LocalIndexArray getEdgeFaceLocalIndices(Index edgeIndex);
\n+
402 LocalIndexArray getVertexFaceLocalIndices(Index vertIndex);
\n+
403 LocalIndexArray getVertexEdgeLocalIndices(Index vertIndex);
\n+
404
\n+
405 // Replace these with access to sharpness buffers/arrays rather than elements:
\n+
406 float& getEdgeSharpness(Index edgeIndex);
\n+
407 float& getVertexSharpness(Index vertIndex);
\n+
408
\n+
409 // Create, destroy and populate face-varying channels:
\n+
410 int createFVarChannel(int fvarValueCount, Sdc::Options const& options);
\n+
411 void destroyFVarChannel(int channel);
\n
412
\n-
413 return ConstIndexArray(&_faceChildFaceIndices[_faceChildFaceCountsAndOffsets[2*parentFace+1]],
\n-
414 _faceChildFaceCountsAndOffsets[2*parentFace]);
\n-
415}
\n+
413 IndexArray getFaceFVarValues(Index faceIndex, int channel);
\n+
414
\n+
415 void completeFVarChannelTopology(int channel, int regBoundaryValence);
\n
416
\n-
417inline IndexArray
\n-
418Refinement::getFaceChildFaces(Index parentFace) {
\n-
419
\n-
420 return IndexArray(&_faceChildFaceIndices[_faceChildFaceCountsAndOffsets[2*parentFace+1]],
\n-
421 _faceChildFaceCountsAndOffsets[2*parentFace]);
\n-
422}
\n-
423
\n-
424inline ConstIndexArray
\n-
425Refinement::getFaceChildEdges(Index parentFace) const {
\n-
426
\n-
427 return ConstIndexArray(&_faceChildEdgeIndices[_faceChildEdgeCountsAndOffsets[2*parentFace+1]],
\n-
428 _faceChildEdgeCountsAndOffsets[2*parentFace]);
\n-
429}
\n-
430inline IndexArray
\n-
431Refinement::getFaceChildEdges(Index parentFace) {
\n-
432
\n-
433 return IndexArray(&_faceChildEdgeIndices[_faceChildEdgeCountsAndOffsets[2*parentFace+1]],
\n-
434 _faceChildEdgeCountsAndOffsets[2*parentFace]);
\n-
435}
\n+
417 // Counts and offsets for all relation types:
\n+
418 // - these may be unwarranted if we let Refinement access members directly...
\n+
419 int getNumFaceVertices( Index faceIndex) const { return _faceVertCountsAndOffsets[2*faceIndex]; }
\n+
420 int getOffsetOfFaceVertices(Index faceIndex) const { return _faceVertCountsAndOffsets[2*faceIndex + 1]; }
\n+
421
\n+
422 int getNumFaceEdges( Index faceIndex) const { return getNumFaceVertices(faceIndex); }
\n+
423 int getOffsetOfFaceEdges(Index faceIndex) const { return getOffsetOfFaceVertices(faceIndex); }
\n+
424
\n+
425 int getNumEdgeVertices( Index ) const { return 2; }
\n+
426 int getOffsetOfEdgeVertices(Index edgeIndex) const { return 2 * edgeIndex; }
\n+
427
\n+
428 int getNumEdgeFaces( Index edgeIndex) const { return _edgeFaceCountsAndOffsets[2*edgeIndex]; }
\n+
429 int getOffsetOfEdgeFaces(Index edgeIndex) const { return _edgeFaceCountsAndOffsets[2*edgeIndex + 1]; }
\n+
430
\n+
431 int getNumVertexFaces( Index vertIndex) const { return _vertFaceCountsAndOffsets[2*vertIndex]; }
\n+
432 int getOffsetOfVertexFaces(Index vertIndex) const { return _vertFaceCountsAndOffsets[2*vertIndex + 1]; }
\n+
433
\n+
434 int getNumVertexEdges( Index vertIndex) const { return _vertEdgeCountsAndOffsets[2*vertIndex]; }
\n+
435 int getOffsetOfVertexEdges(Index vertIndex) const { return _vertEdgeCountsAndOffsets[2*vertIndex + 1]; }
\n
436
\n-
437inline ConstIndexArray
\n-
438Refinement::getEdgeChildEdges(Index parentEdge) const {
\n-
439
\n-
440 return ConstIndexArray(&_edgeChildEdgeIndices[parentEdge*2], 2);
\n-
441}
\n-
442
\n-
443inline IndexArray
\n-
444Refinement::getEdgeChildEdges(Index parentEdge) {
\n-
445
\n-
446 return IndexArray(&_edgeChildEdgeIndices[parentEdge*2], 2);
\n-
447}
\n-
448
\n-
449} // end namespace internal
\n-
450} // end namespace Vtr
\n-
451
\n-
452} // end namespace OPENSUBDIV_VERSION
\n-
453using namespace OPENSUBDIV_VERSION;
\n-
454} // end namespace OpenSubdiv
\n+
437 ConstIndexArray getFaceVertices() const;
\n+
438
\n+
439 //
\n+
440 // Note that for some relations, the size of the relations for a child component
\n+
441 // can vary radically from its parent due to the sparsity of the refinement. So
\n+
442 // in these cases a few additional utilities are provided to help define the set
\n+
443 // of incident components. Assuming adequate memory has been allocated, the
\n+
444 // "resize" methods here initialize the set of incident components by setting
\n+
445 // both the size and the appropriate offset, while "trim" is use to quickly lower
\n+
446 // the size from an upper bound and nothing else.
\n+
447 //
\n+
448 void resizeFaceVertices(Index FaceIndex, int count);
\n+
449
\n+
450 void resizeEdgeFaces(Index edgeIndex, int count);
\n+
451 void trimEdgeFaces( Index edgeIndex, int count);
\n+
452
\n+
453 void resizeVertexFaces(Index vertIndex, int count);
\n+
454 void trimVertexFaces( Index vertIndex, int count);
\n
455
\n-
456#endif /* OPENSUBDIV3_VTR_REFINEMENT_H */
\n+
456 void resizeVertexEdges(Index vertIndex, int count);
\n+
457 void trimVertexEdges( Index vertIndex, int count);
\n+
458
\n+
459public:
\n+
460 //
\n+
461 // Initial plans were to have a few specific classes properly construct the
\n+
462 // topology from scratch, e.g. the Refinement class and a Factory class for
\n+
463 // the base level, by populating all topological relations. The need to have
\n+
464 // a class construct full topology given only a simple face-vertex list, made
\n+
465 // it necessary to write code to define and orient all relations -- and most
\n+
466 // of that seemed best placed here.
\n+
467 //
\n+
468 bool completeTopologyFromFaceVertices();
\n+
469 Index findEdge(Index v0, Index v1, ConstIndexArray v0Edges) const;
\n+
470
\n+
471 // Methods supporting the above:
\n+
472 void orientIncidentComponents();
\n+
473 bool orderVertexFacesAndEdges(Index vIndex, Index* vFaces, Index* vEdges) const;
\n+
474 bool orderVertexFacesAndEdges(Index vIndex);
\n+
475 void populateLocalIndices();
\n+
476
\n+
477 IndexArray shareFaceVertCountsAndOffsets() const;
\n+
478
\n+
479private:
\n+
480 // Refinement classes (including all subclasses) build a Level:
\n+
481 friend class Refinement;
\n+
482 friend class TriRefinement;
\n+
483 friend class QuadRefinement;
\n+
484
\n+
485 //
\n+
486 // A Level is independent of subdivision scheme or options. While it may have been
\n+
487 // affected by them in its construction, they are not associated with it -- a Level
\n+
488 // is pure topology and any subdivision parameters are external.
\n+
489 //
\n+
490
\n+
491 // Simple members for inventory, etc.
\n+
492 int _faceCount;
\n+
493 int _edgeCount;
\n+
494 int _vertCount;
\n+
495
\n+
496 // The "depth" member is clearly useful in both the topological splitting and the
\n+
497 // stencil queries, but arguably it ties the Level to a hierarchy which counters
\n+
498 // the idea of it being independent.
\n+
499 int _depth;
\n+
500
\n+
501 // Maxima to help clients manage sizing of data buffers. Given "max valence",
\n+
502 // the "max edge faces" is strictly redundant as it will always be less, but
\n+
503 // since it will typically be so much less (i.e. 2) it is kept for now.
\n+
504 int _maxEdgeFaces;
\n+
505 int _maxValence;
\n+
506
\n+
507 //
\n+
508 // Topology vectors:
\n+
509 // Note that of all of these, only data for the face-edge relation is not
\n+
510 // stored in the osd::FarTables in any form. The FarTable vectors combine
\n+
511 // the edge-vert and edge-face relations. The eventual goal is that this
\n+
512 // data be part of the osd::Far classes and be a superset of the FarTable
\n+
513 // vectors, i.e. no data duplication or conversion. The fact that FarTable
\n+
514 // already stores 5 of the 6 possible relations should make the topology
\n+
515 // storage as a whole a non-issue.
\n+
516 //
\n+
517 // The vert-face-child and vert-edge-child indices are also arguably not
\n+
518 // a topology relation but more one for parent/child relations. But it is
\n+
519 // a topological relationship, and if named differently would not likely
\n+
520 // raise this. It has been named with "child" in the name as it does play
\n+
521 // a more significant role during subdivision in mapping between parent
\n+
522 // and child components, and so has been named to reflect that more clearly.
\n+
523 //
\n+
524
\n+
525 // Per-face:
\n+
526 std::vector<Index> _faceVertCountsAndOffsets; // 2 per face, redundant after level 0
\n+
527 std::vector<Index> _faceVertIndices; // 3 or 4 per face, variable at level 0
\n+
528 std::vector<Index> _faceEdgeIndices; // matches face-vert indices
\n+
529 std::vector<FTag> _faceTags; // 1 per face: includes "hole" tag
\n+
530
\n+
531 // Per-edge:
\n+
532 std::vector<Index> _edgeVertIndices; // 2 per edge
\n+
533 std::vector<Index> _edgeFaceCountsAndOffsets; // 2 per edge
\n+
534 std::vector<Index> _edgeFaceIndices; // varies with faces per edge
\n+
535 std::vector<LocalIndex> _edgeFaceLocalIndices; // varies with faces per edge
\n+
536
\n+
537 std::vector<float> _edgeSharpness; // 1 per edge
\n+
538 std::vector<ETag> _edgeTags; // 1 per edge: manifold, boundary, etc.
\n+
539
\n+
540 // Per-vertex:
\n+
541 std::vector<Index> _vertFaceCountsAndOffsets; // 2 per vertex
\n+
542 std::vector<Index> _vertFaceIndices; // varies with valence
\n+
543 std::vector<LocalIndex> _vertFaceLocalIndices; // varies with valence, 8-bit for now
\n+
544
\n+
545 std::vector<Index> _vertEdgeCountsAndOffsets; // 2 per vertex
\n+
546 std::vector<Index> _vertEdgeIndices; // varies with valence
\n+
547 std::vector<LocalIndex> _vertEdgeLocalIndices; // varies with valence, 8-bit for now
\n+
548
\n+
549 std::vector<float> _vertSharpness; // 1 per vertex
\n+
550 std::vector<VTag> _vertTags; // 1 per vertex: manifold, Sdc::Rule, etc.
\n+
551
\n+
552 // Face-varying channels:
\n+
553 std::vector<FVarLevel*> _fvarChannels;
\n+
554};
\n+
555
\n+
556//
\n+
557// Access/modify the vertices incident a given face:
\n+
558//
\n+
559inline ConstIndexArray
\n+
560Level::getFaceVertices(Index faceIndex) const {
\n+
561 return ConstIndexArray(&_faceVertIndices[_faceVertCountsAndOffsets[faceIndex*2+1]],
\n+
562 _faceVertCountsAndOffsets[faceIndex*2]);
\n+
563}
\n+
564inline IndexArray
\n+
565Level::getFaceVertices(Index faceIndex) {
\n+
566 return IndexArray(&_faceVertIndices[_faceVertCountsAndOffsets[faceIndex*2+1]],
\n+
567 _faceVertCountsAndOffsets[faceIndex*2]);
\n+
568}
\n+
569
\n+
570inline void
\n+
571Level::resizeFaceVertices(Index faceIndex, int count) {
\n+
572
\n+
573 int* countOffsetPair = &_faceVertCountsAndOffsets[faceIndex*2];
\n+
574
\n+
575 countOffsetPair[0] = count;
\n+
576 countOffsetPair[1] = (faceIndex == 0) ? 0 : (countOffsetPair[-2] + countOffsetPair[-1]);
\n+
577
\n+
578 _maxValence = std::max(_maxValence, count);
\n+
579}
\n+
580
\n+
581inline ConstIndexArray
\n+
582Level::getFaceVertices() const {
\n+
583 return ConstIndexArray(&_faceVertIndices[0], (int)_faceVertIndices.size());
\n+
584}
\n+
585
\n+
586//
\n+
587// Access/modify the edges incident a given face:
\n+
588//
\n+
589inline ConstIndexArray
\n+
590Level::getFaceEdges(Index faceIndex) const {
\n+
591 return ConstIndexArray(&_faceEdgeIndices[_faceVertCountsAndOffsets[faceIndex*2+1]],
\n+
592 _faceVertCountsAndOffsets[faceIndex*2]);
\n+
593}
\n+
594inline IndexArray
\n+
595Level::getFaceEdges(Index faceIndex) {
\n+
596 return IndexArray(&_faceEdgeIndices[_faceVertCountsAndOffsets[faceIndex*2+1]],
\n+
597 _faceVertCountsAndOffsets[faceIndex*2]);
\n+
598}
\n+
599
\n+
600//
\n+
601// Access/modify the faces incident a given vertex:
\n+
602//
\n+
603inline ConstIndexArray
\n+
604Level::getVertexFaces(Index vertIndex) const {
\n+
605 return ConstIndexArray( (&_vertFaceIndices[0]) + _vertFaceCountsAndOffsets[vertIndex*2+1],
\n+
606 _vertFaceCountsAndOffsets[vertIndex*2]);
\n+
607}
\n+
608inline IndexArray
\n+
609Level::getVertexFaces(Index vertIndex) {
\n+
610 return IndexArray( (&_vertFaceIndices[0]) + _vertFaceCountsAndOffsets[vertIndex*2+1],
\n+
611 _vertFaceCountsAndOffsets[vertIndex*2]);
\n+
612}
\n+
613
\n+
614inline ConstLocalIndexArray
\n+
615Level::getVertexFaceLocalIndices(Index vertIndex) const {
\n+
616 return ConstLocalIndexArray( (&_vertFaceLocalIndices[0]) + _vertFaceCountsAndOffsets[vertIndex*2+1],
\n+
617 _vertFaceCountsAndOffsets[vertIndex*2]);
\n+
618}
\n+
619inline LocalIndexArray
\n+
620Level::getVertexFaceLocalIndices(Index vertIndex) {
\n+
621 return LocalIndexArray( (&_vertFaceLocalIndices[0]) + _vertFaceCountsAndOffsets[vertIndex*2+1],
\n+
622 _vertFaceCountsAndOffsets[vertIndex*2]);
\n+
623}
\n+
624
\n+
625inline void
\n+
626Level::resizeVertexFaces(Index vertIndex, int count) {
\n+
627 int* countOffsetPair = &_vertFaceCountsAndOffsets[vertIndex*2];
\n+
628
\n+
629 countOffsetPair[0] = count;
\n+
630 countOffsetPair[1] = (vertIndex == 0) ? 0 : (countOffsetPair[-2] + countOffsetPair[-1]);
\n+
631}
\n+
632inline void
\n+
633Level::trimVertexFaces(Index vertIndex, int count) {
\n+
634 _vertFaceCountsAndOffsets[vertIndex*2] = count;
\n+
635}
\n+
636
\n+
637//
\n+
638// Access/modify the edges incident a given vertex:
\n+
639//
\n+
640inline ConstIndexArray
\n+
641Level::getVertexEdges(Index vertIndex) const {
\n+
642 return ConstIndexArray( (&_vertEdgeIndices[0]) +_vertEdgeCountsAndOffsets[vertIndex*2+1],
\n+
643 _vertEdgeCountsAndOffsets[vertIndex*2]);
\n+
644}
\n+
645inline IndexArray
\n+
646Level::getVertexEdges(Index vertIndex) {
\n+
647 return IndexArray( (&_vertEdgeIndices[0]) +_vertEdgeCountsAndOffsets[vertIndex*2+1],
\n+
648 _vertEdgeCountsAndOffsets[vertIndex*2]);
\n+
649}
\n+
650
\n+
651inline ConstLocalIndexArray
\n+
652Level::getVertexEdgeLocalIndices(Index vertIndex) const {
\n+
653 return ConstLocalIndexArray( (&_vertEdgeLocalIndices[0]) + _vertEdgeCountsAndOffsets[vertIndex*2+1],
\n+
654 _vertEdgeCountsAndOffsets[vertIndex*2]);
\n+
655}
\n+
656inline LocalIndexArray
\n+
657Level::getVertexEdgeLocalIndices(Index vertIndex) {
\n+
658 return LocalIndexArray( (&_vertEdgeLocalIndices[0]) + _vertEdgeCountsAndOffsets[vertIndex*2+1],
\n+
659 _vertEdgeCountsAndOffsets[vertIndex*2]);
\n+
660}
\n+
661
\n+
662inline void
\n+
663Level::resizeVertexEdges(Index vertIndex, int count) {
\n+
664 int* countOffsetPair = &_vertEdgeCountsAndOffsets[vertIndex*2];
\n+
665
\n+
666 countOffsetPair[0] = count;
\n+
667 countOffsetPair[1] = (vertIndex == 0) ? 0 : (countOffsetPair[-2] + countOffsetPair[-1]);
\n+
668
\n+
669 _maxValence = std::max(_maxValence, count);
\n+
670}
\n+
671inline void
\n+
672Level::trimVertexEdges(Index vertIndex, int count) {
\n+
673 _vertEdgeCountsAndOffsets[vertIndex*2] = count;
\n+
674}
\n+
675
\n+
676inline void
\n+
677Level::setMaxValence(int valence) {
\n+
678 _maxValence = valence;
\n+
679}
\n+
680
\n+
681//
\n+
682// Access/modify the vertices incident a given edge:
\n+
683//
\n+
684inline ConstIndexArray
\n+
685Level::getEdgeVertices(Index edgeIndex) const {
\n+
686 return ConstIndexArray(&_edgeVertIndices[edgeIndex*2], 2);
\n+
687}
\n+
688inline IndexArray
\n+
689Level::getEdgeVertices(Index edgeIndex) {
\n+
690 return IndexArray(&_edgeVertIndices[edgeIndex*2], 2);
\n+
691}
\n+
692
\n+
693//
\n+
694// Access/modify the faces incident a given edge:
\n+
695//
\n+
696inline ConstIndexArray
\n+
697Level::getEdgeFaces(Index edgeIndex) const {
\n+
698 return ConstIndexArray(&_edgeFaceIndices[0] +
\n+
699 _edgeFaceCountsAndOffsets[edgeIndex*2+1],
\n+
700 _edgeFaceCountsAndOffsets[edgeIndex*2]);
\n+
701}
\n+
702inline IndexArray
\n+
703Level::getEdgeFaces(Index edgeIndex) {
\n+
704 return IndexArray(&_edgeFaceIndices[0] +
\n+
705 _edgeFaceCountsAndOffsets[edgeIndex*2+1],
\n+
706 _edgeFaceCountsAndOffsets[edgeIndex*2]);
\n+
707}
\n+
708
\n+
709inline ConstLocalIndexArray
\n+
710Level::getEdgeFaceLocalIndices(Index edgeIndex) const {
\n+
711 return ConstLocalIndexArray(&_edgeFaceLocalIndices[0] +
\n+
712 _edgeFaceCountsAndOffsets[edgeIndex*2+1],
\n+
713 _edgeFaceCountsAndOffsets[edgeIndex*2]);
\n+
714}
\n+
715inline LocalIndexArray
\n+
716Level::getEdgeFaceLocalIndices(Index edgeIndex) {
\n+
717 return LocalIndexArray(&_edgeFaceLocalIndices[0] +
\n+
718 _edgeFaceCountsAndOffsets[edgeIndex*2+1],
\n+
719 _edgeFaceCountsAndOffsets[edgeIndex*2]);
\n+
720}
\n+
721
\n+
722inline void
\n+
723Level::resizeEdgeFaces(Index edgeIndex, int count) {
\n+
724 int* countOffsetPair = &_edgeFaceCountsAndOffsets[edgeIndex*2];
\n+
725
\n+
726 countOffsetPair[0] = count;
\n+
727 countOffsetPair[1] = (edgeIndex == 0) ? 0 : (countOffsetPair[-2] + countOffsetPair[-1]);
\n+
728
\n+
729 _maxEdgeFaces = std::max(_maxEdgeFaces, count);
\n+
730}
\n+
731inline void
\n+
732Level::trimEdgeFaces(Index edgeIndex, int count) {
\n+
733 _edgeFaceCountsAndOffsets[edgeIndex*2] = count;
\n+
734}
\n+
735
\n+
736//
\n+
737// Access/modify sharpness values:
\n+
738//
\n+
739inline float
\n+
740Level::getEdgeSharpness(Index edgeIndex) const {
\n+
741 return _edgeSharpness[edgeIndex];
\n+
742}
\n+
743inline float&
\n+
744Level::getEdgeSharpness(Index edgeIndex) {
\n+
745 return _edgeSharpness[edgeIndex];
\n+
746}
\n+
747
\n+
748inline float
\n+
749Level::getVertexSharpness(Index vertIndex) const {
\n+
750 return _vertSharpness[vertIndex];
\n+
751}
\n+
752inline float&
\n+
753Level::getVertexSharpness(Index vertIndex) {
\n+
754 return _vertSharpness[vertIndex];
\n+
755}
\n+
756
\n+
757inline Sdc::Crease::Rule
\n+
758Level::getVertexRule(Index vertIndex) const {
\n+
759 return (Sdc::Crease::Rule) _vertTags[vertIndex]._rule;
\n+
760}
\n+
761
\n+
762//
\n+
763// Access/modify hole tag:
\n+
764//
\n+
765inline void
\n+
766Level::setFaceHole(Index faceIndex, bool b) {
\n+
767 _faceTags[faceIndex]._hole = b;
\n+
768}
\n+
769inline bool
\n+
770Level::isFaceHole(Index faceIndex) const {
\n+
771 return _faceTags[faceIndex]._hole;
\n+
772}
\n+
773
\n+
774//
\n+
775// Access/modify non-manifold tags:
\n+
776//
\n+
777inline void
\n+
778Level::setEdgeNonManifold(Index edgeIndex, bool b) {
\n+
779 _edgeTags[edgeIndex]._nonManifold = b;
\n+
780}
\n+
781inline bool
\n+
782Level::isEdgeNonManifold(Index edgeIndex) const {
\n+
783 return _edgeTags[edgeIndex]._nonManifold;
\n+
784}
\n+
785
\n+
786inline void
\n+
787Level::setVertexNonManifold(Index vertIndex, bool b) {
\n+
788 _vertTags[vertIndex]._nonManifold = b;
\n+
789}
\n+
790inline bool
\n+
791Level::isVertexNonManifold(Index vertIndex) const {
\n+
792 return _vertTags[vertIndex]._nonManifold;
\n+
793}
\n+
794
\n+
795//
\n+
796// Sizing methods to allocate space:
\n+
797//
\n+
798inline void
\n+
799Level::resizeFaces(int faceCount) {
\n+
800 _faceCount = faceCount;
\n+
801 _faceVertCountsAndOffsets.resize(2 * faceCount);
\n+
802
\n+
803 _faceTags.resize(faceCount);
\n+
804 std::memset((void*) &_faceTags[0], 0, _faceCount * sizeof(FTag));
\n+
805}
\n+
806inline void
\n+
807Level::resizeFaceVertices(int totalFaceVertCount) {
\n+
808 _faceVertIndices.resize(totalFaceVertCount);
\n+
809}
\n+
810inline void
\n+
811Level::resizeFaceEdges(int totalFaceEdgeCount) {
\n+
812 _faceEdgeIndices.resize(totalFaceEdgeCount);
\n+
813}
\n+
814
\n+
815inline void
\n+
816Level::resizeEdges(int edgeCount) {
\n+
817
\n+
818 _edgeCount = edgeCount;
\n+
819 _edgeFaceCountsAndOffsets.resize(2 * edgeCount);
\n+
820
\n+
821 _edgeSharpness.resize(edgeCount);
\n+
822 _edgeTags.resize(edgeCount);
\n+
823
\n+
824 if (edgeCount>0) {
\n+
825 std::memset((void*) &_edgeTags[0], 0, _edgeCount * sizeof(ETag));
\n+
826 }
\n+
827}
\n+
828inline void
\n+
829Level::resizeEdgeVertices() {
\n+
830
\n+
831 _edgeVertIndices.resize(2 * _edgeCount);
\n+
832}
\n+
833inline void
\n+
834Level::resizeEdgeFaces(int totalEdgeFaceCount) {
\n+
835
\n+
836 _edgeFaceIndices.resize(totalEdgeFaceCount);
\n+
837 _edgeFaceLocalIndices.resize(totalEdgeFaceCount);
\n+
838}
\n+
839
\n+
840inline void
\n+
841Level::resizeVertices(int vertCount) {
\n+
842
\n+
843 _vertCount = vertCount;
\n+
844 _vertFaceCountsAndOffsets.resize(2 * vertCount);
\n+
845 _vertEdgeCountsAndOffsets.resize(2 * vertCount);
\n+
846
\n+
847 _vertSharpness.resize(vertCount);
\n+
848 _vertTags.resize(vertCount);
\n+
849 std::memset((void*) &_vertTags[0], 0, _vertCount * sizeof(VTag));
\n+
850}
\n+
851inline void
\n+
852Level::resizeVertexFaces(int totalVertFaceCount) {
\n+
853
\n+
854 _vertFaceIndices.resize(totalVertFaceCount);
\n+
855 _vertFaceLocalIndices.resize(totalVertFaceCount);
\n+
856}
\n+
857inline void
\n+
858Level::resizeVertexEdges(int totalVertEdgeCount) {
\n+
859
\n+
860 _vertEdgeIndices.resize(totalVertEdgeCount);
\n+
861 _vertEdgeLocalIndices.resize(totalVertEdgeCount);
\n+
862}
\n+
863
\n+
864inline IndexArray
\n+
865Level::shareFaceVertCountsAndOffsets() const {
\n+
866 // XXXX manuelk we have to force const casting here (classes don't 'share'
\n+
867 // members usually...)
\n+
868 return IndexArray(const_cast<Index *>(&_faceVertCountsAndOffsets[0]),
\n+
869 (int)_faceVertCountsAndOffsets.size());
\n+
870}
\n+
871
\n+
872} // end namespace internal
\n+
873} // end namespace Vtr
\n+
874
\n+
875} // end namespace OPENSUBDIV_VERSION
\n+
876using namespace OPENSUBDIV_VERSION;
\n+
877} // end namespace OpenSubdiv
\n+
878
\n+
879#endif /* OPENSUBDIV3_VTR_LEVEL_H */
\n \n+
OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstLocalIndexArray
ConstArray< LocalIndex > ConstLocalIndexArray
Definition: types.h:83
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Vtr::LocalIndexArray
Array< LocalIndex > LocalIndexArray
Definition: types.h:82
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Vtr::IndexVector
std::vector< Index > IndexVector
Definition: types.h:77
\n
OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstIndexArray
ConstArray< Index > ConstIndexArray
Definition: types.h:80
\n+\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Split
Split
Enumerated type for all face splitting schemes.
Definition: types.h:47
\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-refinement.h\n+level.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2014 DreamWorks Animation LLC.\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -29,548 +29,997 @@\n 17//\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n- 24#ifndef OPENSUBDIV3_VTR_REFINEMENT_H\n- 25#define OPENSUBDIV3_VTR_REFINEMENT_H\n+ 24#ifndef OPENSUBDIV3_VTR_LEVEL_H\n+ 25#define OPENSUBDIV3_VTR_LEVEL_H\n 26\n 27#include \"../version.h\"\n 28\n 29#include \"../sdc/types.h\"\n- 30#include \"../sdc/options.h\"\n- 31#include \"../vtr/types.h\"\n- 32#include \"../vtr/level.h\"\n+ 30#include \"../sdc/crease.h\"\n+ 31#include \"../sdc/options.h\"\n+ 32#include \"../vtr/types.h\"\n 33\n- 34#include \n- 35\n- 36//\n- 37// Declaration for the main refinement class (Refinement) and its pre-\n-requisites:\n- 38//\n- 39namespace OpenSubdiv {\n- 40namespace OPENSUBDIV_VERSION {\n- 41\n- 42namespace Vtr {\n- 43namespace internal {\n- 44\n- 45class FVarRefinement;\n- 46\n- 47//\n- 48// Refinement:\n- 49// A refinement is a mapping between two levels -- relating the components\n-in the original\n- 50// (parent) level to the one refined (child). The refinement may be complete\n-(uniform) or sparse\n- 51// (adaptive or otherwise selective), so not all components in the parent\n-level will spawn\n- 52// components in the child level.\n- 53//\n- 54// Refinement is an abstract class and expects subclasses corresponding to\n-the different types\n- 55// of topological splits that the supported subdivision schemes collectively\n-require, i.e. those\n- 56// listed in Sdc::SplitType. Note the virtual requirements expected of the\n-subclasses in the list\n- 57// of protected methods -- they differ mainly in the topology that is\n-created in the child Level\n- 58// and not the propagation of tags through refinement, subdivision of\n-sharpness values or the\n- 59// treatment of face-varying data. The primary subclasses are QuadRefinement\n-and TriRefinement.\n+ 34#include \n+ 35#include \n+ 36#include \n+ 37#include \n+ 38\n+ 39\n+ 40namespace OpenSubdiv {\n+ 41namespace OPENSUBDIV_VERSION {\n+ 42\n+ 43namespace Vtr {\n+ 44namespace internal {\n+ 45\n+ 46class Refinement;\n+ 47class TriRefinement;\n+ 48class QuadRefinement;\n+ 49class FVarRefinement;\n+ 50class FVarLevel;\n+ 51\n+ 52//\n+ 53// Level:\n+ 54// A refinement level includes a vectorized representation of the topology\n+ 55// for a particular subdivision level. The topology is \"complete\" in that\n+any\n+ 56// level can be used as the base level of another subdivision hierarchy and\n+can\n+ 57// be considered a complete mesh independent of its ancestors. It currently\n+ 58// does contain a \"depth\" member -- as some inferences can then be made\n+about\n+ 59// the topology (i.e. all quads or all tris if not level 0).\n 60//\n- 61// At a high level, all that is necessary in terms of interface is to\n-construct, initialize\n- 62// (linking the two levels), optionally select components for sparse\n-refinement (via use of the\n- 63// SparseSelector) and call the refine() method. This usage is expected of\n-Far::TopologyRefiner.\n- 64//\n- 65// Since we really want this class to be restricted from public access\n-eventually, all methods\n- 66// begin with lower case (as is the convention for protected methods) and\n-the list of friends\n- 67// will be maintained more strictly.\n- 68//\n- 69class Refinement {\n- 70\n- 71public:\n- 72 Refinement(Level const & parent, Level & child, Sdc::Options const&\n-schemeOptions);\n- 73 virtual ~Refinement();\n- 74\n- 75 Level const& parent() const { return *_parent; }\n- 76 Level const& child() const { return *_child; }\n- 77 Level& child() { return *_child; }\n- 78\n- 79 Sdc::Split getSplitType() const { return _splitType; }\n- 80 int getRegularFaceSize() const { return _regFaceSize; }\n- 81 Sdc::Options getOptions() const { return _options; }\n+ 61// This class is intended for private use within the library. There are\n+still\n+ 62// opportunities to specialize levels -- e.g. those supporting N-sided faces\n+vs\n+ 63// those that are purely quads or tris -- so we prefer to insulate it from\n+public\n+ 64// access.\n+ 65//\n+ 66// The representation of topology here is to store six topological\n+relationships\n+ 67// in tables of integers. Each is stored in its own array(s) so the result\n+is\n+ 68// a SOA representation of the topology. The six relations are:\n+ 69//\n+ 70// - face-verts: vertices incident/comprising a face\n+ 71// - face-edges: edges incident a face\n+ 72// - edge-verts: vertices incident/comprising an edge\n+ 73// - edge-faces: faces incident an edge\n+ 74// - vert-faces: faces incident a vertex\n+ 75// - vert-edges: edges incident a vertex\n+ 76//\n+ 77// There is some redundancy here but the intent is not that this be a\n+minimal\n+ 78// representation, the intent is that it be amenable to refinement. Classes\n+in\n+ 79// the Far layer essentially store 5 of these 6 in a permuted form -- we add\n+ 80// the face-edges here to simplify refinement.\n+ 81//\n 82\n- 83 // Face-varying:\n- 84 int getNumFVarChannels() const { return (int) _fvarChannels.size(); }\n- 85\n- 86 FVarRefinement const & getFVarRefinement(int c) const { return\n-*_fvarChannels[c]; }\n- 87\n- 88 //\n- 89 // Options associated with the actual refinement operation, which may end\n-up\n- 90 // quite involved if we want to allow for the refinement of data that is\n-not\n- 91 // of interest to be suppressed. For now we have:\n- 92 //\n- 93 // \"sparse\": the alternative to uniform refinement, which requires that\n- 94 // components be previously selected/marked to be included.\n- 95 //\n- 96 // \"minimal topology\": this is one that may get broken down into a finer\n- 97 // set of options. It suppresses \"full topology\" in the child level\n- 98 // and only generates what is minimally necessary for interpolation --\n- 99 // which requires at least the face-vertices for faces, but also the\n- 100 // vertex-faces for any face-varying channels present. So it will\n- 101 // generate one or two of the six possible topological relations.\n- 102 //\n- 103 // These are strictly controlled right now, e.g. for sparse refinement, we\n- 104 // currently enforce full topology at the finest level to allow for\n-subsequent\n- 105 // patch construction.\n- 106 //\n- 107 struct Options {\n- 108 Options() : _sparse(false),\n- 109 _faceVertsFirst(false),\n- 110 _minimalTopology(false)\n- 111 { }\n- 112\n- 113 unsigned int _sparse : 1;\n- 114 unsigned int _faceVertsFirst : 1;\n- 115 unsigned int _minimalTopology : 1;\n- 116\n- 117 // Still under consideration:\n- 118 //unsigned int _childToParentMap : 1;\n- 119 };\n- 120\n- 121 void refine(Options options = Options());\n+ 83class Level {\n+ 84\n+ 85public:\n+ 86 //\n+ 87 // Simple nested types to hold the tags for each component type -- some of\n+ 88 // which are user-specified features (e.g. whether a face is a hole or not)\n+ 89 // while others indicate the topological nature of the component, how it\n+ 90 // is affected by creasing in its neighborhood, etc.\n+ 91 //\n+ 92 // Most of these properties are passed down to child components during\n+ 93 // refinement, but some -- notably the designation of a component as semi-\n+ 94 // sharp -- require re-determination as sharpness values are reduced at\n+each\n+ 95 // level.\n+ 96 //\n+ 97 struct VTag {\n+ 98 VTag() { }\n+ 99\n+ 100 // When cleared, the VTag ALMOST represents a smooth, regular, interior\n+ 101 // vertex -- the Type enum requires a bit be explicitly set for Smooth,\n+ 102 // so that must be done explicitly if desired on initialization.\n+ 103 void clear() { std::memset((void*) this, 0, sizeof(VTag)); }\n+ 104\n+ 105 typedef unsigned short VTagSize;\n+ 106\n+ 107 VTagSize _nonManifold : 1; // fixed\n+ 108 VTagSize _xordinary : 1; // fixed\n+ 109 VTagSize _boundary : 1; // fixed\n+ 110 VTagSize _corner : 1; // fixed\n+ 111 VTagSize _infSharp : 1; // fixed\n+ 112 VTagSize _semiSharp : 1; // variable\n+ 113 VTagSize _semiSharpEdges : 1; // variable\n+ 114 VTagSize _rule : 4; // variable when _semiSharp\n+ 115\n+ 116 // These next to tags are complementary -- the \"incomplete\" tag is only\n+ 117 // relevant for refined levels while the \"incident an irregular face\" tag\n+ 118 // is only relevant for the base level. They could be combined as both\n+ 119 // indicate \"no full regular ring\" around a vertex\n+ 120 VTagSize _incomplete : 1; // variable only set in refined levels\n+ 121 VTagSize _incidIrregFace : 1; // variable only set in base level\n 122\n- 123 bool hasFaceVerticesFirst() const { return _faceVertsFirst; }\n- 124\n- 125public:\n- 126 //\n- 127 // Access to members -- some testing classes (involving vertex\n-interpolation)\n- 128 // currently make use of these:\n- 129 //\n- 130 int getNumChildFacesFromFaces() const { return _childFaceFromFaceCount; }\n- 131 int getNumChildEdgesFromFaces() const { return _childEdgeFromFaceCount; }\n- 132 int getNumChildEdgesFromEdges() const { return _childEdgeFromEdgeCount; }\n- 133 int getNumChildVerticesFromFaces() const { return _childVertFromFaceCount;\n-}\n- 134 int getNumChildVerticesFromEdges() const { return _childVertFromEdgeCount;\n-}\n- 135 int getNumChildVerticesFromVertices() const { return\n-_childVertFromVertCount; }\n- 136\n- 137 Index getFirstChildFaceFromFaces() const { return _firstChildFaceFromFace;\n-}\n- 138 Index getFirstChildEdgeFromFaces() const { return _firstChildEdgeFromFace;\n-}\n- 139 Index getFirstChildEdgeFromEdges() const { return _firstChildEdgeFromEdge;\n-}\n- 140 Index getFirstChildVertexFromFaces() const { return\n-_firstChildVertFromFace; }\n- 141 Index getFirstChildVertexFromEdges() const { return\n-_firstChildVertFromEdge; }\n- 142 Index getFirstChildVertexFromVertices() const { return\n-_firstChildVertFromVert; }\n- 143\n- 144 Index getFaceChildVertex(Index f) const { return _faceChildVertIndex[f]; }\n- 145 Index getEdgeChildVertex(Index e) const { return _edgeChildVertIndex[e]; }\n- 146 Index getVertexChildVertex(Index v) const { return _vertChildVertIndex[v];\n-}\n+ 123 // Tags indicating incident infinitely-sharp (permanent) features\n+ 124 VTagSize _infSharpEdges : 1; // fixed\n+ 125 VTagSize _infSharpCrease : 1; // fixed\n+ 126 VTagSize _infIrregular : 1; // fixed\n+ 127\n+ 128 // Alternate constructor and accessor for dealing with integer bits\n+directly:\n+ 129 explicit VTag(VTagSize bits) {\n+ 130 std::memcpy(this, &bits, sizeof(bits));\n+ 131 }\n+ 132 VTagSize getBits() const {\n+ 133 VTagSize bits;\n+ 134 std::memcpy(&bits, this, sizeof(bits));\n+ 135 return bits;\n+ 136 }\n+ 137\n+ 138 static VTag BitwiseOr(VTag const vTags[], int size = 4);\n+ 139 };\n+ 140 struct ETag {\n+ 141 ETag() { }\n+ 142\n+ 143 // When cleared, the ETag represents a smooth, manifold, interior edge\n+ 144 void clear() { std::memset((void*) this, 0, sizeof(ETag)); }\n+ 145\n+ 146 typedef unsigned char ETagSize;\n 147\n- 148 ConstIndexArray getFaceChildFaces(Index parentFace) const;\n- 149 ConstIndexArray getFaceChildEdges(Index parentFace) const;\n- 150 ConstIndexArray getEdgeChildEdges(Index parentEdge) const;\n- 151\n- 152 // Child-to-parent relationships\n- 153 bool isChildVertexComplete(Index v) const { return ! _childVertexTag\n-[v]._incomplete; }\n- 154\n- 155 Index getChildFaceParentFace(Index f) const { return _childFaceParentIndex\n-[f]; }\n- 156 int getChildFaceInParentFace(Index f) const { return _childFaceTag\n-[f]._indexInParent; }\n- 157\n- 158 Index getChildEdgeParentIndex(Index e) const { return\n-_childEdgeParentIndex[e]; }\n- 159\n- 160 Index getChildVertexParentIndex(Index v) const { return\n-_childVertexParentIndex[v]; }\n- 161\n- 162//\n- 163// Modifiers intended for internal/protected use:\n- 164//\n- 165public:\n- 166\n- 167 IndexArray getFaceChildFaces(Index parentFace);\n- 168 IndexArray getFaceChildEdges(Index parentFace);\n- 169 IndexArray getEdgeChildEdges(Index parentEdge);\n- 170\n- 171public:\n- 172 //\n- 173 // Tags have now been added per-component in Level, but there is\n-additional need to tag\n- 174 // components within Refinement -- we can't tag the parent level\n-components for any\n- 175 // refinement (in order to keep it const) and tags associated with\n-children that are\n- 176 // specific to the child-to-parent mapping may not be warranted in the\n-child level.\n- 177 //\n- 178 // Parent tags are only required for sparse refinement. The main property\n-to tag is\n- 179 // whether a component was selected, and so a single SparseTag is used for\n-all three\n- 180 // component types. Tagging if a component is \"transitional\" is also\n-useful. This may\n- 181 // only be necessary for edges but is currently packed into a mask per-\n-edge for faces,\n- 182 // which could be deferred, in which case \"transitional\" could be a single\n-bit.\n- 183 //\n- 184 // Child tags are part of the child-to-parent mapping, which consists of\n-the parent\n- 185 // component index for each child component, plus a tag for the child\n-indicating more\n- 186 // about its relationship to its parent, e.g. is it completely defined,\n-what the parent\n- 187 // component type is, what is the index of the child within its parent,\n-etc.\n+ 148 ETagSize _nonManifold : 1; // fixed\n+ 149 ETagSize _boundary : 1; // fixed\n+ 150 ETagSize _infSharp : 1; // fixed\n+ 151 ETagSize _semiSharp : 1; // variable\n+ 152\n+ 153 // Alternate constructor and accessor for dealing with integer bits\n+directly:\n+ 154 explicit ETag(ETagSize bits) {\n+ 155 std::memcpy(this, &bits, sizeof(bits));\n+ 156 }\n+ 157 ETagSize getBits() const {\n+ 158 ETagSize bits;\n+ 159 std::memcpy(&bits, this, sizeof(bits));\n+ 160 return bits;\n+ 161 }\n+ 162\n+ 163 static ETag BitwiseOr(ETag const eTags[], int size = 4);\n+ 164 };\n+ 165 struct FTag {\n+ 166 FTag() { }\n+ 167\n+ 168 void clear() { std::memset((void*) this, 0, sizeof(FTag)); }\n+ 169\n+ 170 typedef unsigned char FTagSize;\n+ 171\n+ 172 FTagSize _hole : 1; // fixed\n+ 173\n+ 174 // On deck -- coming soon...\n+ 175 //FTagSize _hasEdits : 1; // variable\n+ 176 };\n+ 177\n+ 178 // Additional simple struct to identify a \"span\" around a vertex, i.e. a\n+ 179 // subset of the faces around a vertex delimited by some property (e.g. a\n+ 180 // face-varying discontinuity, an inf-sharp edge, etc.)\n+ 181 //\n+ 182 // The span requires an \"origin\" and a \"size\" to fully define its extent.\n+ 183 // Use of the size is required over a leading/trailing pair as the valence\n+ 184 // around a non-manifold vertex cannot be trivially determined from two\n+ 185 // extremeties. Similarly a start face is chosen over an edge as starting\n+ 186 // with a manifold edge is ambiguous. Additional tags also support\n+ 187 // non-manifold cases, e.g. periodic spans at the apex of a double cone.\n 188 //\n- 189 struct SparseTag {\n- 190 SparseTag() : _selected(0), _transitional(0) { }\n- 191\n- 192 unsigned char _selected : 1; // component specifically selected for\n-refinement\n- 193 unsigned char _transitional : 4; // adjacent to a refined component (4-\n-bits for face)\n- 194 };\n+ 189 // Currently setting the size to 0 or leaving the span \"unassigned\" is an\n+ 190 // indication to use the full neighborhood rather than a subset -- prefer\n+ 191 // use of the const method here to direct inspection of the member.\n+ 192 //\n+ 193 struct VSpan {\n+ 194 VSpan() { std::memset((void*) this, 0, sizeof(VSpan)); }\n 195\n- 196 struct ChildTag {\n- 197 ChildTag() { }\n+ 196 void clear() { std::memset((void*) this, 0, sizeof(VSpan)); }\n+ 197 bool isAssigned() const { return _numFaces > 0; }\n 198\n- 199 unsigned char _incomplete : 1; // incomplete neighborhood to represent\n-limit of parent\n- 200 unsigned char _parentType : 2; // type of parent component: vertex, edge\n-or face\n- 201 unsigned char _indexInParent : 2; // index of child wrt parent: 0-3, or\n-iterative if N > 4\n- 202 };\n- 203\n- 204 // Methods to access and modify tags:\n- 205 SparseTag const & getParentFaceSparseTag( Index f) const { return\n-_parentFaceTag[f]; }\n- 206 SparseTag const & getParentEdgeSparseTag( Index e) const { return\n-_parentEdgeTag[e]; }\n- 207 SparseTag const & getParentVertexSparseTag(Index v) const { return\n-_parentVertexTag[v]; }\n- 208\n- 209 SparseTag & getParentFaceSparseTag( Index f) { return _parentFaceTag[f]; }\n- 210 SparseTag & getParentEdgeSparseTag( Index e) { return _parentEdgeTag[e]; }\n- 211 SparseTag & getParentVertexSparseTag(Index v) { return _parentVertexTag\n-[v]; }\n- 212\n- 213 ChildTag const & getChildFaceTag( Index f) const { return _childFaceTag\n-[f]; }\n- 214 ChildTag const & getChildEdgeTag( Index e) const { return _childEdgeTag\n-[e]; }\n- 215 ChildTag const & getChildVertexTag(Index v) const { return _childVertexTag\n-[v]; }\n- 216\n- 217 ChildTag & getChildFaceTag( Index f) { return _childFaceTag[f]; }\n- 218 ChildTag & getChildEdgeTag( Index e) { return _childEdgeTag[e]; }\n- 219 ChildTag & getChildVertexTag(Index v) { return _childVertexTag[v]; }\n- 220\n- 221// Remaining methods should really be protected -- for use by subclasses...\n- 222public:\n- 223 //\n- 224 // Methods involved in constructing the parent-to-child mapping -- when\n-the\n- 225 // refinement is sparse, additional methods are needed to identify the\n-selection:\n- 226 //\n- 227 void populateParentToChildMapping();\n- 228 void populateParentChildIndices();\n- 229 void printParentToChildMapping() const;\n- 230\n- 231 virtual void allocateParentChildIndices() = 0;\n- 232\n- 233 // Supporting method for sparse refinement:\n- 234 void initializeSparseSelectionTags();\n- 235 void markSparseChildComponentIndices();\n- 236 void markSparseVertexChildren();\n- 237 void markSparseEdgeChildren();\n- 238\n- 239 virtual void markSparseFaceChildren() = 0;\n- 240\n- 241 void initializeChildComponentCounts();\n- 242\n- 243 //\n- 244 // Methods involved in constructing the child-to-parent mapping:\n- 245 //\n- 246 void populateChildToParentMapping();\n- 247\n- 248 void populateFaceParentVectors(ChildTag const initialChildTags[2][4]);\n- 249 void populateFaceParentFromParentFaces(ChildTag const initialChildTags[2]\n-[4]);\n+ 199 LocalIndex _numFaces;\n+ 200 LocalIndex _startFace;\n+ 201 LocalIndex _cornerInSpan;\n+ 202\n+ 203 unsigned short _periodic : 1;\n+ 204 unsigned short _sharp : 1;\n+ 205 };\n+ 206\n+ 207public:\n+ 208 Level();\n+ 209 ~Level();\n+ 210\n+ 211 // Simple accessors:\n+ 212 int getDepth() const { return _depth; }\n+ 213\n+ 214 int getNumVertices() const { return _vertCount; }\n+ 215 int getNumFaces() const { return _faceCount; }\n+ 216 int getNumEdges() const { return _edgeCount; }\n+ 217\n+ 218 // More global sizes may prove useful...\n+ 219 int getNumFaceVerticesTotal() const { return (int) _faceVertIndices.size\n+(); }\n+ 220 int getNumFaceEdgesTotal() const { return (int) _faceEdgeIndices.size(); }\n+ 221 int getNumEdgeVerticesTotal() const { return (int) _edgeVertIndices.size\n+(); }\n+ 222 int getNumEdgeFacesTotal() const { return (int) _edgeFaceIndices.size(); }\n+ 223 int getNumVertexFacesTotal() const { return (int) _vertFaceIndices.size();\n+}\n+ 224 int getNumVertexEdgesTotal() const { return (int) _vertEdgeIndices.size();\n+}\n+ 225\n+ 226 int getMaxValence() const { return _maxValence; }\n+ 227 int getMaxEdgeFaces() const { return _maxEdgeFaces; }\n+ 228\n+ 229 // Methods to access the relation tables/indices -- note that for some\n+relations\n+ 230 // (i.e. those where a component is \"contained by\" a neighbor, or more\n+generally\n+ 231 // when the neighbor is a simplex of higher dimension) we store an\n+additional\n+ 232 // \"local index\", e.g. for the case of vert-faces if one of the faces F[i]\n+is\n+ 233 // incident a vertex V, then L[i] is the \"local index\" in F[i] of vertex\n+V.\n+ 234 // Once have only quads (or tris), this local index need only occupy two\n+bits\n+ 235 // and could conceivably be packed into the same integer as the face\n+index, but\n+ 236 // for now, given the need to support faces of potentially high valence\n+we'll\n+ 237 // use an 8- or 16-bit integer.\n+ 238 //\n+ 239 // Methods to access the six topological relations:\n+ 240 ConstIndexArray getFaceVertices(Index faceIndex) const;\n+ 241 ConstIndexArray getFaceEdges(Index faceIndex) const;\n+ 242 ConstIndexArray getEdgeVertices(Index edgeIndex) const;\n+ 243 ConstIndexArray getEdgeFaces(Index edgeIndex) const;\n+ 244 ConstIndexArray getVertexFaces(Index vertIndex) const;\n+ 245 ConstIndexArray getVertexEdges(Index vertIndex) const;\n+ 246\n+ 247 ConstLocalIndexArray getEdgeFaceLocalIndices(Index edgeIndex) const;\n+ 248 ConstLocalIndexArray getVertexFaceLocalIndices(Index vertIndex) const;\n+ 249 ConstLocalIndexArray getVertexEdgeLocalIndices(Index vertIndex) const;\n 250\n- 251 void populateEdgeParentVectors(ChildTag const initialChildTags[2][4]);\n- 252 void populateEdgeParentFromParentFaces(ChildTag const initialChildTags[2]\n-[4]);\n- 253 void populateEdgeParentFromParentEdges(ChildTag const initialChildTags[2]\n-[4]);\n- 254\n- 255 void populateVertexParentVectors(ChildTag const initialChildTags[2][4]);\n- 256 void populateVertexParentFromParentFaces(ChildTag const initialChildTags\n-[2][4]);\n- 257 void populateVertexParentFromParentEdges(ChildTag const initialChildTags\n-[2][4]);\n- 258 void populateVertexParentFromParentVertices(ChildTag const\n-initialChildTags[2][4]);\n- 259\n- 260 //\n- 261 // Methods involved in propagating component tags from parent to child:\n- 262 //\n- 263 void propagateComponentTags();\n- 264\n- 265 void populateFaceTagVectors();\n- 266 void populateFaceTagsFromParentFaces();\n+ 251 // Replace these with access to sharpness buffers/arrays rather than\n+elements:\n+ 252 float getEdgeSharpness(Index edgeIndex) const;\n+ 253 float getVertexSharpness(Index vertIndex) const;\n+ 254 Sdc::Crease::Rule getVertexRule(Index vertIndex) const;\n+ 255\n+ 256 Index findEdge(Index v0Index, Index v1Index) const;\n+ 257\n+ 258 // Holes\n+ 259 void setFaceHole(Index faceIndex, bool b);\n+ 260 bool isFaceHole(Index faceIndex) const;\n+ 261\n+ 262 // Face-varying\n+ 263 Sdc::Options getFVarOptions(int channel) const;\n+ 264 int getNumFVarChannels() const { return (int) _fvarChannels.size(); }\n+ 265 int getNumFVarValues(int channel) const;\n+ 266 ConstIndexArray getFaceFVarValues(Index faceIndex, int channel) const;\n 267\n- 268 void populateEdgeTagVectors();\n- 269 void populateEdgeTagsFromParentFaces();\n- 270 void populateEdgeTagsFromParentEdges();\n- 271\n- 272 void populateVertexTagVectors();\n- 273 void populateVertexTagsFromParentFaces();\n- 274 void populateVertexTagsFromParentEdges();\n- 275 void populateVertexTagsFromParentVertices();\n- 276\n- 277 //\n- 278 // Methods (and types) involved in subdividing the topology -- though not\n- 279 // fully exploited, any subset of the 6 relations can be generated:\n- 280 //\n- 281 struct Relations {\n- 282 unsigned int _faceVertices : 1;\n- 283 unsigned int _faceEdges : 1;\n- 284 unsigned int _edgeVertices : 1;\n- 285 unsigned int _edgeFaces : 1;\n- 286 unsigned int _vertexFaces : 1;\n- 287 unsigned int _vertexEdges : 1;\n+ 268 FVarLevel & getFVarLevel(int channel) { return *_fvarChannels[channel]; }\n+ 269 FVarLevel const & getFVarLevel(int channel) const { return *_fvarChannels\n+[channel]; }\n+ 270\n+ 271 // Manifold/non-manifold tags:\n+ 272 void setEdgeNonManifold(Index edgeIndex, bool b);\n+ 273 bool isEdgeNonManifold(Index edgeIndex) const;\n+ 274\n+ 275 void setVertexNonManifold(Index vertIndex, bool b);\n+ 276 bool isVertexNonManifold(Index vertIndex) const;\n+ 277\n+ 278 // General access to all component tags:\n+ 279 VTag const & getVertexTag(Index vertIndex) const { return _vertTags\n+[vertIndex]; }\n+ 280 ETag const & getEdgeTag(Index edgeIndex) const { return _edgeTags\n+[edgeIndex]; }\n+ 281 FTag const & getFaceTag(Index faceIndex) const { return _faceTags\n+[faceIndex]; }\n+ 282\n+ 283 VTag & getVertexTag(Index vertIndex) { return _vertTags[vertIndex]; }\n+ 284 ETag & getEdgeTag(Index edgeIndex) { return _edgeTags[edgeIndex]; }\n+ 285 FTag & getFaceTag(Index faceIndex) { return _faceTags[faceIndex]; }\n+ 286\n+ 287public:\n 288\n- 289 void setAll(bool enable) {\n- 290 _faceVertices = enable;\n- 291 _faceEdges = enable;\n- 292 _edgeVertices = enable;\n- 293 _edgeFaces = enable;\n- 294 _vertexFaces = enable;\n- 295 _vertexEdges = enable;\n- 296 }\n- 297 };\n- 298\n- 299 void subdivideTopology(Relations const& relationsToSubdivide);\n- 300\n- 301 virtual void populateFaceVertexRelation() = 0;\n- 302 virtual void populateFaceEdgeRelation() = 0;\n- 303 virtual void populateEdgeVertexRelation() = 0;\n- 304 virtual void populateEdgeFaceRelation() = 0;\n- 305 virtual void populateVertexFaceRelation() = 0;\n- 306 virtual void populateVertexEdgeRelation() = 0;\n- 307\n- 308 //\n- 309 // Methods involved in subdividing and inspecting sharpness values:\n- 310 //\n- 311 void subdivideSharpnessValues();\n+ 289 // Debugging aides:\n+ 290 enum TopologyError {\n+ 291 TOPOLOGY_MISSING_EDGE_FACES=0,\n+ 292 TOPOLOGY_MISSING_EDGE_VERTS,\n+ 293 TOPOLOGY_MISSING_FACE_EDGES,\n+ 294 TOPOLOGY_MISSING_FACE_VERTS,\n+ 295 TOPOLOGY_MISSING_VERT_FACES,\n+ 296 TOPOLOGY_MISSING_VERT_EDGES,\n+ 297\n+ 298 TOPOLOGY_FAILED_CORRELATION_EDGE_FACE,\n+ 299 TOPOLOGY_FAILED_CORRELATION_FACE_VERT,\n+ 300 TOPOLOGY_FAILED_CORRELATION_FACE_EDGE,\n+ 301\n+ 302 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_EDGE,\n+ 303 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_FACE,\n+ 304 TOPOLOGY_FAILED_ORIENTATION_INCIDENT_FACES_EDGES,\n+ 305\n+ 306 TOPOLOGY_DEGENERATE_EDGE,\n+ 307 TOPOLOGY_NON_MANIFOLD_EDGE,\n+ 308\n+ 309 TOPOLOGY_INVALID_CREASE_EDGE,\n+ 310 TOPOLOGY_INVALID_CREASE_VERT\n+ 311 };\n 312\n- 313 void subdivideVertexSharpness();\n- 314 void subdivideEdgeSharpness();\n- 315 void reclassifySemisharpVertices();\n+ 313 static char const * getTopologyErrorString(TopologyError errCode);\n+ 314\n+ 315 typedef void (* ValidationCallback)(TopologyError errCode, char const *\n+msg, void const * clientData);\n 316\n- 317 //\n- 318 // Methods involved in subdividing face-varying topology:\n- 319 //\n- 320 void subdivideFVarChannels();\n- 321\n- 322protected:\n- 323 // A debug method of Level prints a Refinement (should really change this)\n- 324 friend void Level::print(const Refinement *) const;\n+ 317 bool validateTopology(ValidationCallback callback=0, void const *\n+clientData=0) const;\n+ 318\n+ 319 void print(const Refinement* parentRefinement = 0) const;\n+ 320\n+ 321public:\n+ 322 // High-level topology queries -- these may be moved elsewhere:\n+ 323\n+ 324 bool isSingleCreasePatch(Index face, float* sharpnessOut=NULL, int*\n+rotationOut=NULL) const;\n 325\n 326 //\n- 327 // Data members -- the logical grouping of some of these (and methods that\n-make use\n- 328 // of them) may lead to grouping them into a few utility classes or\n-structs...\n- 329 //\n- 330\n- 331 // Defined on construction:\n- 332 Level const * _parent;\n- 333 Level * _child;\n- 334 Sdc::Options _options;\n- 335\n- 336 // Defined by the subclass:\n- 337 Sdc::Split _splitType;\n- 338 int _regFaceSize;\n- 339\n- 340 // Determined by the refinement options:\n- 341 bool _uniform;\n- 342 bool _faceVertsFirst;\n+ 327 // When inspecting topology, the component tags -- particularly VTag and\n+ETag -- are most\n+ 328 // often inspected in groups for the face to which they belong. They are\n+designed to be\n+ 329 // bitwise OR'd (the result then referred to as a \"composite\" tag) to make\n+quick decisions\n+ 330 // about the face as a whole to avoid tedious topological inspection.\n+ 331 //\n+ 332 // The same logic can be applied to topology in a FVar channel when tags\n+specific to that\n+ 333 // channel are used. Note that the VTags apply to the FVar values assigned\n+to the corners\n+ 334 // of the face and not the vertex as a whole. The \"composite\" face-varying\n+VTag for a\n+ 335 // vertex is the union of VTags of all distinct FVar values for that\n+vertex.\n+ 336 //\n+ 337 bool doesVertexFVarTopologyMatch(Index vIndex, int fvarChannel) const;\n+ 338 bool doesFaceFVarTopologyMatch( Index fIndex, int fvarChannel) const;\n+ 339 bool doesEdgeFVarTopologyMatch( Index eIndex, int fvarChannel) const;\n+ 340\n+ 341 void getFaceVTags(Index fIndex, VTag vTags[], int fvarChannel = -1) const;\n+ 342 void getFaceETags(Index fIndex, ETag eTags[], int fvarChannel = -1) const;\n 343\n- 344 //\n- 345 // Inventory and ordering of the types of child components:\n- 346 //\n- 347 int _childFaceFromFaceCount; // arguably redundant (all faces originate\n-from faces)\n- 348 int _childEdgeFromFaceCount;\n- 349 int _childEdgeFromEdgeCount;\n- 350 int _childVertFromFaceCount;\n- 351 int _childVertFromEdgeCount;\n- 352 int _childVertFromVertCount;\n- 353\n- 354 int _firstChildFaceFromFace; // arguably redundant (all faces originate\n-from faces)\n- 355 int _firstChildEdgeFromFace;\n- 356 int _firstChildEdgeFromEdge;\n- 357 int _firstChildVertFromFace;\n- 358 int _firstChildVertFromEdge;\n- 359 int _firstChildVertFromVert;\n- 360\n- 361 //\n- 362 // The parent-to-child mapping:\n- 363 // These are vectors sized according to the number of parent components\n-(and\n- 364 // their topology) that contain references/indices to the child components\n-that\n- 365 // result from them by refinement. When refinement is sparse, parent\n-components\n- 366 // that have not spawned all child components will have their missing\n-children\n- 367 // marked as invalid.\n- 368 //\n- 369 // NOTE the \"Array\" members here. Often vectors within the Level can be\n-shared\n- 370 // with the Refinement, and an Array instance is used to do so. If not\n-shared\n- 371 // the subclass just initializes the Array members after allocating its\n-own local\n- 372 // vector members.\n- 373 //\n- 374 IndexArray _faceChildFaceCountsAndOffsets;\n- 375 IndexArray _faceChildEdgeCountsAndOffsets;\n+ 344 VTag getFaceCompositeVTag(Index fIndex, int fvarChannel = -1) const;\n+ 345 VTag getFaceCompositeVTag(ConstIndexArray & fVerts) const;\n+ 346\n+ 347 VTag getVertexCompositeFVarVTag(Index vIndex, int fvarChannel) const;\n+ 348\n+ 349 //\n+ 350 // When gathering \"patch points\" we may want the indices of the vertices\n+or the corresponding\n+ 351 // FVar values for a particular channel. Both are represented and equally\n+accessible within\n+ 352 // the faces, so we allow all to be returned through these methods.\n+Setting the optional FVar\n+ 353 // channel to -1 will retrieve indices of vertices instead of FVar values:\n+ 354 //\n+ 355 int gatherQuadLinearPatchPoints(Index fIndex, Index patchPoints[], int\n+rotation = 0,\n+ 356 int fvarChannel = -1) const;\n+ 357\n+ 358 int gatherQuadRegularInteriorPatchPoints(Index fIndex, Index patchPoints\n+[], int rotation = 0,\n+ 359 int fvarChannel = -1) const;\n+ 360 int gatherQuadRegularBoundaryPatchPoints(Index fIndex, Index patchPoints\n+[], int boundaryEdgeInFace,\n+ 361 int fvarChannel = -1) const;\n+ 362 int gatherQuadRegularCornerPatchPoints( Index fIndex, Index patchPoints[],\n+int cornerVertInFace,\n+ 363 int fvarChannel = -1) const;\n+ 364\n+ 365 int gatherQuadRegularRingAroundVertex(Index vIndex, Index ringPoints[],\n+ 366 int fvarChannel = -1) const;\n+ 367 int gatherQuadRegularPartialRingAroundVertex(Index vIndex, VSpan const &\n+span, Index ringPoints[],\n+ 368 int fvarChannel = -1) const;\n+ 369\n+ 370 // WIP -- for future use, need to extend for face-varying...\n+ 371 int gatherTriRegularInteriorPatchPoints( Index fIndex, Index patchVerts[],\n+int rotation = 0) const;\n+ 372 int gatherTriRegularBoundaryVertexPatchPoints(Index fIndex, Index\n+patchVerts[], int boundaryVertInFace) const;\n+ 373 int gatherTriRegularBoundaryEdgePatchPoints( Index fIndex, Index\n+patchVerts[], int boundaryEdgeInFace) const;\n+ 374 int gatherTriRegularCornerVertexPatchPoints( Index fIndex, Index\n+patchVerts[], int cornerVertInFace) const;\n+ 375 int gatherTriRegularCornerEdgePatchPoints( Index fIndex, Index patchVerts\n+[], int cornerEdgeInFace) const;\n 376\n- 377 IndexVector _faceChildFaceIndices; // *cannot* always use face-vert\n-counts/offsets\n- 378 IndexVector _faceChildEdgeIndices; // can use face-vert counts/offsets\n- 379 IndexVector _faceChildVertIndex;\n- 380\n- 381 IndexVector _edgeChildEdgeIndices; // trivial/corresponding pair for each\n- 382 IndexVector _edgeChildVertIndex;\n- 383\n- 384 IndexVector _vertChildVertIndex;\n- 385\n- 386 //\n- 387 // The child-to-parent mapping:\n- 388 //\n- 389 IndexVector _childFaceParentIndex;\n- 390 IndexVector _childEdgeParentIndex;\n- 391 IndexVector _childVertexParentIndex;\n+ 377public:\n+ 378 // Sizing methods used to construct a level to populate:\n+ 379 void resizeFaces( int numFaces);\n+ 380 void resizeFaceVertices(int numFaceVertsTotal);\n+ 381 void resizeFaceEdges( int numFaceEdgesTotal);\n+ 382\n+ 383 void resizeEdges( int numEdges);\n+ 384 void resizeEdgeVertices(); // always 2*edgeCount\n+ 385 void resizeEdgeFaces(int numEdgeFacesTotal);\n+ 386\n+ 387 void resizeVertices( int numVertices);\n+ 388 void resizeVertexFaces(int numVertexFacesTotal);\n+ 389 void resizeVertexEdges(int numVertexEdgesTotal);\n+ 390\n+ 391 void setMaxValence(int maxValence);\n 392\n- 393 std::vector _childFaceTag;\n- 394 std::vector _childEdgeTag;\n- 395 std::vector _childVertexTag;\n- 396\n- 397 //\n- 398 // Tags for sparse selection of components:\n- 399 //\n- 400 std::vector _parentFaceTag;\n- 401 std::vector _parentEdgeTag;\n- 402 std::vector _parentVertexTag;\n- 403\n- 404 //\n- 405 // Refinement data for face-varying channels present in the Levels being\n-refined:\n- 406 //\n- 407 std::vector _fvarChannels;\n- 408};\n- 409\n- 410inline ConstIndexArray\n- 411Refinement::getFaceChildFaces(Index parentFace) const {\n+ 393 // Modifiers to populate the relations for each component:\n+ 394 IndexArray getFaceVertices(Index faceIndex);\n+ 395 IndexArray getFaceEdges(Index faceIndex);\n+ 396 IndexArray getEdgeVertices(Index edgeIndex);\n+ 397 IndexArray getEdgeFaces(Index edgeIndex);\n+ 398 IndexArray getVertexFaces(Index vertIndex);\n+ 399 IndexArray getVertexEdges(Index vertIndex);\n+ 400\n+ 401 LocalIndexArray getEdgeFaceLocalIndices(Index edgeIndex);\n+ 402 LocalIndexArray getVertexFaceLocalIndices(Index vertIndex);\n+ 403 LocalIndexArray getVertexEdgeLocalIndices(Index vertIndex);\n+ 404\n+ 405 // Replace these with access to sharpness buffers/arrays rather than\n+elements:\n+ 406 float& getEdgeSharpness(Index edgeIndex);\n+ 407 float& getVertexSharpness(Index vertIndex);\n+ 408\n+ 409 // Create, destroy and populate face-varying channels:\n+ 410 int createFVarChannel(int fvarValueCount, Sdc::Options const& options);\n+ 411 void destroyFVarChannel(int channel);\n 412\n- 413 return ConstIndexArray(&_faceChildFaceIndices\n-[_faceChildFaceCountsAndOffsets[2*parentFace+1]],\n- 414 _faceChildFaceCountsAndOffsets[2*parentFace]);\n- 415}\n+ 413 IndexArray getFaceFVarValues(Index faceIndex, int channel);\n+ 414\n+ 415 void completeFVarChannelTopology(int channel, int regBoundaryValence);\n 416\n- 417inline IndexArray\n- 418Refinement::getFaceChildFaces(Index parentFace) {\n- 419\n- 420 return IndexArray(&_faceChildFaceIndices[_faceChildFaceCountsAndOffsets\n-[2*parentFace+1]],\n- 421 _faceChildFaceCountsAndOffsets[2*parentFace]);\n- 422}\n- 423\n- 424inline ConstIndexArray\n- 425Refinement::getFaceChildEdges(Index parentFace) const {\n- 426\n- 427 return ConstIndexArray(&_faceChildEdgeIndices\n-[_faceChildEdgeCountsAndOffsets[2*parentFace+1]],\n- 428 _faceChildEdgeCountsAndOffsets[2*parentFace]);\n- 429}\n- 430inline IndexArray\n- 431Refinement::getFaceChildEdges(Index parentFace) {\n- 432\n- 433 return IndexArray(&_faceChildEdgeIndices[_faceChildEdgeCountsAndOffsets\n-[2*parentFace+1]],\n- 434 _faceChildEdgeCountsAndOffsets[2*parentFace]);\n- 435}\n+ 417 // Counts and offsets for all relation types:\n+ 418 // - these may be unwarranted if we let Refinement access members\n+directly...\n+ 419 int getNumFaceVertices( Index faceIndex) const { return\n+_faceVertCountsAndOffsets[2*faceIndex]; }\n+ 420 int getOffsetOfFaceVertices(Index faceIndex) const { return\n+_faceVertCountsAndOffsets[2*faceIndex + 1]; }\n+ 421\n+ 422 int getNumFaceEdges( Index faceIndex) const { return getNumFaceVertices\n+(faceIndex); }\n+ 423 int getOffsetOfFaceEdges(Index faceIndex) const { return\n+getOffsetOfFaceVertices(faceIndex); }\n+ 424\n+ 425 int getNumEdgeVertices( Index ) const { return 2; }\n+ 426 int getOffsetOfEdgeVertices(Index edgeIndex) const { return 2 * edgeIndex;\n+}\n+ 427\n+ 428 int getNumEdgeFaces( Index edgeIndex) const { return\n+_edgeFaceCountsAndOffsets[2*edgeIndex]; }\n+ 429 int getOffsetOfEdgeFaces(Index edgeIndex) const { return\n+_edgeFaceCountsAndOffsets[2*edgeIndex + 1]; }\n+ 430\n+ 431 int getNumVertexFaces( Index vertIndex) const { return\n+_vertFaceCountsAndOffsets[2*vertIndex]; }\n+ 432 int getOffsetOfVertexFaces(Index vertIndex) const { return\n+_vertFaceCountsAndOffsets[2*vertIndex + 1]; }\n+ 433\n+ 434 int getNumVertexEdges( Index vertIndex) const { return\n+_vertEdgeCountsAndOffsets[2*vertIndex]; }\n+ 435 int getOffsetOfVertexEdges(Index vertIndex) const { return\n+_vertEdgeCountsAndOffsets[2*vertIndex + 1]; }\n 436\n- 437inline ConstIndexArray\n- 438Refinement::getEdgeChildEdges(Index parentEdge) const {\n- 439\n- 440 return ConstIndexArray(&_edgeChildEdgeIndices[parentEdge*2], 2);\n- 441}\n- 442\n- 443inline IndexArray\n- 444Refinement::getEdgeChildEdges(Index parentEdge) {\n- 445\n- 446 return IndexArray(&_edgeChildEdgeIndices[parentEdge*2], 2);\n- 447}\n- 448\n- 449} // end namespace internal\n- 450} // end namespace Vtr\n- 451\n- 452} // end namespace OPENSUBDIV_VERSION\n- 453using namespace OPENSUBDIV_VERSION;\n- 454} // end namespace OpenSubdiv\n+ 437 ConstIndexArray getFaceVertices() const;\n+ 438\n+ 439 //\n+ 440 // Note that for some relations, the size of the relations for a child\n+component\n+ 441 // can vary radically from its parent due to the sparsity of the\n+refinement. So\n+ 442 // in these cases a few additional utilities are provided to help define\n+the set\n+ 443 // of incident components. Assuming adequate memory has been allocated,\n+the\n+ 444 // \"resize\" methods here initialize the set of incident components by\n+setting\n+ 445 // both the size and the appropriate offset, while \"trim\" is use to\n+quickly lower\n+ 446 // the size from an upper bound and nothing else.\n+ 447 //\n+ 448 void resizeFaceVertices(Index FaceIndex, int count);\n+ 449\n+ 450 void resizeEdgeFaces(Index edgeIndex, int count);\n+ 451 void trimEdgeFaces( Index edgeIndex, int count);\n+ 452\n+ 453 void resizeVertexFaces(Index vertIndex, int count);\n+ 454 void trimVertexFaces( Index vertIndex, int count);\n 455\n- 456#endif /* OPENSUBDIV3_VTR_REFINEMENT_H */\n+ 456 void resizeVertexEdges(Index vertIndex, int count);\n+ 457 void trimVertexEdges( Index vertIndex, int count);\n+ 458\n+ 459public:\n+ 460 //\n+ 461 // Initial plans were to have a few specific classes properly construct\n+the\n+ 462 // topology from scratch, e.g. the Refinement class and a Factory class\n+for\n+ 463 // the base level, by populating all topological relations. The need to\n+have\n+ 464 // a class construct full topology given only a simple face-vertex list,\n+made\n+ 465 // it necessary to write code to define and orient all relations -- and\n+most\n+ 466 // of that seemed best placed here.\n+ 467 //\n+ 468 bool completeTopologyFromFaceVertices();\n+ 469 Index findEdge(Index v0, Index v1, ConstIndexArray v0Edges) const;\n+ 470\n+ 471 // Methods supporting the above:\n+ 472 void orientIncidentComponents();\n+ 473 bool orderVertexFacesAndEdges(Index vIndex, Index* vFaces, Index* vEdges)\n+const;\n+ 474 bool orderVertexFacesAndEdges(Index vIndex);\n+ 475 void populateLocalIndices();\n+ 476\n+ 477 IndexArray shareFaceVertCountsAndOffsets() const;\n+ 478\n+ 479private:\n+ 480 // Refinement classes (including all subclasses) build a Level:\n+ 481 friend class Refinement;\n+ 482 friend class TriRefinement;\n+ 483 friend class QuadRefinement;\n+ 484\n+ 485 //\n+ 486 // A Level is independent of subdivision scheme or options. While it may\n+have been\n+ 487 // affected by them in its construction, they are not associated with it -\n+- a Level\n+ 488 // is pure topology and any subdivision parameters are external.\n+ 489 //\n+ 490\n+ 491 // Simple members for inventory, etc.\n+ 492 int _faceCount;\n+ 493 int _edgeCount;\n+ 494 int _vertCount;\n+ 495\n+ 496 // The \"depth\" member is clearly useful in both the topological splitting\n+and the\n+ 497 // stencil queries, but arguably it ties the Level to a hierarchy which\n+counters\n+ 498 // the idea of it being independent.\n+ 499 int _depth;\n+ 500\n+ 501 // Maxima to help clients manage sizing of data buffers. Given \"max\n+valence\",\n+ 502 // the \"max edge faces\" is strictly redundant as it will always be less,\n+but\n+ 503 // since it will typically be so much less (i.e. 2) it is kept for now.\n+ 504 int _maxEdgeFaces;\n+ 505 int _maxValence;\n+ 506\n+ 507 //\n+ 508 // Topology vectors:\n+ 509 // Note that of all of these, only data for the face-edge relation is not\n+ 510 // stored in the osd::FarTables in any form. The FarTable vectors combine\n+ 511 // the edge-vert and edge-face relations. The eventual goal is that this\n+ 512 // data be part of the osd::Far classes and be a superset of the FarTable\n+ 513 // vectors, i.e. no data duplication or conversion. The fact that FarTable\n+ 514 // already stores 5 of the 6 possible relations should make the topology\n+ 515 // storage as a whole a non-issue.\n+ 516 //\n+ 517 // The vert-face-child and vert-edge-child indices are also arguably not\n+ 518 // a topology relation but more one for parent/child relations. But it is\n+ 519 // a topological relationship, and if named differently would not likely\n+ 520 // raise this. It has been named with \"child\" in the name as it does play\n+ 521 // a more significant role during subdivision in mapping between parent\n+ 522 // and child components, and so has been named to reflect that more\n+clearly.\n+ 523 //\n+ 524\n+ 525 // Per-face:\n+ 526 std::vector _faceVertCountsAndOffsets; // 2 per face, redundant\n+after level 0\n+ 527 std::vector _faceVertIndices; // 3 or 4 per face, variable at level\n+0\n+ 528 std::vector _faceEdgeIndices; // matches face-vert indices\n+ 529 std::vector _faceTags; // 1 per face: includes \"hole\" tag\n+ 530\n+ 531 // Per-edge:\n+ 532 std::vector _edgeVertIndices; // 2 per edge\n+ 533 std::vector _edgeFaceCountsAndOffsets; // 2 per edge\n+ 534 std::vector _edgeFaceIndices; // varies with faces per edge\n+ 535 std::vector _edgeFaceLocalIndices; // varies with faces per\n+edge\n+ 536\n+ 537 std::vector _edgeSharpness; // 1 per edge\n+ 538 std::vector _edgeTags; // 1 per edge: manifold, boundary, etc.\n+ 539\n+ 540 // Per-vertex:\n+ 541 std::vector _vertFaceCountsAndOffsets; // 2 per vertex\n+ 542 std::vector _vertFaceIndices; // varies with valence\n+ 543 std::vector _vertFaceLocalIndices; // varies with valence, 8-\n+bit for now\n+ 544\n+ 545 std::vector _vertEdgeCountsAndOffsets; // 2 per vertex\n+ 546 std::vector _vertEdgeIndices; // varies with valence\n+ 547 std::vector _vertEdgeLocalIndices; // varies with valence, 8-\n+bit for now\n+ 548\n+ 549 std::vector _vertSharpness; // 1 per vertex\n+ 550 std::vector _vertTags; // 1 per vertex: manifold, Sdc::Rule, etc.\n+ 551\n+ 552 // Face-varying channels:\n+ 553 std::vector _fvarChannels;\n+ 554};\n+ 555\n+ 556//\n+ 557// Access/modify the vertices incident a given face:\n+ 558//\n+ 559inline ConstIndexArray\n+ 560Level::getFaceVertices(Index faceIndex) const {\n+ 561 return ConstIndexArray(&_faceVertIndices[_faceVertCountsAndOffsets\n+[faceIndex*2+1]],\n+ 562 _faceVertCountsAndOffsets[faceIndex*2]);\n+ 563}\n+ 564inline IndexArray\n+ 565Level::getFaceVertices(Index faceIndex) {\n+ 566 return IndexArray(&_faceVertIndices[_faceVertCountsAndOffsets\n+[faceIndex*2+1]],\n+ 567 _faceVertCountsAndOffsets[faceIndex*2]);\n+ 568}\n+ 569\n+ 570inline void\n+ 571Level::resizeFaceVertices(Index faceIndex, int count) {\n+ 572\n+ 573 int* countOffsetPair = &_faceVertCountsAndOffsets[faceIndex*2];\n+ 574\n+ 575 countOffsetPair[0] = count;\n+ 576 countOffsetPair[1] = (faceIndex == 0) ? 0 : (countOffsetPair[-2] +\n+countOffsetPair[-1]);\n+ 577\n+ 578 _maxValence = std::max(_maxValence, count);\n+ 579}\n+ 580\n+ 581inline ConstIndexArray\n+ 582Level::getFaceVertices() const {\n+ 583 return ConstIndexArray(&_faceVertIndices[0], (int)_faceVertIndices.size\n+());\n+ 584}\n+ 585\n+ 586//\n+ 587// Access/modify the edges incident a given face:\n+ 588//\n+ 589inline ConstIndexArray\n+ 590Level::getFaceEdges(Index faceIndex) const {\n+ 591 return ConstIndexArray(&_faceEdgeIndices[_faceVertCountsAndOffsets\n+[faceIndex*2+1]],\n+ 592 _faceVertCountsAndOffsets[faceIndex*2]);\n+ 593}\n+ 594inline IndexArray\n+ 595Level::getFaceEdges(Index faceIndex) {\n+ 596 return IndexArray(&_faceEdgeIndices[_faceVertCountsAndOffsets\n+[faceIndex*2+1]],\n+ 597 _faceVertCountsAndOffsets[faceIndex*2]);\n+ 598}\n+ 599\n+ 600//\n+ 601// Access/modify the faces incident a given vertex:\n+ 602//\n+ 603inline ConstIndexArray\n+ 604Level::getVertexFaces(Index vertIndex) const {\n+ 605 return ConstIndexArray( (&_vertFaceIndices[0]) + _vertFaceCountsAndOffsets\n+[vertIndex*2+1],\n+ 606 _vertFaceCountsAndOffsets[vertIndex*2]);\n+ 607}\n+ 608inline IndexArray\n+ 609Level::getVertexFaces(Index vertIndex) {\n+ 610 return IndexArray( (&_vertFaceIndices[0]) + _vertFaceCountsAndOffsets\n+[vertIndex*2+1],\n+ 611 _vertFaceCountsAndOffsets[vertIndex*2]);\n+ 612}\n+ 613\n+ 614inline ConstLocalIndexArray\n+ 615Level::getVertexFaceLocalIndices(Index vertIndex) const {\n+ 616 return ConstLocalIndexArray( (&_vertFaceLocalIndices[0]) +\n+_vertFaceCountsAndOffsets[vertIndex*2+1],\n+ 617 _vertFaceCountsAndOffsets[vertIndex*2]);\n+ 618}\n+ 619inline LocalIndexArray\n+ 620Level::getVertexFaceLocalIndices(Index vertIndex) {\n+ 621 return LocalIndexArray( (&_vertFaceLocalIndices[0]) +\n+_vertFaceCountsAndOffsets[vertIndex*2+1],\n+ 622 _vertFaceCountsAndOffsets[vertIndex*2]);\n+ 623}\n+ 624\n+ 625inline void\n+ 626Level::resizeVertexFaces(Index vertIndex, int count) {\n+ 627 int* countOffsetPair = &_vertFaceCountsAndOffsets[vertIndex*2];\n+ 628\n+ 629 countOffsetPair[0] = count;\n+ 630 countOffsetPair[1] = (vertIndex == 0) ? 0 : (countOffsetPair[-2] +\n+countOffsetPair[-1]);\n+ 631}\n+ 632inline void\n+ 633Level::trimVertexFaces(Index vertIndex, int count) {\n+ 634 _vertFaceCountsAndOffsets[vertIndex*2] = count;\n+ 635}\n+ 636\n+ 637//\n+ 638// Access/modify the edges incident a given vertex:\n+ 639//\n+ 640inline ConstIndexArray\n+ 641Level::getVertexEdges(Index vertIndex) const {\n+ 642 return ConstIndexArray( (&_vertEdgeIndices[0]) +_vertEdgeCountsAndOffsets\n+[vertIndex*2+1],\n+ 643 _vertEdgeCountsAndOffsets[vertIndex*2]);\n+ 644}\n+ 645inline IndexArray\n+ 646Level::getVertexEdges(Index vertIndex) {\n+ 647 return IndexArray( (&_vertEdgeIndices[0]) +_vertEdgeCountsAndOffsets\n+[vertIndex*2+1],\n+ 648 _vertEdgeCountsAndOffsets[vertIndex*2]);\n+ 649}\n+ 650\n+ 651inline ConstLocalIndexArray\n+ 652Level::getVertexEdgeLocalIndices(Index vertIndex) const {\n+ 653 return ConstLocalIndexArray( (&_vertEdgeLocalIndices[0]) +\n+_vertEdgeCountsAndOffsets[vertIndex*2+1],\n+ 654 _vertEdgeCountsAndOffsets[vertIndex*2]);\n+ 655}\n+ 656inline LocalIndexArray\n+ 657Level::getVertexEdgeLocalIndices(Index vertIndex) {\n+ 658 return LocalIndexArray( (&_vertEdgeLocalIndices[0]) +\n+_vertEdgeCountsAndOffsets[vertIndex*2+1],\n+ 659 _vertEdgeCountsAndOffsets[vertIndex*2]);\n+ 660}\n+ 661\n+ 662inline void\n+ 663Level::resizeVertexEdges(Index vertIndex, int count) {\n+ 664 int* countOffsetPair = &_vertEdgeCountsAndOffsets[vertIndex*2];\n+ 665\n+ 666 countOffsetPair[0] = count;\n+ 667 countOffsetPair[1] = (vertIndex == 0) ? 0 : (countOffsetPair[-2] +\n+countOffsetPair[-1]);\n+ 668\n+ 669 _maxValence = std::max(_maxValence, count);\n+ 670}\n+ 671inline void\n+ 672Level::trimVertexEdges(Index vertIndex, int count) {\n+ 673 _vertEdgeCountsAndOffsets[vertIndex*2] = count;\n+ 674}\n+ 675\n+ 676inline void\n+ 677Level::setMaxValence(int valence) {\n+ 678 _maxValence = valence;\n+ 679}\n+ 680\n+ 681//\n+ 682// Access/modify the vertices incident a given edge:\n+ 683//\n+ 684inline ConstIndexArray\n+ 685Level::getEdgeVertices(Index edgeIndex) const {\n+ 686 return ConstIndexArray(&_edgeVertIndices[edgeIndex*2], 2);\n+ 687}\n+ 688inline IndexArray\n+ 689Level::getEdgeVertices(Index edgeIndex) {\n+ 690 return IndexArray(&_edgeVertIndices[edgeIndex*2], 2);\n+ 691}\n+ 692\n+ 693//\n+ 694// Access/modify the faces incident a given edge:\n+ 695//\n+ 696inline ConstIndexArray\n+ 697Level::getEdgeFaces(Index edgeIndex) const {\n+ 698 return ConstIndexArray(&_edgeFaceIndices[0] +\n+ 699 _edgeFaceCountsAndOffsets[edgeIndex*2+1],\n+ 700 _edgeFaceCountsAndOffsets[edgeIndex*2]);\n+ 701}\n+ 702inline IndexArray\n+ 703Level::getEdgeFaces(Index edgeIndex) {\n+ 704 return IndexArray(&_edgeFaceIndices[0] +\n+ 705 _edgeFaceCountsAndOffsets[edgeIndex*2+1],\n+ 706 _edgeFaceCountsAndOffsets[edgeIndex*2]);\n+ 707}\n+ 708\n+ 709inline ConstLocalIndexArray\n+ 710Level::getEdgeFaceLocalIndices(Index edgeIndex) const {\n+ 711 return ConstLocalIndexArray(&_edgeFaceLocalIndices[0] +\n+ 712 _edgeFaceCountsAndOffsets[edgeIndex*2+1],\n+ 713 _edgeFaceCountsAndOffsets[edgeIndex*2]);\n+ 714}\n+ 715inline LocalIndexArray\n+ 716Level::getEdgeFaceLocalIndices(Index edgeIndex) {\n+ 717 return LocalIndexArray(&_edgeFaceLocalIndices[0] +\n+ 718 _edgeFaceCountsAndOffsets[edgeIndex*2+1],\n+ 719 _edgeFaceCountsAndOffsets[edgeIndex*2]);\n+ 720}\n+ 721\n+ 722inline void\n+ 723Level::resizeEdgeFaces(Index edgeIndex, int count) {\n+ 724 int* countOffsetPair = &_edgeFaceCountsAndOffsets[edgeIndex*2];\n+ 725\n+ 726 countOffsetPair[0] = count;\n+ 727 countOffsetPair[1] = (edgeIndex == 0) ? 0 : (countOffsetPair[-2] +\n+countOffsetPair[-1]);\n+ 728\n+ 729 _maxEdgeFaces = std::max(_maxEdgeFaces, count);\n+ 730}\n+ 731inline void\n+ 732Level::trimEdgeFaces(Index edgeIndex, int count) {\n+ 733 _edgeFaceCountsAndOffsets[edgeIndex*2] = count;\n+ 734}\n+ 735\n+ 736//\n+ 737// Access/modify sharpness values:\n+ 738//\n+ 739inline float\n+ 740Level::getEdgeSharpness(Index edgeIndex) const {\n+ 741 return _edgeSharpness[edgeIndex];\n+ 742}\n+ 743inline float&\n+ 744Level::getEdgeSharpness(Index edgeIndex) {\n+ 745 return _edgeSharpness[edgeIndex];\n+ 746}\n+ 747\n+ 748inline float\n+ 749Level::getVertexSharpness(Index vertIndex) const {\n+ 750 return _vertSharpness[vertIndex];\n+ 751}\n+ 752inline float&\n+ 753Level::getVertexSharpness(Index vertIndex) {\n+ 754 return _vertSharpness[vertIndex];\n+ 755}\n+ 756\n+ 757inline Sdc::Crease::Rule\n+ 758Level::getVertexRule(Index vertIndex) const {\n+ 759 return (Sdc::Crease::Rule) _vertTags[vertIndex]._rule;\n+ 760}\n+ 761\n+ 762//\n+ 763// Access/modify hole tag:\n+ 764//\n+ 765inline void\n+ 766Level::setFaceHole(Index faceIndex, bool b) {\n+ 767 _faceTags[faceIndex]._hole = b;\n+ 768}\n+ 769inline bool\n+ 770Level::isFaceHole(Index faceIndex) const {\n+ 771 return _faceTags[faceIndex]._hole;\n+ 772}\n+ 773\n+ 774//\n+ 775// Access/modify non-manifold tags:\n+ 776//\n+ 777inline void\n+ 778Level::setEdgeNonManifold(Index edgeIndex, bool b) {\n+ 779 _edgeTags[edgeIndex]._nonManifold = b;\n+ 780}\n+ 781inline bool\n+ 782Level::isEdgeNonManifold(Index edgeIndex) const {\n+ 783 return _edgeTags[edgeIndex]._nonManifold;\n+ 784}\n+ 785\n+ 786inline void\n+ 787Level::setVertexNonManifold(Index vertIndex, bool b) {\n+ 788 _vertTags[vertIndex]._nonManifold = b;\n+ 789}\n+ 790inline bool\n+ 791Level::isVertexNonManifold(Index vertIndex) const {\n+ 792 return _vertTags[vertIndex]._nonManifold;\n+ 793}\n+ 794\n+ 795//\n+ 796// Sizing methods to allocate space:\n+ 797//\n+ 798inline void\n+ 799Level::resizeFaces(int faceCount) {\n+ 800 _faceCount = faceCount;\n+ 801 _faceVertCountsAndOffsets.resize(2 * faceCount);\n+ 802\n+ 803 _faceTags.resize(faceCount);\n+ 804 std::memset((void*) &_faceTags[0], 0, _faceCount * sizeof(FTag));\n+ 805}\n+ 806inline void\n+ 807Level::resizeFaceVertices(int totalFaceVertCount) {\n+ 808 _faceVertIndices.resize(totalFaceVertCount);\n+ 809}\n+ 810inline void\n+ 811Level::resizeFaceEdges(int totalFaceEdgeCount) {\n+ 812 _faceEdgeIndices.resize(totalFaceEdgeCount);\n+ 813}\n+ 814\n+ 815inline void\n+ 816Level::resizeEdges(int edgeCount) {\n+ 817\n+ 818 _edgeCount = edgeCount;\n+ 819 _edgeFaceCountsAndOffsets.resize(2 * edgeCount);\n+ 820\n+ 821 _edgeSharpness.resize(edgeCount);\n+ 822 _edgeTags.resize(edgeCount);\n+ 823\n+ 824 if (edgeCount>0) {\n+ 825 std::memset((void*) &_edgeTags[0], 0, _edgeCount * sizeof(ETag));\n+ 826 }\n+ 827}\n+ 828inline void\n+ 829Level::resizeEdgeVertices() {\n+ 830\n+ 831 _edgeVertIndices.resize(2 * _edgeCount);\n+ 832}\n+ 833inline void\n+ 834Level::resizeEdgeFaces(int totalEdgeFaceCount) {\n+ 835\n+ 836 _edgeFaceIndices.resize(totalEdgeFaceCount);\n+ 837 _edgeFaceLocalIndices.resize(totalEdgeFaceCount);\n+ 838}\n+ 839\n+ 840inline void\n+ 841Level::resizeVertices(int vertCount) {\n+ 842\n+ 843 _vertCount = vertCount;\n+ 844 _vertFaceCountsAndOffsets.resize(2 * vertCount);\n+ 845 _vertEdgeCountsAndOffsets.resize(2 * vertCount);\n+ 846\n+ 847 _vertSharpness.resize(vertCount);\n+ 848 _vertTags.resize(vertCount);\n+ 849 std::memset((void*) &_vertTags[0], 0, _vertCount * sizeof(VTag));\n+ 850}\n+ 851inline void\n+ 852Level::resizeVertexFaces(int totalVertFaceCount) {\n+ 853\n+ 854 _vertFaceIndices.resize(totalVertFaceCount);\n+ 855 _vertFaceLocalIndices.resize(totalVertFaceCount);\n+ 856}\n+ 857inline void\n+ 858Level::resizeVertexEdges(int totalVertEdgeCount) {\n+ 859\n+ 860 _vertEdgeIndices.resize(totalVertEdgeCount);\n+ 861 _vertEdgeLocalIndices.resize(totalVertEdgeCount);\n+ 862}\n+ 863\n+ 864inline IndexArray\n+ 865Level::shareFaceVertCountsAndOffsets() const {\n+ 866 // XXXX manuelk we have to force const casting here (classes don't 'share'\n+ 867 // members usually...)\n+ 868 return IndexArray(const_cast(&_faceVertCountsAndOffsets[0]),\n+ 869 (int)_faceVertCountsAndOffsets.size());\n+ 870}\n+ 871\n+ 872} // end namespace internal\n+ 873} // end namespace Vtr\n+ 874\n+ 875} // end namespace OPENSUBDIV_VERSION\n+ 876using namespace OPENSUBDIV_VERSION;\n+ 877} // end namespace OpenSubdiv\n+ 878\n+ 879#endif /* OPENSUBDIV3_VTR_LEVEL_H */\n OpenSubdiv\n Definition: limits.h:32\n+OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstLocalIndexArray\n+ConstArray< LocalIndex > ConstLocalIndexArray\n+Definition: types.h:83\n+OpenSubdiv::OPENSUBDIV_VERSION::Vtr::LocalIndexArray\n+Array< LocalIndex > LocalIndexArray\n+Definition: types.h:82\n OpenSubdiv::OPENSUBDIV_VERSION::Vtr::Index\n int Index\n Definition: types.h:54\n-OpenSubdiv::OPENSUBDIV_VERSION::Vtr::IndexVector\n-std::vector< Index > IndexVector\n-Definition: types.h:77\n OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstIndexArray\n ConstArray< Index > ConstIndexArray\n Definition: types.h:80\n+OpenSubdiv::OPENSUBDIV_VERSION::Vtr::LocalIndex\n+unsigned short LocalIndex\n+Definition: types.h:66\n OpenSubdiv::OPENSUBDIV_VERSION::Vtr::IndexArray\n Array< Index > IndexArray\n Definition: types.h:79\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Split\n-Split\n-Enumerated type for all face splitting schemes.\n-Definition: types.h:47\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::Rule\n+Rule\n+Definition: crease.h:82\n * opensubdiv\n * vtr\n- * refinement.h\n+ * level.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00758.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00758.html", "unified_diff": "@@ -89,15 +89,15 @@\n
fvarRefinement.h File Reference
\n \n
\n
#include "../version.h"
\n #include "../sdc/types.h"
\n #include "../sdc/crease.h"
\n #include "../vtr/types.h"
\n-#include "../vtr/refinement.h"
\n+#include "../vtr/refinement.h"
\n #include "../vtr/fvarLevel.h"
\n #include <vector>
\n #include <cassert>
\n #include <cstring>
\n
\n

Go to the source code of this file.

\n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00761.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00761.html", "unified_diff": "@@ -90,15 +90,15 @@\n \n
\n
#include "../version.h"
\n #include "../sdc/types.h"
\n #include "../sdc/crease.h"
\n #include "../sdc/options.h"
\n #include "../vtr/types.h"
\n-#include "../vtr/level.h"
\n+#include "../vtr/level.h"
\n #include <vector>
\n #include <cassert>
\n #include <cstring>
\n
\n

Go to the source code of this file.

\n
\n

\n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00782.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00782.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/sdc/bilinearScheme.h File Reference\n+OpenSubdiv: opensubdiv/sdc/catmarkScheme.h File Reference\n \n \n \n \n \n \n \n@@ -82,19 +82,21 @@\n name=\"MSearchResults\" id=\"MSearchResults\">\n \n \n \n
\n
\n Namespaces
\n-
bilinearScheme.h File Reference
\n+
catmarkScheme.h File Reference
\n
\n
\n
#include "../version.h"
\n #include "../sdc/scheme.h"
\n+#include <cassert>
\n+#include <cmath>
\n
\n

Go to the source code of this file.

\n \n \n \n \n@@ -104,13 +106,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,22 +5,24 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Namespaces\n-bilinearScheme.h File Reference\n+catmarkScheme.h File Reference\n #include \"../version.h\"\n #include \"../sdc/scheme.h\"\n+#include \n+#include \n Go_to_the_source_code_of_this_file.\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Sdc\n \u00a0\n * opensubdiv\n * sdc\n- * bilinearScheme.h\n+ * catmarkScheme.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00782_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00782_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/sdc/bilinearScheme.h Source File\n+OpenSubdiv: opensubdiv/sdc/catmarkScheme.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
bilinearScheme.h
\n+
catmarkScheme.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2014 DreamWorks Animation LLC.
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -106,175 +106,548 @@\n
17//
\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n-
24#ifndef OPENSUBDIV3_SDC_BILINEAR_SCHEME_H
\n-
25#define OPENSUBDIV3_SDC_BILINEAR_SCHEME_H
\n+
24#ifndef OPENSUBDIV3_SDC_CATMARK_SCHEME_H
\n+
25#define OPENSUBDIV3_SDC_CATMARK_SCHEME_H
\n
26
\n
27#include "../version.h"
\n
28
\n
29#include "../sdc/scheme.h"
\n
30
\n-
31namespace OpenSubdiv {
\n-
32namespace OPENSUBDIV_VERSION {
\n+
31#include <cassert>
\n+
32#include <cmath>
\n
33
\n-
34namespace Sdc {
\n-
35
\n-
36//
\n-
37// Specializations for Scheme<SCHEME_BILINEAR>:
\n-
38//
\n-
39
\n-
40//
\n-
41// Bilinear traits:
\n-
42//
\n-
43template <>
\n-
44inline Split Scheme<SCHEME_BILINEAR>::GetTopologicalSplitType() { return SPLIT_TO_QUADS; }
\n-
45
\n+
34namespace OpenSubdiv {
\n+
35namespace OPENSUBDIV_VERSION {
\n+
36
\n+
37namespace Sdc {
\n+
38
\n+
39//
\n+
40// Specializations for Scheme<SCHEME_CATMARK>:
\n+
41//
\n+
42
\n+
43//
\n+
44// Catmark traits:
\n+
45//
\n
46template <>
\n-
47inline int Scheme<SCHEME_BILINEAR>::GetRegularFaceSize() { return 4; }
\n+
47inline Split Scheme<SCHEME_CATMARK>::GetTopologicalSplitType() { return SPLIT_TO_QUADS; }
\n
48
\n
49template <>
\n-
50inline int Scheme<SCHEME_BILINEAR>::GetRegularVertexValence() { return 4; }
\n+
50inline int Scheme<SCHEME_CATMARK>::GetRegularFaceSize() { return 4; }
\n
51
\n
52template <>
\n-
53inline int Scheme<SCHEME_BILINEAR>::GetLocalNeighborhoodSize() { return 0; }
\n+
53inline int Scheme<SCHEME_CATMARK>::GetRegularVertexValence() { return 4; }
\n
54
\n-
55
\n-
56//
\n-
57// Refinement masks:
\n-
58//
\n-
59template <>
\n-
60template <typename EDGE, typename MASK>
\n-
61void
\n-
62Scheme<SCHEME_BILINEAR>::ComputeEdgeVertexMask(EDGE const& edge, MASK& mask,
\n-
63 Crease::Rule, Crease::Rule) const {
\n-
64 // This should be inline, otherwise trivially replicate it:
\n-
65 assignCreaseMaskForEdge(edge, mask);
\n-
66}
\n-
67
\n+
55template <>
\n+
56inline int Scheme<SCHEME_CATMARK>::GetLocalNeighborhoodSize() { return 1; }
\n+
57
\n+
58
\n+
59//
\n+
60// Masks for edge-vertices: the hard Crease mask does not need to be specialized
\n+
61// (simply the midpoint), so all that is left is the Smooth case:
\n+
62//
\n+
63// The Smooth mask is complicated by the need to support the "triangle subdivision"
\n+
64// option, which applies different weighting in the presence of triangles. It is
\n+
65// up for debate as to whether this is useful or not -- we may be able to deprecate
\n+
66// this option.
\n+
67//
\n
68template <>
\n-
69template <typename VERTEX, typename MASK>
\n-
70void
\n-
71Scheme<SCHEME_BILINEAR>::ComputeVertexVertexMask(VERTEX const& vertex, MASK& mask,
\n-
72 Crease::Rule, Crease::Rule) const {
\n-
73 // This should be inline, otherwise trivially replicate it:
\n-
74 assignCornerMaskForVertex(vertex, mask);
\n-
75}
\n+
69template <typename EDGE, typename MASK>
\n+
70inline void
\n+
71Scheme<SCHEME_CATMARK>::assignSmoothMaskForEdge(EDGE const& edge, MASK& mask) const {
\n+
72
\n+
73 typedef typename MASK::Weight Weight;
\n+
74
\n+
75 int faceCount = edge.GetNumFaces();
\n
76
\n-
77
\n-
78//
\n-
79// Limit masks for position -- the limit position of all vertices is the refined vertex.
\n-
80//
\n-
81template <>
\n-
82template <typename VERTEX, typename MASK>
\n-
83inline void
\n-
84Scheme<SCHEME_BILINEAR>::assignCornerLimitMask(VERTEX const& /* vertex */, MASK& posMask) const {
\n-
85
\n-
86 posMask.SetNumVertexWeights(1);
\n-
87 posMask.SetNumEdgeWeights(0);
\n-
88 posMask.SetNumFaceWeights(0);
\n-
89 posMask.SetFaceWeightsForFaceCenters(false);
\n-
90
\n-
91 posMask.VertexWeight(0) = 1.0f;
\n-
92}
\n-
93
\n-
94template <>
\n-
95template <typename VERTEX, typename MASK>
\n-
96inline void
\n-
97Scheme<SCHEME_BILINEAR>::assignCreaseLimitMask(VERTEX const& vertex, MASK& posMask,
\n-
98 int const /* creaseEnds */[2]) const {
\n-
99
\n-
100 assignCornerLimitMask(vertex, posMask);
\n-
101}
\n-
102
\n-
103template <>
\n-
104template <typename VERTEX, typename MASK>
\n-
105inline void
\n-
106Scheme<SCHEME_BILINEAR>::assignSmoothLimitMask(VERTEX const& vertex, MASK& posMask) const {
\n-
107
\n-
108 assignCornerLimitMask(vertex, posMask);
\n-
109}
\n+
77 mask.SetNumVertexWeights(2);
\n+
78 mask.SetNumEdgeWeights(0);
\n+
79 mask.SetNumFaceWeights(faceCount);
\n+
80 mask.SetFaceWeightsForFaceCenters(true);
\n+
81
\n+
82 //
\n+
83 // Determine if we need to inspect incident faces and apply alternate weighting for
\n+
84 // triangles -- and if so, determine which of the two are triangles.
\n+
85 //
\n+
86 bool face0IsTri = false;
\n+
87 bool face1IsTri = false;
\n+
88 bool useTriangleOption = (_options.GetTriangleSubdivision() == Options::TRI_SUB_SMOOTH);
\n+
89 if (useTriangleOption) {
\n+
90 if (faceCount == 2) {
\n+
91 //
\n+
92 // Ideally we want to avoid this inspection when we have already subdivided at
\n+
93 // least once -- need something in the Edge interface to help avoid this, e.g.
\n+
94 // an IsRegular() query, the subdivision level...
\n+
95 //
\n+
96 int vertsPerFace[2];
\n+
97 edge.GetNumVerticesPerFace(vertsPerFace);
\n+
98
\n+
99 face0IsTri = (vertsPerFace[0] == 3);
\n+
100 face1IsTri = (vertsPerFace[1] == 3);
\n+
101 useTriangleOption = face0IsTri || face1IsTri;
\n+
102 } else {
\n+
103 useTriangleOption = false;
\n+
104 }
\n+
105 }
\n+
106
\n+
107 if (! useTriangleOption) {
\n+
108 mask.VertexWeight(0) = 0.25f;
\n+
109 mask.VertexWeight(1) = 0.25f;
\n
110
\n-
111//
\n-
112// Limit masks for tangents -- these are ambiguous around all vertices. Provide
\n-
113// the tangents based on the incident edges of the first face.
\n-
114//
\n-
115template <>
\n-
116template <typename VERTEX, typename MASK>
\n-
117inline void
\n-
118Scheme<SCHEME_BILINEAR>::assignCornerLimitTangentMasks(VERTEX const& /* vertex */,
\n-
119 MASK& tan1Mask, MASK& tan2Mask) const {
\n-
120
\n-
121 tan1Mask.SetNumVertexWeights(1);
\n-
122 tan1Mask.SetNumEdgeWeights(2);
\n-
123 tan1Mask.SetNumFaceWeights(0);
\n-
124 tan1Mask.SetFaceWeightsForFaceCenters(false);
\n+
111 if (faceCount == 2) {
\n+
112 mask.FaceWeight(0) = 0.25f;
\n+
113 mask.FaceWeight(1) = 0.25f;
\n+
114 } else {
\n+
115 Weight fWeight = 0.5f / (Weight)faceCount;
\n+
116 for (int i = 0; i < faceCount; ++i) {
\n+
117 mask.FaceWeight(i) = fWeight;
\n+
118 }
\n+
119 }
\n+
120 } else {
\n+
121 //
\n+
122 // This mimics the implementation in Hbr in terms of order of operations.
\n+
123 //
\n+
124 const Weight CATMARK_SMOOTH_TRI_EDGE_WEIGHT = (Weight) 0.470;
\n
125
\n-
126 tan2Mask.SetNumVertexWeights(1);
\n-
127 tan2Mask.SetNumEdgeWeights(2);
\n-
128 tan2Mask.SetNumFaceWeights(0);
\n-
129 tan2Mask.SetFaceWeightsForFaceCenters(false);
\n-
130
\n-
131 tan1Mask.VertexWeight(0) = -1.0f;
\n-
132 tan1Mask.EdgeWeight(0) = 1.0f;
\n-
133 tan1Mask.EdgeWeight(1) = 0.0f;
\n+
126 Weight f0Weight = face0IsTri ? CATMARK_SMOOTH_TRI_EDGE_WEIGHT : 0.25f;
\n+
127 Weight f1Weight = face1IsTri ? CATMARK_SMOOTH_TRI_EDGE_WEIGHT : 0.25f;
\n+
128
\n+
129 Weight fWeight = 0.5f * (f0Weight + f1Weight);
\n+
130 Weight vWeight = 0.5f * (1.0f - 2.0f * fWeight);
\n+
131
\n+
132 mask.VertexWeight(0) = vWeight;
\n+
133 mask.VertexWeight(1) = vWeight;
\n
134
\n-
135 tan2Mask.VertexWeight(0) = -1.0f;
\n-
136 tan2Mask.EdgeWeight(0) = 0.0f;
\n-
137 tan2Mask.EdgeWeight(1) = 1.0f;
\n+
135 mask.FaceWeight(0) = fWeight;
\n+
136 mask.FaceWeight(1) = fWeight;
\n+
137 }
\n
138}
\n
139
\n-
140template <>
\n-
141template <typename VERTEX, typename MASK>
\n-
142inline void
\n-
143Scheme<SCHEME_BILINEAR>::assignCreaseLimitTangentMasks(VERTEX const& vertex,
\n-
144 MASK& tan1Mask, MASK& tan2Mask, int const /* creaseEnds */[2]) const {
\n-
145
\n-
146 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);
\n-
147}
\n-
148
\n-
149template <>
\n-
150template <typename VERTEX, typename MASK>
\n-
151inline void
\n-
152Scheme<SCHEME_BILINEAR>::assignSmoothLimitTangentMasks(VERTEX const& vertex,
\n-
153 MASK& tan1Mask, MASK& tan2Mask) const {
\n-
154
\n-
155 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);
\n-
156}
\n-
157
\n-
158} // end namespace sdc
\n-
159
\n-
160} // end namespace OPENSUBDIV_VERSION
\n-
161using namespace OPENSUBDIV_VERSION;
\n-
162} // end namespace OpenSubdiv
\n-
163
\n-
164#endif /* OPENSUBDIV3_SDC_BILINEAR_SCHEME_H */
\n+
140
\n+
141//
\n+
142// Masks for vertex-vertices: the hard Corner mask does not need to be specialized
\n+
143// (simply the vertex itself), leaving the Crease and Smooth cases (Dart is smooth):
\n+
144//
\n+
145template <>
\n+
146template <typename VERTEX, typename MASK>
\n+
147inline void
\n+
148Scheme<SCHEME_CATMARK>::assignCreaseMaskForVertex(VERTEX const& vertex, MASK& mask,
\n+
149 int const creaseEnds[2]) const {
\n+
150 typedef typename MASK::Weight Weight;
\n+
151
\n+
152 int valence = vertex.GetNumEdges();
\n+
153
\n+
154 mask.SetNumVertexWeights(1);
\n+
155 mask.SetNumEdgeWeights(valence);
\n+
156 mask.SetNumFaceWeights(0);
\n+
157 mask.SetFaceWeightsForFaceCenters(false);
\n+
158
\n+
159 Weight vWeight = 0.75f;
\n+
160 Weight eWeight = 0.125f;
\n+
161
\n+
162 mask.VertexWeight(0) = vWeight;
\n+
163 for (int i = 0; i < valence; ++i) {
\n+
164 mask.EdgeWeight(i) = 0.0f;
\n+
165 }
\n+
166 mask.EdgeWeight(creaseEnds[0]) = eWeight;
\n+
167 mask.EdgeWeight(creaseEnds[1]) = eWeight;
\n+
168}
\n+
169
\n+
170template <>
\n+
171template <typename VERTEX, typename MASK>
\n+
172inline void
\n+
173Scheme<SCHEME_CATMARK>::assignSmoothMaskForVertex(VERTEX const& vertex, MASK& mask) const {
\n+
174
\n+
175 typedef typename MASK::Weight Weight;
\n+
176
\n+
177 //
\n+
178 // A Smooth vertex must be manifold and interior -- manifold boundary vertices will be
\n+
179 // Creases and non-manifold vertices of any kind will be Corners or Creases. If smooth
\n+
180 // rules for non-manifold vertices are ever defined, this will need adjusting:
\n+
181 //
\n+
182 assert(vertex.GetNumFaces() == vertex.GetNumEdges());
\n+
183
\n+
184 int valence = vertex.GetNumFaces();
\n+
185
\n+
186 mask.SetNumVertexWeights(1);
\n+
187 mask.SetNumEdgeWeights(valence);
\n+
188 mask.SetNumFaceWeights(valence);
\n+
189 mask.SetFaceWeightsForFaceCenters(true);
\n+
190
\n+
191 Weight vWeight = (Weight)(valence - 2) / (Weight)valence;
\n+
192 Weight fWeight = 1.0f / (Weight)(valence * valence);
\n+
193 Weight eWeight = fWeight;
\n+
194
\n+
195 mask.VertexWeight(0) = vWeight;
\n+
196 for (int i = 0; i < valence; ++i) {
\n+
197 mask.EdgeWeight(i) = eWeight;
\n+
198 mask.FaceWeight(i) = fWeight;
\n+
199 }
\n+
200}
\n+
201
\n+
202//
\n+
203// Limit masks for position:
\n+
204//
\n+
205template <>
\n+
206template <typename VERTEX, typename MASK>
\n+
207inline void
\n+
208Scheme<SCHEME_CATMARK>::assignCornerLimitMask(VERTEX const& /* vertex */, MASK& posMask) const {
\n+
209
\n+
210 posMask.SetNumVertexWeights(1);
\n+
211 posMask.SetNumEdgeWeights(0);
\n+
212 posMask.SetNumFaceWeights(0);
\n+
213 posMask.SetFaceWeightsForFaceCenters(false);
\n+
214
\n+
215 posMask.VertexWeight(0) = 1.0f;
\n+
216}
\n+
217
\n+
218template <>
\n+
219template <typename VERTEX, typename MASK>
\n+
220inline void
\n+
221Scheme<SCHEME_CATMARK>::assignCreaseLimitMask(VERTEX const& vertex, MASK& posMask,
\n+
222 int const creaseEnds[2]) const {
\n+
223
\n+
224 typedef typename MASK::Weight Weight;
\n+
225
\n+
226 int valence = vertex.GetNumEdges();
\n+
227
\n+
228 posMask.SetNumVertexWeights(1);
\n+
229 posMask.SetNumEdgeWeights(valence);
\n+
230 posMask.SetNumFaceWeights(0);
\n+
231 posMask.SetFaceWeightsForFaceCenters(false);
\n+
232
\n+
233 Weight vWeight = (Weight)(2.0 / 3.0);
\n+
234 Weight eWeight = (Weight)(1.0 / 6.0);
\n+
235
\n+
236 posMask.VertexWeight(0) = vWeight;
\n+
237 for (int i = 0; i < valence; ++i) {
\n+
238 posMask.EdgeWeight(i) = 0.0f;
\n+
239 }
\n+
240 posMask.EdgeWeight(creaseEnds[0]) = eWeight;
\n+
241 posMask.EdgeWeight(creaseEnds[1]) = eWeight;
\n+
242}
\n+
243
\n+
244template <>
\n+
245template <typename VERTEX, typename MASK>
\n+
246inline void
\n+
247Scheme<SCHEME_CATMARK>::assignSmoothLimitMask(VERTEX const& vertex, MASK& posMask) const {
\n+
248
\n+
249 typedef typename MASK::Weight Weight;
\n+
250
\n+
251 int valence = vertex.GetNumFaces();
\n+
252 if (valence == 2) {
\n+
253 assignCornerLimitMask(vertex, posMask);
\n+
254 return;
\n+
255 }
\n+
256
\n+
257 posMask.SetNumVertexWeights(1);
\n+
258 posMask.SetNumEdgeWeights(valence);
\n+
259 posMask.SetNumFaceWeights(valence);
\n+
260 posMask.SetFaceWeightsForFaceCenters(false);
\n+
261
\n+
262 // Specialize for the regular case:
\n+
263 if (valence == 4) {
\n+
264 Weight fWeight = (Weight)(1.0 / 36.0);
\n+
265 Weight eWeight = (Weight)(1.0 / 9.0);
\n+
266 Weight vWeight = (Weight)(4.0 / 9.0);
\n+
267
\n+
268 posMask.VertexWeight(0) = vWeight;
\n+
269
\n+
270 posMask.EdgeWeight(0) = eWeight;
\n+
271 posMask.EdgeWeight(1) = eWeight;
\n+
272 posMask.EdgeWeight(2) = eWeight;
\n+
273 posMask.EdgeWeight(3) = eWeight;
\n+
274
\n+
275 posMask.FaceWeight(0) = fWeight;
\n+
276 posMask.FaceWeight(1) = fWeight;
\n+
277 posMask.FaceWeight(2) = fWeight;
\n+
278 posMask.FaceWeight(3) = fWeight;
\n+
279 } else {
\n+
280 Weight Valence = (Weight) valence;
\n+
281
\n+
282 Weight fWeight = 1.0f / (Valence * (Valence + 5.0f));
\n+
283 Weight eWeight = 4.0f * fWeight;
\n+
284 Weight vWeight = 1.0f - Valence * (eWeight + fWeight);
\n+
285
\n+
286 posMask.VertexWeight(0) = vWeight;
\n+
287 for (int i = 0; i < valence; ++i) {
\n+
288 posMask.EdgeWeight(i) = eWeight;
\n+
289 posMask.FaceWeight(i) = fWeight;
\n+
290 }
\n+
291 }
\n+
292}
\n+
293
\n+
294//
\n+
295// Limit masks for tangents -- these are stubs for now, or have a temporary
\n+
296// implementation
\n+
297//
\n+
298template <>
\n+
299template <typename VERTEX, typename MASK>
\n+
300inline void
\n+
301Scheme<SCHEME_CATMARK>::assignCornerLimitTangentMasks(VERTEX const& vertex,
\n+
302 MASK& tan1Mask, MASK& tan2Mask) const {
\n+
303
\n+
304 int valence = vertex.GetNumEdges();
\n+
305
\n+
306 tan1Mask.SetNumVertexWeights(1);
\n+
307 tan1Mask.SetNumEdgeWeights(valence);
\n+
308 tan1Mask.SetNumFaceWeights(0);
\n+
309 tan1Mask.SetFaceWeightsForFaceCenters(false);
\n+
310
\n+
311 tan2Mask.SetNumVertexWeights(1);
\n+
312 tan2Mask.SetNumEdgeWeights(valence);
\n+
313 tan2Mask.SetNumFaceWeights(0);
\n+
314 tan2Mask.SetFaceWeightsForFaceCenters(false);
\n+
315
\n+
316 // Should be at least 2 edges -- be sure to clear weights for any more:
\n+
317 tan1Mask.VertexWeight(0) = -1.0f;
\n+
318 tan1Mask.EdgeWeight(0) = 1.0f;
\n+
319 tan1Mask.EdgeWeight(1) = 0.0f;
\n+
320
\n+
321 tan2Mask.VertexWeight(0) = -1.0f;
\n+
322 tan2Mask.EdgeWeight(0) = 0.0f;
\n+
323 tan2Mask.EdgeWeight(1) = 1.0f;
\n+
324
\n+
325 for (int i = 2; i < valence; ++i) {
\n+
326 tan1Mask.EdgeWeight(i) = 0.0f;
\n+
327 tan2Mask.EdgeWeight(i) = 0.0f;
\n+
328 }
\n+
329}
\n+
330
\n+
331template <>
\n+
332template <typename VERTEX, typename MASK>
\n+
333inline void
\n+
334Scheme<SCHEME_CATMARK>::assignCreaseLimitTangentMasks(VERTEX const& vertex,
\n+
335 MASK& tan1Mask, MASK& tan2Mask, int const creaseEnds[2]) const {
\n+
336
\n+
337 typedef typename MASK::Weight Weight;
\n+
338
\n+
339 //
\n+
340 // First, the tangent along the crease:
\n+
341 // The first crease edge is considered the "leading" edge of the span
\n+
342 // of surface for which we are evaluating tangents and the second edge the
\n+
343 // "trailing edge". By convention, the tangent along the crease is oriented
\n+
344 // in the direction of the leading edge.
\n+
345 //
\n+
346 int numEdges = vertex.GetNumEdges();
\n+
347 int numFaces = vertex.GetNumFaces();
\n+
348
\n+
349 tan1Mask.SetNumVertexWeights(1);
\n+
350 tan1Mask.SetNumEdgeWeights(numEdges);
\n+
351 tan1Mask.SetNumFaceWeights(numFaces);
\n+
352 tan1Mask.SetFaceWeightsForFaceCenters(false);
\n+
353
\n+
354 tan1Mask.VertexWeight(0) = 0.0f;
\n+
355 for (int i = 0; i < numEdges; ++i) {
\n+
356 tan1Mask.EdgeWeight(i) = 0.0f;
\n+
357 }
\n+
358 for (int i = 0; i < numFaces; ++i) {
\n+
359 tan1Mask.FaceWeight(i) = 0.0f;
\n+
360 }
\n+
361
\n+
362 tan1Mask.EdgeWeight(creaseEnds[0]) = 0.5f;
\n+
363 tan1Mask.EdgeWeight(creaseEnds[1]) = -0.5f;
\n+
364
\n+
365 //
\n+
366 // Second, the tangent across the interior faces:
\n+
367 // Note this is ambiguous for an interior vertex. We currently return
\n+
368 // the tangent for the surface in the counter-clockwise span between the
\n+
369 // leading and trailing edges that form the crease. Given the expected
\n+
370 // computation of a surface normal as Tan1 X Tan2, this tangent should be
\n+
371 // oriented "inward" from the crease/boundary -- across the surface rather
\n+
372 // than outward and away from it.
\n+
373 //
\n+
374 tan2Mask.SetNumVertexWeights(1);
\n+
375 tan2Mask.SetNumEdgeWeights(numEdges);
\n+
376 tan2Mask.SetNumFaceWeights(numFaces);
\n+
377 tan2Mask.SetFaceWeightsForFaceCenters(false);
\n+
378
\n+
379 // Prepend weights of 0 preceding the crease:
\n+
380 for (int i = 0; i < creaseEnds[0]; ++i) {
\n+
381 tan2Mask.EdgeWeight(i) = 0.0f;
\n+
382 tan2Mask.FaceWeight(i) = 0.0f;
\n+
383 }
\n+
384
\n+
385 // Assign weights to crease edge and interior points:
\n+
386 int interiorEdgeCount = creaseEnds[1] - creaseEnds[0] - 1;
\n+
387 if (interiorEdgeCount == 1) {
\n+
388 // The regular case -- uniform B-spline cross-tangent:
\n+
389
\n+
390 tan2Mask.VertexWeight(0) = (Weight)(-4.0 / 6.0);
\n+
391
\n+
392 tan2Mask.EdgeWeight(creaseEnds[0]) = (Weight)(-1.0 / 6.0);
\n+
393 tan2Mask.EdgeWeight(creaseEnds[0] + 1) = (Weight)( 4.0 / 6.0);
\n+
394 tan2Mask.EdgeWeight(creaseEnds[1]) = (Weight)(-1.0 / 6.0);
\n+
395
\n+
396 tan2Mask.FaceWeight(creaseEnds[0]) = (Weight)(1.0 / 6.0);
\n+
397 tan2Mask.FaceWeight(creaseEnds[0] + 1) = (Weight)(1.0 / 6.0);
\n+
398 } else if (interiorEdgeCount > 1) {
\n+
399 // The irregular case -- formulae from Biermann et al:
\n+
400
\n+
401 double k = (double) (interiorEdgeCount + 1);
\n+
402 double theta = M_PI / k;
\n+
403
\n+
404 double cosTheta = std::cos(theta);
\n+
405 double sinTheta = std::sin(theta);
\n+
406
\n+
407 // Loop/Schaefer use a different divisor here (3*k + cos(theta)):
\n+
408 double commonDenom = 1.0f / (k * (3.0f + cosTheta));
\n+
409 double R = (cosTheta + 1.0f) / sinTheta;
\n+
410
\n+
411 double vertexWeight = 4.0f * R * (cosTheta - 1.0f);
\n+
412 double creaseWeight = -R * (1.0f + 2.0f * cosTheta);
\n+
413
\n+
414 tan2Mask.VertexWeight(0) = (Weight) (vertexWeight * commonDenom);
\n+
415
\n+
416 tan2Mask.EdgeWeight(creaseEnds[0]) = (Weight) (creaseWeight * commonDenom);
\n+
417 tan2Mask.EdgeWeight(creaseEnds[1]) = (Weight) (creaseWeight * commonDenom);
\n+
418
\n+
419 tan2Mask.FaceWeight(creaseEnds[0]) = (Weight) (sinTheta * commonDenom);
\n+
420
\n+
421 double sinThetaI = 0.0f;
\n+
422 double sinThetaIplus1 = sinTheta;
\n+
423 for (int i = 1; i < k; ++i) {
\n+
424 sinThetaI = sinThetaIplus1;
\n+
425 sinThetaIplus1 = std::sin((i+1)*theta);
\n+
426
\n+
427 tan2Mask.EdgeWeight(creaseEnds[0] + i) = (Weight) ((4.0f * sinThetaI) * commonDenom);
\n+
428 tan2Mask.FaceWeight(creaseEnds[0] + i) = (Weight) ((sinThetaI + sinThetaIplus1) * commonDenom);
\n+
429 }
\n+
430 } else {
\n+
431 // Special case for a single face -- simple average of boundary edges:
\n+
432
\n+
433 tan2Mask.VertexWeight(0) = -6.0f;
\n+
434
\n+
435 tan2Mask.EdgeWeight(creaseEnds[0]) = 3.0f;
\n+
436 tan2Mask.EdgeWeight(creaseEnds[1]) = 3.0f;
\n+
437
\n+
438 tan2Mask.FaceWeight(creaseEnds[0]) = 0.0f;
\n+
439 }
\n+
440
\n+
441 // Append weights of 0 following the crease:
\n+
442 for (int i = creaseEnds[1]; i < numFaces; ++i) {
\n+
443 tan2Mask.FaceWeight(i) = 0.0f;
\n+
444 }
\n+
445 for (int i = creaseEnds[1] + 1; i < numEdges; ++i) {
\n+
446 tan2Mask.EdgeWeight(i) = 0.0f;
\n+
447 }
\n+
448}
\n+
449
\n+
450template <>
\n+
451template <typename VERTEX, typename MASK>
\n+
452inline void
\n+
453Scheme<SCHEME_CATMARK>::assignSmoothLimitTangentMasks(VERTEX const& vertex,
\n+
454 MASK& tan1Mask, MASK& tan2Mask) const {
\n+
455
\n+
456 typedef typename MASK::Weight Weight;
\n+
457
\n+
458 int valence = vertex.GetNumFaces();
\n+
459 if (valence == 2) {
\n+
460 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);
\n+
461 return;
\n+
462 }
\n+
463
\n+
464 // Compute tan1 initially -- tan2 is simply a rotation:
\n+
465 tan1Mask.SetNumVertexWeights(1);
\n+
466 tan1Mask.SetNumEdgeWeights(valence);
\n+
467 tan1Mask.SetNumFaceWeights(valence);
\n+
468 tan1Mask.SetFaceWeightsForFaceCenters(false);
\n+
469
\n+
470 tan1Mask.VertexWeight(0) = 0.0f;
\n+
471
\n+
472 if (valence == 4) {
\n+
473 tan1Mask.EdgeWeight(0) = 4.0f;
\n+
474 tan1Mask.EdgeWeight(1) = 0.0f;
\n+
475 tan1Mask.EdgeWeight(2) = -4.0f;
\n+
476 tan1Mask.EdgeWeight(3) = 0.0f;
\n+
477
\n+
478 tan1Mask.FaceWeight(0) = 1.0f;
\n+
479 tan1Mask.FaceWeight(1) = -1.0f;
\n+
480 tan1Mask.FaceWeight(2) = -1.0f;
\n+
481 tan1Mask.FaceWeight(3) = 1.0f;
\n+
482 } else {
\n+
483 double theta = 2.0f * M_PI / (double)valence;
\n+
484
\n+
485 double cosTheta = std::cos(theta);
\n+
486 double cosHalfTheta = std::cos(theta * 0.5f);
\n+
487
\n+
488 double lambda = (5.0 / 16.0) + (1.0 / 16.0) *
\n+
489 (cosTheta + cosHalfTheta * std::sqrt(2.0f * (9.0f + cosTheta)));
\n+
490
\n+
491 double edgeWeightScale = 4.0f;
\n+
492 double faceWeightScale = 1.0f / (4.0f * lambda - 1.0f);
\n+
493
\n+
494 for (int i = 0; i < valence; ++i) {
\n+
495 double cosThetaI = std::cos( i * theta);
\n+
496 double cosThetaIplus1 = std::cos((i+1)* theta);
\n+
497
\n+
498 tan1Mask.EdgeWeight(i) = (Weight) (edgeWeightScale * cosThetaI);
\n+
499 tan1Mask.FaceWeight(i) = (Weight) (faceWeightScale * (cosThetaI + cosThetaIplus1));
\n+
500 }
\n+
501 }
\n+
502
\n+
503 // Now rotate/copy tan1 weights to tan2:
\n+
504 tan2Mask.SetNumVertexWeights(1);
\n+
505 tan2Mask.SetNumEdgeWeights(valence);
\n+
506 tan2Mask.SetNumFaceWeights(valence);
\n+
507 tan2Mask.SetFaceWeightsForFaceCenters(false);
\n+
508
\n+
509 tan2Mask.VertexWeight(0) = 0.0f;
\n+
510 if (valence == 4) {
\n+
511 tan2Mask.EdgeWeight(0) = 0.0f;
\n+
512 tan2Mask.EdgeWeight(1) = 4.0f;
\n+
513 tan2Mask.EdgeWeight(2) = 0.0f;
\n+
514 tan2Mask.EdgeWeight(3) = -4.0f;
\n+
515
\n+
516 tan2Mask.FaceWeight(0) = 1.0f;
\n+
517 tan2Mask.FaceWeight(1) = 1.0f;
\n+
518 tan2Mask.FaceWeight(2) = -1.0f;
\n+
519 tan2Mask.FaceWeight(3) = -1.0f;
\n+
520 } else {
\n+
521 tan2Mask.EdgeWeight(0) = tan1Mask.EdgeWeight(valence-1);
\n+
522 tan2Mask.FaceWeight(0) = tan1Mask.FaceWeight(valence-1);
\n+
523 for (int i = 1; i < valence; ++i) {
\n+
524 tan2Mask.EdgeWeight(i) = tan1Mask.EdgeWeight(i-1);
\n+
525 tan2Mask.FaceWeight(i) = tan1Mask.FaceWeight(i-1);
\n+
526 }
\n+
527 }
\n+
528}
\n+
529
\n+
530} // end namespace sdc
\n+
531
\n+
532} // end namespace OPENSUBDIV_VERSION
\n+
533using namespace OPENSUBDIV_VERSION;
\n+
534} // end namespace OpenSubdiv
\n+
535
\n+
536#endif /* OPENSUBDIV3_SDC_CATMARK_SCHEME_H */
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Split
Split
Enumerated type for all face splitting schemes.
Definition: types.h:47
\n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SPLIT_TO_QUADS
@ SPLIT_TO_QUADS
Used by Catmark and Bilinear.
Definition: types.h:48
\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Options::TRI_SUB_SMOOTH
@ TRI_SUB_SMOOTH
"smooth triangle" weights (Catmark scheme only)
Definition: options.h:74
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothMaskForEdge
void assignSmoothMaskForEdge(EDGE const &edge, MASK &mask) const
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::ComputeVertexVertexMask
void ComputeVertexVertexMask(VERTEX const &vertexNeighborhood, MASK &vertexVertexMask, Crease::Rule parentRule=Crease::RULE_UNKNOWN, Crease::Rule childRule=Crease::RULE_UNKNOWN) const
Vertex-vertex masks If known, a single Rule or pair of Rules can be specified (indicating a crease tr...
Definition: scheme.h:512
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseMaskForVertex
void assignCreaseMaskForVertex(VERTEX const &edge, MASK &mask, int const creaseEnds[2]) const
\n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseLimitTangentMasks
void assignCreaseLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK &tan2, int const creaseEnds[2]) const
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::ComputeEdgeVertexMask
void ComputeEdgeVertexMask(EDGE const &edgeNeighborhood, MASK &edgeVertexMask, Crease::Rule parentRule=Crease::RULE_UNKNOWN, Crease::Rule childRule=Crease::RULE_UNKNOWN) const
Edge-vertex masks If known, the Rule for the edge and/or the derived vertex can be specified to accel...
Definition: scheme.h:392
\n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothLimitMask
void assignSmoothLimitMask(VERTEX const &vertex, MASK &pos) const
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseLimitMask
void assignCreaseLimitMask(VERTEX const &vertex, MASK &pos, int const creaseEnds[2]) const
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothMaskForVertex
void assignSmoothMaskForVertex(VERTEX const &edge, MASK &mask) const
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothLimitTangentMasks
void assignSmoothLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK &tan2) const
\n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCornerLimitMask
void assignCornerLimitMask(VERTEX const &vertex, MASK &pos) const
\n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCornerLimitTangentMasks
void assignCornerLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK &tan2) const
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-bilinearScheme.h\n+catmarkScheme.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2014 DreamWorks Animation LLC.\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -29,211 +29,595 @@\n 17//\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n- 24#ifndef OPENSUBDIV3_SDC_BILINEAR_SCHEME_H\n- 25#define OPENSUBDIV3_SDC_BILINEAR_SCHEME_H\n+ 24#ifndef OPENSUBDIV3_SDC_CATMARK_SCHEME_H\n+ 25#define OPENSUBDIV3_SDC_CATMARK_SCHEME_H\n 26\n 27#include \"../version.h\"\n 28\n 29#include \"../sdc/scheme.h\"\n 30\n- 31namespace OpenSubdiv {\n- 32namespace OPENSUBDIV_VERSION {\n+ 31#include \n+ 32#include \n 33\n-34namespace Sdc {\n- 35\n- 36//\n- 37// Specializations for Scheme:\n- 38//\n- 39\n- 40//\n- 41// Bilinear traits:\n- 42//\n- 43template <>\n-44inline Split Scheme::GetTopologicalSplitType() { return\n-SPLIT_TO_QUADS; }\n- 45\n+ 34namespace OpenSubdiv {\n+ 35namespace OPENSUBDIV_VERSION {\n+ 36\n+ 37namespace Sdc {\n+ 38\n+ 39//\n+ 40// Specializations for Scheme:\n+ 41//\n+ 42\n+ 43//\n+ 44// Catmark traits:\n+ 45//\n 46template <>\n-47inline int Scheme::GetRegularFaceSize() { return 4; }\n+47inline Split Scheme::GetTopologicalSplitType() { return\n+SPLIT_TO_QUADS; }\n 48\n 49template <>\n-50inline int Scheme::GetRegularVertexValence() { return 4; }\n+50inline int Scheme::GetRegularFaceSize() { return 4; }\n 51\n 52template <>\n-53inline int Scheme::GetLocalNeighborhoodSize() { return 0; }\n+53inline int Scheme::GetRegularVertexValence() { return 4; }\n 54\n- 55\n- 56//\n- 57// Refinement masks:\n- 58//\n- 59template <>\n- 60template \n- 61void\n-62Scheme::ComputeEdgeVertexMask(EDGE const& edge, MASK& mask,\n- 63 Crease::Rule, Crease::Rule) const {\n- 64 // This should be inline, otherwise trivially replicate it:\n- 65 assignCreaseMaskForEdge(edge, mask);\n- 66}\n- 67\n+ 55template <>\n+56inline int Scheme::GetLocalNeighborhoodSize() { return 1; }\n+ 57\n+ 58\n+ 59//\n+ 60// Masks for edge-vertices: the hard Crease mask does not need to be\n+specialized\n+ 61// (simply the midpoint), so all that is left is the Smooth case:\n+ 62//\n+ 63// The Smooth mask is complicated by the need to support the \"triangle\n+subdivision\"\n+ 64// option, which applies different weighting in the presence of triangles.\n+It is\n+ 65// up for debate as to whether this is useful or not -- we may be able to\n+deprecate\n+ 66// this option.\n+ 67//\n 68template <>\n- 69template \n- 70void\n-71Scheme::ComputeVertexVertexMask(VERTEX const& vertex, MASK&\n-mask,\n- 72 Crease::Rule, Crease::Rule) const {\n- 73 // This should be inline, otherwise trivially replicate it:\n- 74 assignCornerMaskForVertex(vertex, mask);\n- 75}\n+ 69template \n+ 70inline void\n+71Scheme::assignSmoothMaskForEdge(EDGE const& edge, MASK& mask)\n+const {\n+ 72\n+ 73 typedef typename MASK::Weight Weight;\n+ 74\n+ 75 int faceCount = edge.GetNumFaces();\n 76\n- 77\n- 78//\n- 79// Limit masks for position -- the limit position of all vertices is the\n-refined vertex.\n- 80//\n- 81template <>\n- 82template \n- 83inline void\n-84Scheme::assignCornerLimitMask(VERTEX const& /* vertex */,\n-MASK& posMask) const {\n- 85\n- 86 posMask.SetNumVertexWeights(1);\n- 87 posMask.SetNumEdgeWeights(0);\n- 88 posMask.SetNumFaceWeights(0);\n- 89 posMask.SetFaceWeightsForFaceCenters(false);\n- 90\n- 91 posMask.VertexWeight(0) = 1.0f;\n- 92}\n- 93\n- 94template <>\n- 95template \n- 96inline void\n-97Scheme::assignCreaseLimitMask(VERTEX const& vertex, MASK&\n-posMask,\n- 98 int const /* creaseEnds */[2]) const {\n- 99\n- 100 assignCornerLimitMask(vertex, posMask);\n- 101}\n- 102\n- 103template <>\n- 104template \n- 105inline void\n-106Scheme::assignSmoothLimitMask(VERTEX const& vertex, MASK&\n-posMask) const {\n- 107\n- 108 assignCornerLimitMask(vertex, posMask);\n- 109}\n+ 77 mask.SetNumVertexWeights(2);\n+ 78 mask.SetNumEdgeWeights(0);\n+ 79 mask.SetNumFaceWeights(faceCount);\n+ 80 mask.SetFaceWeightsForFaceCenters(true);\n+ 81\n+ 82 //\n+ 83 // Determine if we need to inspect incident faces and apply alternate\n+weighting for\n+ 84 // triangles -- and if so, determine which of the two are triangles.\n+ 85 //\n+ 86 bool face0IsTri = false;\n+ 87 bool face1IsTri = false;\n+ 88 bool useTriangleOption = (_options.GetTriangleSubdivision() == Options::\n+TRI_SUB_SMOOTH);\n+ 89 if (useTriangleOption) {\n+ 90 if (faceCount == 2) {\n+ 91 //\n+ 92 // Ideally we want to avoid this inspection when we have already subdivided\n+at\n+ 93 // least once -- need something in the Edge interface to help avoid this,\n+e.g.\n+ 94 // an IsRegular() query, the subdivision level...\n+ 95 //\n+ 96 int vertsPerFace[2];\n+ 97 edge.GetNumVerticesPerFace(vertsPerFace);\n+ 98\n+ 99 face0IsTri = (vertsPerFace[0] == 3);\n+ 100 face1IsTri = (vertsPerFace[1] == 3);\n+ 101 useTriangleOption = face0IsTri || face1IsTri;\n+ 102 } else {\n+ 103 useTriangleOption = false;\n+ 104 }\n+ 105 }\n+ 106\n+ 107 if (! useTriangleOption) {\n+ 108 mask.VertexWeight(0) = 0.25f;\n+ 109 mask.VertexWeight(1) = 0.25f;\n 110\n- 111//\n- 112// Limit masks for tangents -- these are ambiguous around all vertices.\n-Provide\n- 113// the tangents based on the incident edges of the first face.\n- 114//\n- 115template <>\n- 116template \n- 117inline void\n-118Scheme::assignCornerLimitTangentMasks(VERTEX const& /\n-* vertex */,\n- 119 MASK& tan1Mask, MASK& tan2Mask) const {\n- 120\n- 121 tan1Mask.SetNumVertexWeights(1);\n- 122 tan1Mask.SetNumEdgeWeights(2);\n- 123 tan1Mask.SetNumFaceWeights(0);\n- 124 tan1Mask.SetFaceWeightsForFaceCenters(false);\n+ 111 if (faceCount == 2) {\n+ 112 mask.FaceWeight(0) = 0.25f;\n+ 113 mask.FaceWeight(1) = 0.25f;\n+ 114 } else {\n+ 115 Weight fWeight = 0.5f / (Weight)faceCount;\n+ 116 for (int i = 0; i < faceCount; ++i) {\n+ 117 mask.FaceWeight(i) = fWeight;\n+ 118 }\n+ 119 }\n+ 120 } else {\n+ 121 //\n+ 122 // This mimics the implementation in Hbr in terms of order of operations.\n+ 123 //\n+ 124 const Weight CATMARK_SMOOTH_TRI_EDGE_WEIGHT = (Weight) 0.470;\n 125\n- 126 tan2Mask.SetNumVertexWeights(1);\n- 127 tan2Mask.SetNumEdgeWeights(2);\n- 128 tan2Mask.SetNumFaceWeights(0);\n- 129 tan2Mask.SetFaceWeightsForFaceCenters(false);\n- 130\n- 131 tan1Mask.VertexWeight(0) = -1.0f;\n- 132 tan1Mask.EdgeWeight(0) = 1.0f;\n- 133 tan1Mask.EdgeWeight(1) = 0.0f;\n+ 126 Weight f0Weight = face0IsTri ? CATMARK_SMOOTH_TRI_EDGE_WEIGHT : 0.25f;\n+ 127 Weight f1Weight = face1IsTri ? CATMARK_SMOOTH_TRI_EDGE_WEIGHT : 0.25f;\n+ 128\n+ 129 Weight fWeight = 0.5f * (f0Weight + f1Weight);\n+ 130 Weight vWeight = 0.5f * (1.0f - 2.0f * fWeight);\n+ 131\n+ 132 mask.VertexWeight(0) = vWeight;\n+ 133 mask.VertexWeight(1) = vWeight;\n 134\n- 135 tan2Mask.VertexWeight(0) = -1.0f;\n- 136 tan2Mask.EdgeWeight(0) = 0.0f;\n- 137 tan2Mask.EdgeWeight(1) = 1.0f;\n+ 135 mask.FaceWeight(0) = fWeight;\n+ 136 mask.FaceWeight(1) = fWeight;\n+ 137 }\n 138}\n 139\n- 140template <>\n- 141template \n- 142inline void\n-143Scheme::assignCreaseLimitTangentMasks(VERTEX const& vertex,\n- 144 MASK& tan1Mask, MASK& tan2Mask, int const /* creaseEnds */[2]) const {\n- 145\n- 146 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);\n- 147}\n- 148\n- 149template <>\n- 150template \n- 151inline void\n-152Scheme::assignSmoothLimitTangentMasks(VERTEX const& vertex,\n- 153 MASK& tan1Mask, MASK& tan2Mask) const {\n- 154\n- 155 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);\n- 156}\n- 157\n- 158} // end namespace sdc\n- 159\n- 160} // end namespace OPENSUBDIV_VERSION\n- 161using namespace OPENSUBDIV_VERSION;\n- 162} // end namespace OpenSubdiv\n- 163\n- 164#endif /* OPENSUBDIV3_SDC_BILINEAR_SCHEME_H */\n+ 140\n+ 141//\n+ 142// Masks for vertex-vertices: the hard Corner mask does not need to be\n+specialized\n+ 143// (simply the vertex itself), leaving the Crease and Smooth cases (Dart is\n+smooth):\n+ 144//\n+ 145template <>\n+ 146template \n+ 147inline void\n+148Scheme::assignCreaseMaskForVertex(VERTEX const& vertex,\n+MASK& mask,\n+ 149 int const creaseEnds[2]) const {\n+ 150 typedef typename MASK::Weight Weight;\n+ 151\n+ 152 int valence = vertex.GetNumEdges();\n+ 153\n+ 154 mask.SetNumVertexWeights(1);\n+ 155 mask.SetNumEdgeWeights(valence);\n+ 156 mask.SetNumFaceWeights(0);\n+ 157 mask.SetFaceWeightsForFaceCenters(false);\n+ 158\n+ 159 Weight vWeight = 0.75f;\n+ 160 Weight eWeight = 0.125f;\n+ 161\n+ 162 mask.VertexWeight(0) = vWeight;\n+ 163 for (int i = 0; i < valence; ++i) {\n+ 164 mask.EdgeWeight(i) = 0.0f;\n+ 165 }\n+ 166 mask.EdgeWeight(creaseEnds[0]) = eWeight;\n+ 167 mask.EdgeWeight(creaseEnds[1]) = eWeight;\n+ 168}\n+ 169\n+ 170template <>\n+ 171template \n+ 172inline void\n+173Scheme::assignSmoothMaskForVertex(VERTEX const& vertex,\n+MASK& mask) const {\n+ 174\n+ 175 typedef typename MASK::Weight Weight;\n+ 176\n+ 177 //\n+ 178 // A Smooth vertex must be manifold and interior -- manifold boundary\n+vertices will be\n+ 179 // Creases and non-manifold vertices of any kind will be Corners or\n+Creases. If smooth\n+ 180 // rules for non-manifold vertices are ever defined, this will need\n+adjusting:\n+ 181 //\n+ 182 assert(vertex.GetNumFaces() == vertex.GetNumEdges());\n+ 183\n+ 184 int valence = vertex.GetNumFaces();\n+ 185\n+ 186 mask.SetNumVertexWeights(1);\n+ 187 mask.SetNumEdgeWeights(valence);\n+ 188 mask.SetNumFaceWeights(valence);\n+ 189 mask.SetFaceWeightsForFaceCenters(true);\n+ 190\n+ 191 Weight vWeight = (Weight)(valence - 2) / (Weight)valence;\n+ 192 Weight fWeight = 1.0f / (Weight)(valence * valence);\n+ 193 Weight eWeight = fWeight;\n+ 194\n+ 195 mask.VertexWeight(0) = vWeight;\n+ 196 for (int i = 0; i < valence; ++i) {\n+ 197 mask.EdgeWeight(i) = eWeight;\n+ 198 mask.FaceWeight(i) = fWeight;\n+ 199 }\n+ 200}\n+ 201\n+ 202//\n+ 203// Limit masks for position:\n+ 204//\n+ 205template <>\n+ 206template \n+ 207inline void\n+208Scheme::assignCornerLimitMask(VERTEX const& /* vertex */,\n+MASK& posMask) const {\n+ 209\n+ 210 posMask.SetNumVertexWeights(1);\n+ 211 posMask.SetNumEdgeWeights(0);\n+ 212 posMask.SetNumFaceWeights(0);\n+ 213 posMask.SetFaceWeightsForFaceCenters(false);\n+ 214\n+ 215 posMask.VertexWeight(0) = 1.0f;\n+ 216}\n+ 217\n+ 218template <>\n+ 219template \n+ 220inline void\n+221Scheme::assignCreaseLimitMask(VERTEX const& vertex, MASK&\n+posMask,\n+ 222 int const creaseEnds[2]) const {\n+ 223\n+ 224 typedef typename MASK::Weight Weight;\n+ 225\n+ 226 int valence = vertex.GetNumEdges();\n+ 227\n+ 228 posMask.SetNumVertexWeights(1);\n+ 229 posMask.SetNumEdgeWeights(valence);\n+ 230 posMask.SetNumFaceWeights(0);\n+ 231 posMask.SetFaceWeightsForFaceCenters(false);\n+ 232\n+ 233 Weight vWeight = (Weight)(2.0 / 3.0);\n+ 234 Weight eWeight = (Weight)(1.0 / 6.0);\n+ 235\n+ 236 posMask.VertexWeight(0) = vWeight;\n+ 237 for (int i = 0; i < valence; ++i) {\n+ 238 posMask.EdgeWeight(i) = 0.0f;\n+ 239 }\n+ 240 posMask.EdgeWeight(creaseEnds[0]) = eWeight;\n+ 241 posMask.EdgeWeight(creaseEnds[1]) = eWeight;\n+ 242}\n+ 243\n+ 244template <>\n+ 245template \n+ 246inline void\n+247Scheme::assignSmoothLimitMask(VERTEX const& vertex, MASK&\n+posMask) const {\n+ 248\n+ 249 typedef typename MASK::Weight Weight;\n+ 250\n+ 251 int valence = vertex.GetNumFaces();\n+ 252 if (valence == 2) {\n+ 253 assignCornerLimitMask(vertex, posMask);\n+ 254 return;\n+ 255 }\n+ 256\n+ 257 posMask.SetNumVertexWeights(1);\n+ 258 posMask.SetNumEdgeWeights(valence);\n+ 259 posMask.SetNumFaceWeights(valence);\n+ 260 posMask.SetFaceWeightsForFaceCenters(false);\n+ 261\n+ 262 // Specialize for the regular case:\n+ 263 if (valence == 4) {\n+ 264 Weight fWeight = (Weight)(1.0 / 36.0);\n+ 265 Weight eWeight = (Weight)(1.0 / 9.0);\n+ 266 Weight vWeight = (Weight)(4.0 / 9.0);\n+ 267\n+ 268 posMask.VertexWeight(0) = vWeight;\n+ 269\n+ 270 posMask.EdgeWeight(0) = eWeight;\n+ 271 posMask.EdgeWeight(1) = eWeight;\n+ 272 posMask.EdgeWeight(2) = eWeight;\n+ 273 posMask.EdgeWeight(3) = eWeight;\n+ 274\n+ 275 posMask.FaceWeight(0) = fWeight;\n+ 276 posMask.FaceWeight(1) = fWeight;\n+ 277 posMask.FaceWeight(2) = fWeight;\n+ 278 posMask.FaceWeight(3) = fWeight;\n+ 279 } else {\n+ 280 Weight Valence = (Weight) valence;\n+ 281\n+ 282 Weight fWeight = 1.0f / (Valence * (Valence + 5.0f));\n+ 283 Weight eWeight = 4.0f * fWeight;\n+ 284 Weight vWeight = 1.0f - Valence * (eWeight + fWeight);\n+ 285\n+ 286 posMask.VertexWeight(0) = vWeight;\n+ 287 for (int i = 0; i < valence; ++i) {\n+ 288 posMask.EdgeWeight(i) = eWeight;\n+ 289 posMask.FaceWeight(i) = fWeight;\n+ 290 }\n+ 291 }\n+ 292}\n+ 293\n+ 294//\n+ 295// Limit masks for tangents -- these are stubs for now, or have a temporary\n+ 296// implementation\n+ 297//\n+ 298template <>\n+ 299template \n+ 300inline void\n+301Scheme::assignCornerLimitTangentMasks(VERTEX const& vertex,\n+ 302 MASK& tan1Mask, MASK& tan2Mask) const {\n+ 303\n+ 304 int valence = vertex.GetNumEdges();\n+ 305\n+ 306 tan1Mask.SetNumVertexWeights(1);\n+ 307 tan1Mask.SetNumEdgeWeights(valence);\n+ 308 tan1Mask.SetNumFaceWeights(0);\n+ 309 tan1Mask.SetFaceWeightsForFaceCenters(false);\n+ 310\n+ 311 tan2Mask.SetNumVertexWeights(1);\n+ 312 tan2Mask.SetNumEdgeWeights(valence);\n+ 313 tan2Mask.SetNumFaceWeights(0);\n+ 314 tan2Mask.SetFaceWeightsForFaceCenters(false);\n+ 315\n+ 316 // Should be at least 2 edges -- be sure to clear weights for any more:\n+ 317 tan1Mask.VertexWeight(0) = -1.0f;\n+ 318 tan1Mask.EdgeWeight(0) = 1.0f;\n+ 319 tan1Mask.EdgeWeight(1) = 0.0f;\n+ 320\n+ 321 tan2Mask.VertexWeight(0) = -1.0f;\n+ 322 tan2Mask.EdgeWeight(0) = 0.0f;\n+ 323 tan2Mask.EdgeWeight(1) = 1.0f;\n+ 324\n+ 325 for (int i = 2; i < valence; ++i) {\n+ 326 tan1Mask.EdgeWeight(i) = 0.0f;\n+ 327 tan2Mask.EdgeWeight(i) = 0.0f;\n+ 328 }\n+ 329}\n+ 330\n+ 331template <>\n+ 332template \n+ 333inline void\n+334Scheme::assignCreaseLimitTangentMasks(VERTEX const& vertex,\n+ 335 MASK& tan1Mask, MASK& tan2Mask, int const creaseEnds[2]) const {\n+ 336\n+ 337 typedef typename MASK::Weight Weight;\n+ 338\n+ 339 //\n+ 340 // First, the tangent along the crease:\n+ 341 // The first crease edge is considered the \"leading\" edge of the span\n+ 342 // of surface for which we are evaluating tangents and the second edge the\n+ 343 // \"trailing edge\". By convention, the tangent along the crease is\n+oriented\n+ 344 // in the direction of the leading edge.\n+ 345 //\n+ 346 int numEdges = vertex.GetNumEdges();\n+ 347 int numFaces = vertex.GetNumFaces();\n+ 348\n+ 349 tan1Mask.SetNumVertexWeights(1);\n+ 350 tan1Mask.SetNumEdgeWeights(numEdges);\n+ 351 tan1Mask.SetNumFaceWeights(numFaces);\n+ 352 tan1Mask.SetFaceWeightsForFaceCenters(false);\n+ 353\n+ 354 tan1Mask.VertexWeight(0) = 0.0f;\n+ 355 for (int i = 0; i < numEdges; ++i) {\n+ 356 tan1Mask.EdgeWeight(i) = 0.0f;\n+ 357 }\n+ 358 for (int i = 0; i < numFaces; ++i) {\n+ 359 tan1Mask.FaceWeight(i) = 0.0f;\n+ 360 }\n+ 361\n+ 362 tan1Mask.EdgeWeight(creaseEnds[0]) = 0.5f;\n+ 363 tan1Mask.EdgeWeight(creaseEnds[1]) = -0.5f;\n+ 364\n+ 365 //\n+ 366 // Second, the tangent across the interior faces:\n+ 367 // Note this is ambiguous for an interior vertex. We currently return\n+ 368 // the tangent for the surface in the counter-clockwise span between the\n+ 369 // leading and trailing edges that form the crease. Given the expected\n+ 370 // computation of a surface normal as Tan1 X Tan2, this tangent should be\n+ 371 // oriented \"inward\" from the crease/boundary -- across the surface rather\n+ 372 // than outward and away from it.\n+ 373 //\n+ 374 tan2Mask.SetNumVertexWeights(1);\n+ 375 tan2Mask.SetNumEdgeWeights(numEdges);\n+ 376 tan2Mask.SetNumFaceWeights(numFaces);\n+ 377 tan2Mask.SetFaceWeightsForFaceCenters(false);\n+ 378\n+ 379 // Prepend weights of 0 preceding the crease:\n+ 380 for (int i = 0; i < creaseEnds[0]; ++i) {\n+ 381 tan2Mask.EdgeWeight(i) = 0.0f;\n+ 382 tan2Mask.FaceWeight(i) = 0.0f;\n+ 383 }\n+ 384\n+ 385 // Assign weights to crease edge and interior points:\n+ 386 int interiorEdgeCount = creaseEnds[1] - creaseEnds[0] - 1;\n+ 387 if (interiorEdgeCount == 1) {\n+ 388 // The regular case -- uniform B-spline cross-tangent:\n+ 389\n+ 390 tan2Mask.VertexWeight(0) = (Weight)(-4.0 / 6.0);\n+ 391\n+ 392 tan2Mask.EdgeWeight(creaseEnds[0]) = (Weight)(-1.0 / 6.0);\n+ 393 tan2Mask.EdgeWeight(creaseEnds[0] + 1) = (Weight)( 4.0 / 6.0);\n+ 394 tan2Mask.EdgeWeight(creaseEnds[1]) = (Weight)(-1.0 / 6.0);\n+ 395\n+ 396 tan2Mask.FaceWeight(creaseEnds[0]) = (Weight)(1.0 / 6.0);\n+ 397 tan2Mask.FaceWeight(creaseEnds[0] + 1) = (Weight)(1.0 / 6.0);\n+ 398 } else if (interiorEdgeCount > 1) {\n+ 399 // The irregular case -- formulae from Biermann et al:\n+ 400\n+ 401 double k = (double) (interiorEdgeCount + 1);\n+ 402 double theta = M_PI / k;\n+ 403\n+ 404 double cosTheta = std::cos(theta);\n+ 405 double sinTheta = std::sin(theta);\n+ 406\n+ 407 // Loop/Schaefer use a different divisor here (3*k + cos(theta)):\n+ 408 double commonDenom = 1.0f / (k * (3.0f + cosTheta));\n+ 409 double R = (cosTheta + 1.0f) / sinTheta;\n+ 410\n+ 411 double vertexWeight = 4.0f * R * (cosTheta - 1.0f);\n+ 412 double creaseWeight = -R * (1.0f + 2.0f * cosTheta);\n+ 413\n+ 414 tan2Mask.VertexWeight(0) = (Weight) (vertexWeight * commonDenom);\n+ 415\n+ 416 tan2Mask.EdgeWeight(creaseEnds[0]) = (Weight) (creaseWeight *\n+commonDenom);\n+ 417 tan2Mask.EdgeWeight(creaseEnds[1]) = (Weight) (creaseWeight *\n+commonDenom);\n+ 418\n+ 419 tan2Mask.FaceWeight(creaseEnds[0]) = (Weight) (sinTheta * commonDenom);\n+ 420\n+ 421 double sinThetaI = 0.0f;\n+ 422 double sinThetaIplus1 = sinTheta;\n+ 423 for (int i = 1; i < k; ++i) {\n+ 424 sinThetaI = sinThetaIplus1;\n+ 425 sinThetaIplus1 = std::sin((i+1)*theta);\n+ 426\n+ 427 tan2Mask.EdgeWeight(creaseEnds[0] + i) = (Weight) ((4.0f * sinThetaI) *\n+commonDenom);\n+ 428 tan2Mask.FaceWeight(creaseEnds[0] + i) = (Weight) ((sinThetaI +\n+sinThetaIplus1) * commonDenom);\n+ 429 }\n+ 430 } else {\n+ 431 // Special case for a single face -- simple average of boundary edges:\n+ 432\n+ 433 tan2Mask.VertexWeight(0) = -6.0f;\n+ 434\n+ 435 tan2Mask.EdgeWeight(creaseEnds[0]) = 3.0f;\n+ 436 tan2Mask.EdgeWeight(creaseEnds[1]) = 3.0f;\n+ 437\n+ 438 tan2Mask.FaceWeight(creaseEnds[0]) = 0.0f;\n+ 439 }\n+ 440\n+ 441 // Append weights of 0 following the crease:\n+ 442 for (int i = creaseEnds[1]; i < numFaces; ++i) {\n+ 443 tan2Mask.FaceWeight(i) = 0.0f;\n+ 444 }\n+ 445 for (int i = creaseEnds[1] + 1; i < numEdges; ++i) {\n+ 446 tan2Mask.EdgeWeight(i) = 0.0f;\n+ 447 }\n+ 448}\n+ 449\n+ 450template <>\n+ 451template \n+ 452inline void\n+453Scheme::assignSmoothLimitTangentMasks(VERTEX const& vertex,\n+ 454 MASK& tan1Mask, MASK& tan2Mask) const {\n+ 455\n+ 456 typedef typename MASK::Weight Weight;\n+ 457\n+ 458 int valence = vertex.GetNumFaces();\n+ 459 if (valence == 2) {\n+ 460 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);\n+ 461 return;\n+ 462 }\n+ 463\n+ 464 // Compute tan1 initially -- tan2 is simply a rotation:\n+ 465 tan1Mask.SetNumVertexWeights(1);\n+ 466 tan1Mask.SetNumEdgeWeights(valence);\n+ 467 tan1Mask.SetNumFaceWeights(valence);\n+ 468 tan1Mask.SetFaceWeightsForFaceCenters(false);\n+ 469\n+ 470 tan1Mask.VertexWeight(0) = 0.0f;\n+ 471\n+ 472 if (valence == 4) {\n+ 473 tan1Mask.EdgeWeight(0) = 4.0f;\n+ 474 tan1Mask.EdgeWeight(1) = 0.0f;\n+ 475 tan1Mask.EdgeWeight(2) = -4.0f;\n+ 476 tan1Mask.EdgeWeight(3) = 0.0f;\n+ 477\n+ 478 tan1Mask.FaceWeight(0) = 1.0f;\n+ 479 tan1Mask.FaceWeight(1) = -1.0f;\n+ 480 tan1Mask.FaceWeight(2) = -1.0f;\n+ 481 tan1Mask.FaceWeight(3) = 1.0f;\n+ 482 } else {\n+ 483 double theta = 2.0f * M_PI / (double)valence;\n+ 484\n+ 485 double cosTheta = std::cos(theta);\n+ 486 double cosHalfTheta = std::cos(theta * 0.5f);\n+ 487\n+ 488 double lambda = (5.0 / 16.0) + (1.0 / 16.0) *\n+ 489 (cosTheta + cosHalfTheta * std::sqrt(2.0f * (9.0f + cosTheta)));\n+ 490\n+ 491 double edgeWeightScale = 4.0f;\n+ 492 double faceWeightScale = 1.0f / (4.0f * lambda - 1.0f);\n+ 493\n+ 494 for (int i = 0; i < valence; ++i) {\n+ 495 double cosThetaI = std::cos( i * theta);\n+ 496 double cosThetaIplus1 = std::cos((i+1)* theta);\n+ 497\n+ 498 tan1Mask.EdgeWeight(i) = (Weight) (edgeWeightScale * cosThetaI);\n+ 499 tan1Mask.FaceWeight(i) = (Weight) (faceWeightScale * (cosThetaI +\n+cosThetaIplus1));\n+ 500 }\n+ 501 }\n+ 502\n+ 503 // Now rotate/copy tan1 weights to tan2:\n+ 504 tan2Mask.SetNumVertexWeights(1);\n+ 505 tan2Mask.SetNumEdgeWeights(valence);\n+ 506 tan2Mask.SetNumFaceWeights(valence);\n+ 507 tan2Mask.SetFaceWeightsForFaceCenters(false);\n+ 508\n+ 509 tan2Mask.VertexWeight(0) = 0.0f;\n+ 510 if (valence == 4) {\n+ 511 tan2Mask.EdgeWeight(0) = 0.0f;\n+ 512 tan2Mask.EdgeWeight(1) = 4.0f;\n+ 513 tan2Mask.EdgeWeight(2) = 0.0f;\n+ 514 tan2Mask.EdgeWeight(3) = -4.0f;\n+ 515\n+ 516 tan2Mask.FaceWeight(0) = 1.0f;\n+ 517 tan2Mask.FaceWeight(1) = 1.0f;\n+ 518 tan2Mask.FaceWeight(2) = -1.0f;\n+ 519 tan2Mask.FaceWeight(3) = -1.0f;\n+ 520 } else {\n+ 521 tan2Mask.EdgeWeight(0) = tan1Mask.EdgeWeight(valence-1);\n+ 522 tan2Mask.FaceWeight(0) = tan1Mask.FaceWeight(valence-1);\n+ 523 for (int i = 1; i < valence; ++i) {\n+ 524 tan2Mask.EdgeWeight(i) = tan1Mask.EdgeWeight(i-1);\n+ 525 tan2Mask.FaceWeight(i) = tan1Mask.FaceWeight(i-1);\n+ 526 }\n+ 527 }\n+ 528}\n+ 529\n+ 530} // end namespace sdc\n+ 531\n+ 532} // end namespace OPENSUBDIV_VERSION\n+ 533using namespace OPENSUBDIV_VERSION;\n+ 534} // end namespace OpenSubdiv\n+ 535\n+ 536#endif /* OPENSUBDIV3_SDC_CATMARK_SCHEME_H */\n OpenSubdiv\n Definition: limits.h:32\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Split\n Split\n Enumerated type for all face splitting schemes.\n Definition: types.h:47\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SPLIT_TO_QUADS\n @ SPLIT_TO_QUADS\n Used by Catmark and Bilinear.\n Definition: types.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::Rule\n-Rule\n-Definition: crease.h:82\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Options::TRI_SUB_SMOOTH\n+@ TRI_SUB_SMOOTH\n+\"smooth triangle\" weights (Catmark scheme only)\n+Definition: options.h:74\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothMaskForEdge\n+void assignSmoothMaskForEdge(EDGE const &edge, MASK &mask) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::GetTopologicalSplitType\n static Split GetTopologicalSplitType()\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::ComputeVertexVertexMask\n-void ComputeVertexVertexMask(VERTEX const &vertexNeighborhood, MASK\n-&vertexVertexMask, Crease::Rule parentRule=Crease::RULE_UNKNOWN, Crease::Rule\n-childRule=Crease::RULE_UNKNOWN) const\n-Vertex-vertex masks If known, a single Rule or pair of Rules can be specified\n-(indicating a crease tr...\n-Definition: scheme.h:512\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseMaskForVertex\n+void assignCreaseMaskForVertex(VERTEX const &edge, MASK &mask, int const\n+creaseEnds[2]) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseLimitTangentMasks\n void assignCreaseLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK\n &tan2, int const creaseEnds[2]) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::GetRegularFaceSize\n static int GetRegularFaceSize()\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::ComputeEdgeVertexMask\n-void ComputeEdgeVertexMask(EDGE const &edgeNeighborhood, MASK &edgeVertexMask,\n-Crease::Rule parentRule=Crease::RULE_UNKNOWN, Crease::Rule childRule=Crease::\n-RULE_UNKNOWN) const\n-Edge-vertex masks If known, the Rule for the edge and/or the derived vertex can\n-be specified to accel...\n-Definition: scheme.h:392\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothLimitMask\n void assignSmoothLimitMask(VERTEX const &vertex, MASK &pos) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::GetLocalNeighborhoodSize\n static int GetLocalNeighborhoodSize()\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseLimitMask\n void assignCreaseLimitMask(VERTEX const &vertex, MASK &pos, int const\n creaseEnds[2]) const\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothMaskForVertex\n+void assignSmoothMaskForVertex(VERTEX const &edge, MASK &mask) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::GetRegularVertexValence\n static int GetRegularVertexValence()\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothLimitTangentMasks\n void assignSmoothLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK\n &tan2) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCornerLimitMask\n void assignCornerLimitMask(VERTEX const &vertex, MASK &pos) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCornerLimitTangentMasks\n void assignCornerLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK\n &tan2) const\n * opensubdiv\n * sdc\n- * bilinearScheme.h\n+ * catmarkScheme.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00785.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00785.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/sdc/catmarkScheme.h File Reference\n+OpenSubdiv: opensubdiv/sdc/bilinearScheme.h File Reference\n \n \n \n \n \n \n \n@@ -82,21 +82,19 @@\n name=\"MSearchResults\" id=\"MSearchResults\">\n \n \n \n
\n
\n Namespaces
\n-
catmarkScheme.h File Reference
\n+
bilinearScheme.h File Reference
\n
\n
\n
#include "../version.h"
\n #include "../sdc/scheme.h"
\n-#include <cassert>
\n-#include <cmath>
\n
\n

Go to the source code of this file.

\n \n \n \n \n@@ -106,13 +104,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,24 +5,22 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Namespaces\n-catmarkScheme.h File Reference\n+bilinearScheme.h File Reference\n #include \"../version.h\"\n #include \"../sdc/scheme.h\"\n-#include \n-#include \n Go_to_the_source_code_of_this_file.\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Sdc\n \u00a0\n * opensubdiv\n * sdc\n- * catmarkScheme.h\n+ * bilinearScheme.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00785_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00785_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/sdc/catmarkScheme.h Source File\n+OpenSubdiv: opensubdiv/sdc/bilinearScheme.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
catmarkScheme.h
\n+
bilinearScheme.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2014 DreamWorks Animation LLC.
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -106,548 +106,175 @@\n
17//
\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n-
24#ifndef OPENSUBDIV3_SDC_CATMARK_SCHEME_H
\n-
25#define OPENSUBDIV3_SDC_CATMARK_SCHEME_H
\n+
24#ifndef OPENSUBDIV3_SDC_BILINEAR_SCHEME_H
\n+
25#define OPENSUBDIV3_SDC_BILINEAR_SCHEME_H
\n
26
\n
27#include "../version.h"
\n
28
\n
29#include "../sdc/scheme.h"
\n
30
\n-
31#include <cassert>
\n-
32#include <cmath>
\n+
31namespace OpenSubdiv {
\n+
32namespace OPENSUBDIV_VERSION {
\n
33
\n-
34namespace OpenSubdiv {
\n-
35namespace OPENSUBDIV_VERSION {
\n-
36
\n-
37namespace Sdc {
\n-
38
\n-
39//
\n-
40// Specializations for Scheme<SCHEME_CATMARK>:
\n-
41//
\n-
42
\n-
43//
\n-
44// Catmark traits:
\n-
45//
\n+
34namespace Sdc {
\n+
35
\n+
36//
\n+
37// Specializations for Scheme<SCHEME_BILINEAR>:
\n+
38//
\n+
39
\n+
40//
\n+
41// Bilinear traits:
\n+
42//
\n+
43template <>
\n+
44inline Split Scheme<SCHEME_BILINEAR>::GetTopologicalSplitType() { return SPLIT_TO_QUADS; }
\n+
45
\n
46template <>
\n-
47inline Split Scheme<SCHEME_CATMARK>::GetTopologicalSplitType() { return SPLIT_TO_QUADS; }
\n+
47inline int Scheme<SCHEME_BILINEAR>::GetRegularFaceSize() { return 4; }
\n
48
\n
49template <>
\n-
50inline int Scheme<SCHEME_CATMARK>::GetRegularFaceSize() { return 4; }
\n+
50inline int Scheme<SCHEME_BILINEAR>::GetRegularVertexValence() { return 4; }
\n
51
\n
52template <>
\n-
53inline int Scheme<SCHEME_CATMARK>::GetRegularVertexValence() { return 4; }
\n+
53inline int Scheme<SCHEME_BILINEAR>::GetLocalNeighborhoodSize() { return 0; }
\n
54
\n-
55template <>
\n-
56inline int Scheme<SCHEME_CATMARK>::GetLocalNeighborhoodSize() { return 1; }
\n-
57
\n-
58
\n-
59//
\n-
60// Masks for edge-vertices: the hard Crease mask does not need to be specialized
\n-
61// (simply the midpoint), so all that is left is the Smooth case:
\n-
62//
\n-
63// The Smooth mask is complicated by the need to support the "triangle subdivision"
\n-
64// option, which applies different weighting in the presence of triangles. It is
\n-
65// up for debate as to whether this is useful or not -- we may be able to deprecate
\n-
66// this option.
\n-
67//
\n+
55
\n+
56//
\n+
57// Refinement masks:
\n+
58//
\n+
59template <>
\n+
60template <typename EDGE, typename MASK>
\n+
61void
\n+
62Scheme<SCHEME_BILINEAR>::ComputeEdgeVertexMask(EDGE const& edge, MASK& mask,
\n+
63 Crease::Rule, Crease::Rule) const {
\n+
64 // This should be inline, otherwise trivially replicate it:
\n+
65 assignCreaseMaskForEdge(edge, mask);
\n+
66}
\n+
67
\n
68template <>
\n-
69template <typename EDGE, typename MASK>
\n-
70inline void
\n-
71Scheme<SCHEME_CATMARK>::assignSmoothMaskForEdge(EDGE const& edge, MASK& mask) const {
\n-
72
\n-
73 typedef typename MASK::Weight Weight;
\n-
74
\n-
75 int faceCount = edge.GetNumFaces();
\n+
69template <typename VERTEX, typename MASK>
\n+
70void
\n+
71Scheme<SCHEME_BILINEAR>::ComputeVertexVertexMask(VERTEX const& vertex, MASK& mask,
\n+
72 Crease::Rule, Crease::Rule) const {
\n+
73 // This should be inline, otherwise trivially replicate it:
\n+
74 assignCornerMaskForVertex(vertex, mask);
\n+
75}
\n
76
\n-
77 mask.SetNumVertexWeights(2);
\n-
78 mask.SetNumEdgeWeights(0);
\n-
79 mask.SetNumFaceWeights(faceCount);
\n-
80 mask.SetFaceWeightsForFaceCenters(true);
\n-
81
\n-
82 //
\n-
83 // Determine if we need to inspect incident faces and apply alternate weighting for
\n-
84 // triangles -- and if so, determine which of the two are triangles.
\n-
85 //
\n-
86 bool face0IsTri = false;
\n-
87 bool face1IsTri = false;
\n-
88 bool useTriangleOption = (_options.GetTriangleSubdivision() == Options::TRI_SUB_SMOOTH);
\n-
89 if (useTriangleOption) {
\n-
90 if (faceCount == 2) {
\n-
91 //
\n-
92 // Ideally we want to avoid this inspection when we have already subdivided at
\n-
93 // least once -- need something in the Edge interface to help avoid this, e.g.
\n-
94 // an IsRegular() query, the subdivision level...
\n-
95 //
\n-
96 int vertsPerFace[2];
\n-
97 edge.GetNumVerticesPerFace(vertsPerFace);
\n-
98
\n-
99 face0IsTri = (vertsPerFace[0] == 3);
\n-
100 face1IsTri = (vertsPerFace[1] == 3);
\n-
101 useTriangleOption = face0IsTri || face1IsTri;
\n-
102 } else {
\n-
103 useTriangleOption = false;
\n-
104 }
\n-
105 }
\n-
106
\n-
107 if (! useTriangleOption) {
\n-
108 mask.VertexWeight(0) = 0.25f;
\n-
109 mask.VertexWeight(1) = 0.25f;
\n+
77
\n+
78//
\n+
79// Limit masks for position -- the limit position of all vertices is the refined vertex.
\n+
80//
\n+
81template <>
\n+
82template <typename VERTEX, typename MASK>
\n+
83inline void
\n+
84Scheme<SCHEME_BILINEAR>::assignCornerLimitMask(VERTEX const& /* vertex */, MASK& posMask) const {
\n+
85
\n+
86 posMask.SetNumVertexWeights(1);
\n+
87 posMask.SetNumEdgeWeights(0);
\n+
88 posMask.SetNumFaceWeights(0);
\n+
89 posMask.SetFaceWeightsForFaceCenters(false);
\n+
90
\n+
91 posMask.VertexWeight(0) = 1.0f;
\n+
92}
\n+
93
\n+
94template <>
\n+
95template <typename VERTEX, typename MASK>
\n+
96inline void
\n+
97Scheme<SCHEME_BILINEAR>::assignCreaseLimitMask(VERTEX const& vertex, MASK& posMask,
\n+
98 int const /* creaseEnds */[2]) const {
\n+
99
\n+
100 assignCornerLimitMask(vertex, posMask);
\n+
101}
\n+
102
\n+
103template <>
\n+
104template <typename VERTEX, typename MASK>
\n+
105inline void
\n+
106Scheme<SCHEME_BILINEAR>::assignSmoothLimitMask(VERTEX const& vertex, MASK& posMask) const {
\n+
107
\n+
108 assignCornerLimitMask(vertex, posMask);
\n+
109}
\n
110
\n-
111 if (faceCount == 2) {
\n-
112 mask.FaceWeight(0) = 0.25f;
\n-
113 mask.FaceWeight(1) = 0.25f;
\n-
114 } else {
\n-
115 Weight fWeight = 0.5f / (Weight)faceCount;
\n-
116 for (int i = 0; i < faceCount; ++i) {
\n-
117 mask.FaceWeight(i) = fWeight;
\n-
118 }
\n-
119 }
\n-
120 } else {
\n-
121 //
\n-
122 // This mimics the implementation in Hbr in terms of order of operations.
\n-
123 //
\n-
124 const Weight CATMARK_SMOOTH_TRI_EDGE_WEIGHT = (Weight) 0.470;
\n+
111//
\n+
112// Limit masks for tangents -- these are ambiguous around all vertices. Provide
\n+
113// the tangents based on the incident edges of the first face.
\n+
114//
\n+
115template <>
\n+
116template <typename VERTEX, typename MASK>
\n+
117inline void
\n+
118Scheme<SCHEME_BILINEAR>::assignCornerLimitTangentMasks(VERTEX const& /* vertex */,
\n+
119 MASK& tan1Mask, MASK& tan2Mask) const {
\n+
120
\n+
121 tan1Mask.SetNumVertexWeights(1);
\n+
122 tan1Mask.SetNumEdgeWeights(2);
\n+
123 tan1Mask.SetNumFaceWeights(0);
\n+
124 tan1Mask.SetFaceWeightsForFaceCenters(false);
\n
125
\n-
126 Weight f0Weight = face0IsTri ? CATMARK_SMOOTH_TRI_EDGE_WEIGHT : 0.25f;
\n-
127 Weight f1Weight = face1IsTri ? CATMARK_SMOOTH_TRI_EDGE_WEIGHT : 0.25f;
\n-
128
\n-
129 Weight fWeight = 0.5f * (f0Weight + f1Weight);
\n-
130 Weight vWeight = 0.5f * (1.0f - 2.0f * fWeight);
\n-
131
\n-
132 mask.VertexWeight(0) = vWeight;
\n-
133 mask.VertexWeight(1) = vWeight;
\n+
126 tan2Mask.SetNumVertexWeights(1);
\n+
127 tan2Mask.SetNumEdgeWeights(2);
\n+
128 tan2Mask.SetNumFaceWeights(0);
\n+
129 tan2Mask.SetFaceWeightsForFaceCenters(false);
\n+
130
\n+
131 tan1Mask.VertexWeight(0) = -1.0f;
\n+
132 tan1Mask.EdgeWeight(0) = 1.0f;
\n+
133 tan1Mask.EdgeWeight(1) = 0.0f;
\n
134
\n-
135 mask.FaceWeight(0) = fWeight;
\n-
136 mask.FaceWeight(1) = fWeight;
\n-
137 }
\n+
135 tan2Mask.VertexWeight(0) = -1.0f;
\n+
136 tan2Mask.EdgeWeight(0) = 0.0f;
\n+
137 tan2Mask.EdgeWeight(1) = 1.0f;
\n
138}
\n
139
\n-
140
\n-
141//
\n-
142// Masks for vertex-vertices: the hard Corner mask does not need to be specialized
\n-
143// (simply the vertex itself), leaving the Crease and Smooth cases (Dart is smooth):
\n-
144//
\n-
145template <>
\n-
146template <typename VERTEX, typename MASK>
\n-
147inline void
\n-
148Scheme<SCHEME_CATMARK>::assignCreaseMaskForVertex(VERTEX const& vertex, MASK& mask,
\n-
149 int const creaseEnds[2]) const {
\n-
150 typedef typename MASK::Weight Weight;
\n-
151
\n-
152 int valence = vertex.GetNumEdges();
\n-
153
\n-
154 mask.SetNumVertexWeights(1);
\n-
155 mask.SetNumEdgeWeights(valence);
\n-
156 mask.SetNumFaceWeights(0);
\n-
157 mask.SetFaceWeightsForFaceCenters(false);
\n-
158
\n-
159 Weight vWeight = 0.75f;
\n-
160 Weight eWeight = 0.125f;
\n-
161
\n-
162 mask.VertexWeight(0) = vWeight;
\n-
163 for (int i = 0; i < valence; ++i) {
\n-
164 mask.EdgeWeight(i) = 0.0f;
\n-
165 }
\n-
166 mask.EdgeWeight(creaseEnds[0]) = eWeight;
\n-
167 mask.EdgeWeight(creaseEnds[1]) = eWeight;
\n-
168}
\n-
169
\n-
170template <>
\n-
171template <typename VERTEX, typename MASK>
\n-
172inline void
\n-
173Scheme<SCHEME_CATMARK>::assignSmoothMaskForVertex(VERTEX const& vertex, MASK& mask) const {
\n-
174
\n-
175 typedef typename MASK::Weight Weight;
\n-
176
\n-
177 //
\n-
178 // A Smooth vertex must be manifold and interior -- manifold boundary vertices will be
\n-
179 // Creases and non-manifold vertices of any kind will be Corners or Creases. If smooth
\n-
180 // rules for non-manifold vertices are ever defined, this will need adjusting:
\n-
181 //
\n-
182 assert(vertex.GetNumFaces() == vertex.GetNumEdges());
\n-
183
\n-
184 int valence = vertex.GetNumFaces();
\n-
185
\n-
186 mask.SetNumVertexWeights(1);
\n-
187 mask.SetNumEdgeWeights(valence);
\n-
188 mask.SetNumFaceWeights(valence);
\n-
189 mask.SetFaceWeightsForFaceCenters(true);
\n-
190
\n-
191 Weight vWeight = (Weight)(valence - 2) / (Weight)valence;
\n-
192 Weight fWeight = 1.0f / (Weight)(valence * valence);
\n-
193 Weight eWeight = fWeight;
\n-
194
\n-
195 mask.VertexWeight(0) = vWeight;
\n-
196 for (int i = 0; i < valence; ++i) {
\n-
197 mask.EdgeWeight(i) = eWeight;
\n-
198 mask.FaceWeight(i) = fWeight;
\n-
199 }
\n-
200}
\n-
201
\n-
202//
\n-
203// Limit masks for position:
\n-
204//
\n-
205template <>
\n-
206template <typename VERTEX, typename MASK>
\n-
207inline void
\n-
208Scheme<SCHEME_CATMARK>::assignCornerLimitMask(VERTEX const& /* vertex */, MASK& posMask) const {
\n-
209
\n-
210 posMask.SetNumVertexWeights(1);
\n-
211 posMask.SetNumEdgeWeights(0);
\n-
212 posMask.SetNumFaceWeights(0);
\n-
213 posMask.SetFaceWeightsForFaceCenters(false);
\n-
214
\n-
215 posMask.VertexWeight(0) = 1.0f;
\n-
216}
\n-
217
\n-
218template <>
\n-
219template <typename VERTEX, typename MASK>
\n-
220inline void
\n-
221Scheme<SCHEME_CATMARK>::assignCreaseLimitMask(VERTEX const& vertex, MASK& posMask,
\n-
222 int const creaseEnds[2]) const {
\n-
223
\n-
224 typedef typename MASK::Weight Weight;
\n-
225
\n-
226 int valence = vertex.GetNumEdges();
\n-
227
\n-
228 posMask.SetNumVertexWeights(1);
\n-
229 posMask.SetNumEdgeWeights(valence);
\n-
230 posMask.SetNumFaceWeights(0);
\n-
231 posMask.SetFaceWeightsForFaceCenters(false);
\n-
232
\n-
233 Weight vWeight = (Weight)(2.0 / 3.0);
\n-
234 Weight eWeight = (Weight)(1.0 / 6.0);
\n-
235
\n-
236 posMask.VertexWeight(0) = vWeight;
\n-
237 for (int i = 0; i < valence; ++i) {
\n-
238 posMask.EdgeWeight(i) = 0.0f;
\n-
239 }
\n-
240 posMask.EdgeWeight(creaseEnds[0]) = eWeight;
\n-
241 posMask.EdgeWeight(creaseEnds[1]) = eWeight;
\n-
242}
\n-
243
\n-
244template <>
\n-
245template <typename VERTEX, typename MASK>
\n-
246inline void
\n-
247Scheme<SCHEME_CATMARK>::assignSmoothLimitMask(VERTEX const& vertex, MASK& posMask) const {
\n-
248
\n-
249 typedef typename MASK::Weight Weight;
\n-
250
\n-
251 int valence = vertex.GetNumFaces();
\n-
252 if (valence == 2) {
\n-
253 assignCornerLimitMask(vertex, posMask);
\n-
254 return;
\n-
255 }
\n-
256
\n-
257 posMask.SetNumVertexWeights(1);
\n-
258 posMask.SetNumEdgeWeights(valence);
\n-
259 posMask.SetNumFaceWeights(valence);
\n-
260 posMask.SetFaceWeightsForFaceCenters(false);
\n-
261
\n-
262 // Specialize for the regular case:
\n-
263 if (valence == 4) {
\n-
264 Weight fWeight = (Weight)(1.0 / 36.0);
\n-
265 Weight eWeight = (Weight)(1.0 / 9.0);
\n-
266 Weight vWeight = (Weight)(4.0 / 9.0);
\n-
267
\n-
268 posMask.VertexWeight(0) = vWeight;
\n-
269
\n-
270 posMask.EdgeWeight(0) = eWeight;
\n-
271 posMask.EdgeWeight(1) = eWeight;
\n-
272 posMask.EdgeWeight(2) = eWeight;
\n-
273 posMask.EdgeWeight(3) = eWeight;
\n-
274
\n-
275 posMask.FaceWeight(0) = fWeight;
\n-
276 posMask.FaceWeight(1) = fWeight;
\n-
277 posMask.FaceWeight(2) = fWeight;
\n-
278 posMask.FaceWeight(3) = fWeight;
\n-
279 } else {
\n-
280 Weight Valence = (Weight) valence;
\n-
281
\n-
282 Weight fWeight = 1.0f / (Valence * (Valence + 5.0f));
\n-
283 Weight eWeight = 4.0f * fWeight;
\n-
284 Weight vWeight = 1.0f - Valence * (eWeight + fWeight);
\n-
285
\n-
286 posMask.VertexWeight(0) = vWeight;
\n-
287 for (int i = 0; i < valence; ++i) {
\n-
288 posMask.EdgeWeight(i) = eWeight;
\n-
289 posMask.FaceWeight(i) = fWeight;
\n-
290 }
\n-
291 }
\n-
292}
\n-
293
\n-
294//
\n-
295// Limit masks for tangents -- these are stubs for now, or have a temporary
\n-
296// implementation
\n-
297//
\n-
298template <>
\n-
299template <typename VERTEX, typename MASK>
\n-
300inline void
\n-
301Scheme<SCHEME_CATMARK>::assignCornerLimitTangentMasks(VERTEX const& vertex,
\n-
302 MASK& tan1Mask, MASK& tan2Mask) const {
\n-
303
\n-
304 int valence = vertex.GetNumEdges();
\n-
305
\n-
306 tan1Mask.SetNumVertexWeights(1);
\n-
307 tan1Mask.SetNumEdgeWeights(valence);
\n-
308 tan1Mask.SetNumFaceWeights(0);
\n-
309 tan1Mask.SetFaceWeightsForFaceCenters(false);
\n-
310
\n-
311 tan2Mask.SetNumVertexWeights(1);
\n-
312 tan2Mask.SetNumEdgeWeights(valence);
\n-
313 tan2Mask.SetNumFaceWeights(0);
\n-
314 tan2Mask.SetFaceWeightsForFaceCenters(false);
\n-
315
\n-
316 // Should be at least 2 edges -- be sure to clear weights for any more:
\n-
317 tan1Mask.VertexWeight(0) = -1.0f;
\n-
318 tan1Mask.EdgeWeight(0) = 1.0f;
\n-
319 tan1Mask.EdgeWeight(1) = 0.0f;
\n-
320
\n-
321 tan2Mask.VertexWeight(0) = -1.0f;
\n-
322 tan2Mask.EdgeWeight(0) = 0.0f;
\n-
323 tan2Mask.EdgeWeight(1) = 1.0f;
\n-
324
\n-
325 for (int i = 2; i < valence; ++i) {
\n-
326 tan1Mask.EdgeWeight(i) = 0.0f;
\n-
327 tan2Mask.EdgeWeight(i) = 0.0f;
\n-
328 }
\n-
329}
\n-
330
\n-
331template <>
\n-
332template <typename VERTEX, typename MASK>
\n-
333inline void
\n-
334Scheme<SCHEME_CATMARK>::assignCreaseLimitTangentMasks(VERTEX const& vertex,
\n-
335 MASK& tan1Mask, MASK& tan2Mask, int const creaseEnds[2]) const {
\n-
336
\n-
337 typedef typename MASK::Weight Weight;
\n-
338
\n-
339 //
\n-
340 // First, the tangent along the crease:
\n-
341 // The first crease edge is considered the "leading" edge of the span
\n-
342 // of surface for which we are evaluating tangents and the second edge the
\n-
343 // "trailing edge". By convention, the tangent along the crease is oriented
\n-
344 // in the direction of the leading edge.
\n-
345 //
\n-
346 int numEdges = vertex.GetNumEdges();
\n-
347 int numFaces = vertex.GetNumFaces();
\n-
348
\n-
349 tan1Mask.SetNumVertexWeights(1);
\n-
350 tan1Mask.SetNumEdgeWeights(numEdges);
\n-
351 tan1Mask.SetNumFaceWeights(numFaces);
\n-
352 tan1Mask.SetFaceWeightsForFaceCenters(false);
\n-
353
\n-
354 tan1Mask.VertexWeight(0) = 0.0f;
\n-
355 for (int i = 0; i < numEdges; ++i) {
\n-
356 tan1Mask.EdgeWeight(i) = 0.0f;
\n-
357 }
\n-
358 for (int i = 0; i < numFaces; ++i) {
\n-
359 tan1Mask.FaceWeight(i) = 0.0f;
\n-
360 }
\n-
361
\n-
362 tan1Mask.EdgeWeight(creaseEnds[0]) = 0.5f;
\n-
363 tan1Mask.EdgeWeight(creaseEnds[1]) = -0.5f;
\n-
364
\n-
365 //
\n-
366 // Second, the tangent across the interior faces:
\n-
367 // Note this is ambiguous for an interior vertex. We currently return
\n-
368 // the tangent for the surface in the counter-clockwise span between the
\n-
369 // leading and trailing edges that form the crease. Given the expected
\n-
370 // computation of a surface normal as Tan1 X Tan2, this tangent should be
\n-
371 // oriented "inward" from the crease/boundary -- across the surface rather
\n-
372 // than outward and away from it.
\n-
373 //
\n-
374 tan2Mask.SetNumVertexWeights(1);
\n-
375 tan2Mask.SetNumEdgeWeights(numEdges);
\n-
376 tan2Mask.SetNumFaceWeights(numFaces);
\n-
377 tan2Mask.SetFaceWeightsForFaceCenters(false);
\n-
378
\n-
379 // Prepend weights of 0 preceding the crease:
\n-
380 for (int i = 0; i < creaseEnds[0]; ++i) {
\n-
381 tan2Mask.EdgeWeight(i) = 0.0f;
\n-
382 tan2Mask.FaceWeight(i) = 0.0f;
\n-
383 }
\n-
384
\n-
385 // Assign weights to crease edge and interior points:
\n-
386 int interiorEdgeCount = creaseEnds[1] - creaseEnds[0] - 1;
\n-
387 if (interiorEdgeCount == 1) {
\n-
388 // The regular case -- uniform B-spline cross-tangent:
\n-
389
\n-
390 tan2Mask.VertexWeight(0) = (Weight)(-4.0 / 6.0);
\n-
391
\n-
392 tan2Mask.EdgeWeight(creaseEnds[0]) = (Weight)(-1.0 / 6.0);
\n-
393 tan2Mask.EdgeWeight(creaseEnds[0] + 1) = (Weight)( 4.0 / 6.0);
\n-
394 tan2Mask.EdgeWeight(creaseEnds[1]) = (Weight)(-1.0 / 6.0);
\n-
395
\n-
396 tan2Mask.FaceWeight(creaseEnds[0]) = (Weight)(1.0 / 6.0);
\n-
397 tan2Mask.FaceWeight(creaseEnds[0] + 1) = (Weight)(1.0 / 6.0);
\n-
398 } else if (interiorEdgeCount > 1) {
\n-
399 // The irregular case -- formulae from Biermann et al:
\n-
400
\n-
401 double k = (double) (interiorEdgeCount + 1);
\n-
402 double theta = M_PI / k;
\n-
403
\n-
404 double cosTheta = std::cos(theta);
\n-
405 double sinTheta = std::sin(theta);
\n-
406
\n-
407 // Loop/Schaefer use a different divisor here (3*k + cos(theta)):
\n-
408 double commonDenom = 1.0f / (k * (3.0f + cosTheta));
\n-
409 double R = (cosTheta + 1.0f) / sinTheta;
\n-
410
\n-
411 double vertexWeight = 4.0f * R * (cosTheta - 1.0f);
\n-
412 double creaseWeight = -R * (1.0f + 2.0f * cosTheta);
\n-
413
\n-
414 tan2Mask.VertexWeight(0) = (Weight) (vertexWeight * commonDenom);
\n-
415
\n-
416 tan2Mask.EdgeWeight(creaseEnds[0]) = (Weight) (creaseWeight * commonDenom);
\n-
417 tan2Mask.EdgeWeight(creaseEnds[1]) = (Weight) (creaseWeight * commonDenom);
\n-
418
\n-
419 tan2Mask.FaceWeight(creaseEnds[0]) = (Weight) (sinTheta * commonDenom);
\n-
420
\n-
421 double sinThetaI = 0.0f;
\n-
422 double sinThetaIplus1 = sinTheta;
\n-
423 for (int i = 1; i < k; ++i) {
\n-
424 sinThetaI = sinThetaIplus1;
\n-
425 sinThetaIplus1 = std::sin((i+1)*theta);
\n-
426
\n-
427 tan2Mask.EdgeWeight(creaseEnds[0] + i) = (Weight) ((4.0f * sinThetaI) * commonDenom);
\n-
428 tan2Mask.FaceWeight(creaseEnds[0] + i) = (Weight) ((sinThetaI + sinThetaIplus1) * commonDenom);
\n-
429 }
\n-
430 } else {
\n-
431 // Special case for a single face -- simple average of boundary edges:
\n-
432
\n-
433 tan2Mask.VertexWeight(0) = -6.0f;
\n-
434
\n-
435 tan2Mask.EdgeWeight(creaseEnds[0]) = 3.0f;
\n-
436 tan2Mask.EdgeWeight(creaseEnds[1]) = 3.0f;
\n-
437
\n-
438 tan2Mask.FaceWeight(creaseEnds[0]) = 0.0f;
\n-
439 }
\n-
440
\n-
441 // Append weights of 0 following the crease:
\n-
442 for (int i = creaseEnds[1]; i < numFaces; ++i) {
\n-
443 tan2Mask.FaceWeight(i) = 0.0f;
\n-
444 }
\n-
445 for (int i = creaseEnds[1] + 1; i < numEdges; ++i) {
\n-
446 tan2Mask.EdgeWeight(i) = 0.0f;
\n-
447 }
\n-
448}
\n-
449
\n-
450template <>
\n-
451template <typename VERTEX, typename MASK>
\n-
452inline void
\n-
453Scheme<SCHEME_CATMARK>::assignSmoothLimitTangentMasks(VERTEX const& vertex,
\n-
454 MASK& tan1Mask, MASK& tan2Mask) const {
\n-
455
\n-
456 typedef typename MASK::Weight Weight;
\n-
457
\n-
458 int valence = vertex.GetNumFaces();
\n-
459 if (valence == 2) {
\n-
460 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);
\n-
461 return;
\n-
462 }
\n-
463
\n-
464 // Compute tan1 initially -- tan2 is simply a rotation:
\n-
465 tan1Mask.SetNumVertexWeights(1);
\n-
466 tan1Mask.SetNumEdgeWeights(valence);
\n-
467 tan1Mask.SetNumFaceWeights(valence);
\n-
468 tan1Mask.SetFaceWeightsForFaceCenters(false);
\n-
469
\n-
470 tan1Mask.VertexWeight(0) = 0.0f;
\n-
471
\n-
472 if (valence == 4) {
\n-
473 tan1Mask.EdgeWeight(0) = 4.0f;
\n-
474 tan1Mask.EdgeWeight(1) = 0.0f;
\n-
475 tan1Mask.EdgeWeight(2) = -4.0f;
\n-
476 tan1Mask.EdgeWeight(3) = 0.0f;
\n-
477
\n-
478 tan1Mask.FaceWeight(0) = 1.0f;
\n-
479 tan1Mask.FaceWeight(1) = -1.0f;
\n-
480 tan1Mask.FaceWeight(2) = -1.0f;
\n-
481 tan1Mask.FaceWeight(3) = 1.0f;
\n-
482 } else {
\n-
483 double theta = 2.0f * M_PI / (double)valence;
\n-
484
\n-
485 double cosTheta = std::cos(theta);
\n-
486 double cosHalfTheta = std::cos(theta * 0.5f);
\n-
487
\n-
488 double lambda = (5.0 / 16.0) + (1.0 / 16.0) *
\n-
489 (cosTheta + cosHalfTheta * std::sqrt(2.0f * (9.0f + cosTheta)));
\n-
490
\n-
491 double edgeWeightScale = 4.0f;
\n-
492 double faceWeightScale = 1.0f / (4.0f * lambda - 1.0f);
\n-
493
\n-
494 for (int i = 0; i < valence; ++i) {
\n-
495 double cosThetaI = std::cos( i * theta);
\n-
496 double cosThetaIplus1 = std::cos((i+1)* theta);
\n-
497
\n-
498 tan1Mask.EdgeWeight(i) = (Weight) (edgeWeightScale * cosThetaI);
\n-
499 tan1Mask.FaceWeight(i) = (Weight) (faceWeightScale * (cosThetaI + cosThetaIplus1));
\n-
500 }
\n-
501 }
\n-
502
\n-
503 // Now rotate/copy tan1 weights to tan2:
\n-
504 tan2Mask.SetNumVertexWeights(1);
\n-
505 tan2Mask.SetNumEdgeWeights(valence);
\n-
506 tan2Mask.SetNumFaceWeights(valence);
\n-
507 tan2Mask.SetFaceWeightsForFaceCenters(false);
\n-
508
\n-
509 tan2Mask.VertexWeight(0) = 0.0f;
\n-
510 if (valence == 4) {
\n-
511 tan2Mask.EdgeWeight(0) = 0.0f;
\n-
512 tan2Mask.EdgeWeight(1) = 4.0f;
\n-
513 tan2Mask.EdgeWeight(2) = 0.0f;
\n-
514 tan2Mask.EdgeWeight(3) = -4.0f;
\n-
515
\n-
516 tan2Mask.FaceWeight(0) = 1.0f;
\n-
517 tan2Mask.FaceWeight(1) = 1.0f;
\n-
518 tan2Mask.FaceWeight(2) = -1.0f;
\n-
519 tan2Mask.FaceWeight(3) = -1.0f;
\n-
520 } else {
\n-
521 tan2Mask.EdgeWeight(0) = tan1Mask.EdgeWeight(valence-1);
\n-
522 tan2Mask.FaceWeight(0) = tan1Mask.FaceWeight(valence-1);
\n-
523 for (int i = 1; i < valence; ++i) {
\n-
524 tan2Mask.EdgeWeight(i) = tan1Mask.EdgeWeight(i-1);
\n-
525 tan2Mask.FaceWeight(i) = tan1Mask.FaceWeight(i-1);
\n-
526 }
\n-
527 }
\n-
528}
\n-
529
\n-
530} // end namespace sdc
\n-
531
\n-
532} // end namespace OPENSUBDIV_VERSION
\n-
533using namespace OPENSUBDIV_VERSION;
\n-
534} // end namespace OpenSubdiv
\n-
535
\n-
536#endif /* OPENSUBDIV3_SDC_CATMARK_SCHEME_H */
\n+
140template <>
\n+
141template <typename VERTEX, typename MASK>
\n+
142inline void
\n+
143Scheme<SCHEME_BILINEAR>::assignCreaseLimitTangentMasks(VERTEX const& vertex,
\n+
144 MASK& tan1Mask, MASK& tan2Mask, int const /* creaseEnds */[2]) const {
\n+
145
\n+
146 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);
\n+
147}
\n+
148
\n+
149template <>
\n+
150template <typename VERTEX, typename MASK>
\n+
151inline void
\n+
152Scheme<SCHEME_BILINEAR>::assignSmoothLimitTangentMasks(VERTEX const& vertex,
\n+
153 MASK& tan1Mask, MASK& tan2Mask) const {
\n+
154
\n+
155 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);
\n+
156}
\n+
157
\n+
158} // end namespace sdc
\n+
159
\n+
160} // end namespace OPENSUBDIV_VERSION
\n+
161using namespace OPENSUBDIV_VERSION;
\n+
162} // end namespace OpenSubdiv
\n+
163
\n+
164#endif /* OPENSUBDIV3_SDC_BILINEAR_SCHEME_H */
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Split
Split
Enumerated type for all face splitting schemes.
Definition: types.h:47
\n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SPLIT_TO_QUADS
@ SPLIT_TO_QUADS
Used by Catmark and Bilinear.
Definition: types.h:48
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Options::TRI_SUB_SMOOTH
@ TRI_SUB_SMOOTH
"smooth triangle" weights (Catmark scheme only)
Definition: options.h:74
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothMaskForEdge
void assignSmoothMaskForEdge(EDGE const &edge, MASK &mask) const
\n+\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseMaskForVertex
void assignCreaseMaskForVertex(VERTEX const &edge, MASK &mask, int const creaseEnds[2]) const
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::ComputeVertexVertexMask
void ComputeVertexVertexMask(VERTEX const &vertexNeighborhood, MASK &vertexVertexMask, Crease::Rule parentRule=Crease::RULE_UNKNOWN, Crease::Rule childRule=Crease::RULE_UNKNOWN) const
Vertex-vertex masks If known, a single Rule or pair of Rules can be specified (indicating a crease tr...
Definition: scheme.h:512
\n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseLimitTangentMasks
void assignCreaseLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK &tan2, int const creaseEnds[2]) const
\n \n+
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::ComputeEdgeVertexMask
void ComputeEdgeVertexMask(EDGE const &edgeNeighborhood, MASK &edgeVertexMask, Crease::Rule parentRule=Crease::RULE_UNKNOWN, Crease::Rule childRule=Crease::RULE_UNKNOWN) const
Edge-vertex masks If known, the Rule for the edge and/or the derived vertex can be specified to accel...
Definition: scheme.h:392
\n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothLimitMask
void assignSmoothLimitMask(VERTEX const &vertex, MASK &pos) const
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseLimitMask
void assignCreaseLimitMask(VERTEX const &vertex, MASK &pos, int const creaseEnds[2]) const
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothMaskForVertex
void assignSmoothMaskForVertex(VERTEX const &edge, MASK &mask) const
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothLimitTangentMasks
void assignSmoothLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK &tan2) const
\n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCornerLimitMask
void assignCornerLimitMask(VERTEX const &vertex, MASK &pos) const
\n
OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCornerLimitTangentMasks
void assignCornerLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK &tan2) const
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-catmarkScheme.h\n+bilinearScheme.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2014 DreamWorks Animation LLC.\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -29,595 +29,211 @@\n 17//\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n- 24#ifndef OPENSUBDIV3_SDC_CATMARK_SCHEME_H\n- 25#define OPENSUBDIV3_SDC_CATMARK_SCHEME_H\n+ 24#ifndef OPENSUBDIV3_SDC_BILINEAR_SCHEME_H\n+ 25#define OPENSUBDIV3_SDC_BILINEAR_SCHEME_H\n 26\n 27#include \"../version.h\"\n 28\n 29#include \"../sdc/scheme.h\"\n 30\n- 31#include \n- 32#include \n+ 31namespace OpenSubdiv {\n+ 32namespace OPENSUBDIV_VERSION {\n 33\n- 34namespace OpenSubdiv {\n- 35namespace OPENSUBDIV_VERSION {\n- 36\n- 37namespace Sdc {\n- 38\n- 39//\n- 40// Specializations for Scheme:\n- 41//\n- 42\n- 43//\n- 44// Catmark traits:\n- 45//\n- 46template <>\n-47inline Split Scheme::GetTopologicalSplitType() { return\n+34namespace Sdc {\n+ 35\n+ 36//\n+ 37// Specializations for Scheme:\n+ 38//\n+ 39\n+ 40//\n+ 41// Bilinear traits:\n+ 42//\n+ 43template <>\n+44inline Split Scheme::GetTopologicalSplitType() { return\n SPLIT_TO_QUADS; }\n+ 45\n+ 46template <>\n+47inline int Scheme::GetRegularFaceSize() { return 4; }\n 48\n 49template <>\n-50inline int Scheme::GetRegularFaceSize() { return 4; }\n+50inline int Scheme::GetRegularVertexValence() { return 4; }\n 51\n 52template <>\n-53inline int Scheme::GetRegularVertexValence() { return 4; }\n+53inline int Scheme::GetLocalNeighborhoodSize() { return 0; }\n 54\n- 55template <>\n-56inline int Scheme::GetLocalNeighborhoodSize() { return 1; }\n- 57\n- 58\n- 59//\n- 60// Masks for edge-vertices: the hard Crease mask does not need to be\n-specialized\n- 61// (simply the midpoint), so all that is left is the Smooth case:\n- 62//\n- 63// The Smooth mask is complicated by the need to support the \"triangle\n-subdivision\"\n- 64// option, which applies different weighting in the presence of triangles.\n-It is\n- 65// up for debate as to whether this is useful or not -- we may be able to\n-deprecate\n- 66// this option.\n- 67//\n+ 55\n+ 56//\n+ 57// Refinement masks:\n+ 58//\n+ 59template <>\n+ 60template \n+ 61void\n+62Scheme::ComputeEdgeVertexMask(EDGE const& edge, MASK& mask,\n+ 63 Crease::Rule, Crease::Rule) const {\n+ 64 // This should be inline, otherwise trivially replicate it:\n+ 65 assignCreaseMaskForEdge(edge, mask);\n+ 66}\n+ 67\n 68template <>\n- 69template \n- 70inline void\n-71Scheme::assignSmoothMaskForEdge(EDGE const& edge, MASK& mask)\n-const {\n- 72\n- 73 typedef typename MASK::Weight Weight;\n- 74\n- 75 int faceCount = edge.GetNumFaces();\n+ 69template \n+ 70void\n+71Scheme::ComputeVertexVertexMask(VERTEX const& vertex, MASK&\n+mask,\n+ 72 Crease::Rule, Crease::Rule) const {\n+ 73 // This should be inline, otherwise trivially replicate it:\n+ 74 assignCornerMaskForVertex(vertex, mask);\n+ 75}\n 76\n- 77 mask.SetNumVertexWeights(2);\n- 78 mask.SetNumEdgeWeights(0);\n- 79 mask.SetNumFaceWeights(faceCount);\n- 80 mask.SetFaceWeightsForFaceCenters(true);\n- 81\n- 82 //\n- 83 // Determine if we need to inspect incident faces and apply alternate\n-weighting for\n- 84 // triangles -- and if so, determine which of the two are triangles.\n- 85 //\n- 86 bool face0IsTri = false;\n- 87 bool face1IsTri = false;\n- 88 bool useTriangleOption = (_options.GetTriangleSubdivision() == Options::\n-TRI_SUB_SMOOTH);\n- 89 if (useTriangleOption) {\n- 90 if (faceCount == 2) {\n- 91 //\n- 92 // Ideally we want to avoid this inspection when we have already subdivided\n-at\n- 93 // least once -- need something in the Edge interface to help avoid this,\n-e.g.\n- 94 // an IsRegular() query, the subdivision level...\n- 95 //\n- 96 int vertsPerFace[2];\n- 97 edge.GetNumVerticesPerFace(vertsPerFace);\n- 98\n- 99 face0IsTri = (vertsPerFace[0] == 3);\n- 100 face1IsTri = (vertsPerFace[1] == 3);\n- 101 useTriangleOption = face0IsTri || face1IsTri;\n- 102 } else {\n- 103 useTriangleOption = false;\n- 104 }\n- 105 }\n- 106\n- 107 if (! useTriangleOption) {\n- 108 mask.VertexWeight(0) = 0.25f;\n- 109 mask.VertexWeight(1) = 0.25f;\n+ 77\n+ 78//\n+ 79// Limit masks for position -- the limit position of all vertices is the\n+refined vertex.\n+ 80//\n+ 81template <>\n+ 82template \n+ 83inline void\n+84Scheme::assignCornerLimitMask(VERTEX const& /* vertex */,\n+MASK& posMask) const {\n+ 85\n+ 86 posMask.SetNumVertexWeights(1);\n+ 87 posMask.SetNumEdgeWeights(0);\n+ 88 posMask.SetNumFaceWeights(0);\n+ 89 posMask.SetFaceWeightsForFaceCenters(false);\n+ 90\n+ 91 posMask.VertexWeight(0) = 1.0f;\n+ 92}\n+ 93\n+ 94template <>\n+ 95template \n+ 96inline void\n+97Scheme::assignCreaseLimitMask(VERTEX const& vertex, MASK&\n+posMask,\n+ 98 int const /* creaseEnds */[2]) const {\n+ 99\n+ 100 assignCornerLimitMask(vertex, posMask);\n+ 101}\n+ 102\n+ 103template <>\n+ 104template \n+ 105inline void\n+106Scheme::assignSmoothLimitMask(VERTEX const& vertex, MASK&\n+posMask) const {\n+ 107\n+ 108 assignCornerLimitMask(vertex, posMask);\n+ 109}\n 110\n- 111 if (faceCount == 2) {\n- 112 mask.FaceWeight(0) = 0.25f;\n- 113 mask.FaceWeight(1) = 0.25f;\n- 114 } else {\n- 115 Weight fWeight = 0.5f / (Weight)faceCount;\n- 116 for (int i = 0; i < faceCount; ++i) {\n- 117 mask.FaceWeight(i) = fWeight;\n- 118 }\n- 119 }\n- 120 } else {\n- 121 //\n- 122 // This mimics the implementation in Hbr in terms of order of operations.\n- 123 //\n- 124 const Weight CATMARK_SMOOTH_TRI_EDGE_WEIGHT = (Weight) 0.470;\n+ 111//\n+ 112// Limit masks for tangents -- these are ambiguous around all vertices.\n+Provide\n+ 113// the tangents based on the incident edges of the first face.\n+ 114//\n+ 115template <>\n+ 116template \n+ 117inline void\n+118Scheme::assignCornerLimitTangentMasks(VERTEX const& /\n+* vertex */,\n+ 119 MASK& tan1Mask, MASK& tan2Mask) const {\n+ 120\n+ 121 tan1Mask.SetNumVertexWeights(1);\n+ 122 tan1Mask.SetNumEdgeWeights(2);\n+ 123 tan1Mask.SetNumFaceWeights(0);\n+ 124 tan1Mask.SetFaceWeightsForFaceCenters(false);\n 125\n- 126 Weight f0Weight = face0IsTri ? CATMARK_SMOOTH_TRI_EDGE_WEIGHT : 0.25f;\n- 127 Weight f1Weight = face1IsTri ? CATMARK_SMOOTH_TRI_EDGE_WEIGHT : 0.25f;\n- 128\n- 129 Weight fWeight = 0.5f * (f0Weight + f1Weight);\n- 130 Weight vWeight = 0.5f * (1.0f - 2.0f * fWeight);\n- 131\n- 132 mask.VertexWeight(0) = vWeight;\n- 133 mask.VertexWeight(1) = vWeight;\n+ 126 tan2Mask.SetNumVertexWeights(1);\n+ 127 tan2Mask.SetNumEdgeWeights(2);\n+ 128 tan2Mask.SetNumFaceWeights(0);\n+ 129 tan2Mask.SetFaceWeightsForFaceCenters(false);\n+ 130\n+ 131 tan1Mask.VertexWeight(0) = -1.0f;\n+ 132 tan1Mask.EdgeWeight(0) = 1.0f;\n+ 133 tan1Mask.EdgeWeight(1) = 0.0f;\n 134\n- 135 mask.FaceWeight(0) = fWeight;\n- 136 mask.FaceWeight(1) = fWeight;\n- 137 }\n+ 135 tan2Mask.VertexWeight(0) = -1.0f;\n+ 136 tan2Mask.EdgeWeight(0) = 0.0f;\n+ 137 tan2Mask.EdgeWeight(1) = 1.0f;\n 138}\n 139\n- 140\n- 141//\n- 142// Masks for vertex-vertices: the hard Corner mask does not need to be\n-specialized\n- 143// (simply the vertex itself), leaving the Crease and Smooth cases (Dart is\n-smooth):\n- 144//\n- 145template <>\n- 146template \n- 147inline void\n-148Scheme::assignCreaseMaskForVertex(VERTEX const& vertex,\n-MASK& mask,\n- 149 int const creaseEnds[2]) const {\n- 150 typedef typename MASK::Weight Weight;\n- 151\n- 152 int valence = vertex.GetNumEdges();\n- 153\n- 154 mask.SetNumVertexWeights(1);\n- 155 mask.SetNumEdgeWeights(valence);\n- 156 mask.SetNumFaceWeights(0);\n- 157 mask.SetFaceWeightsForFaceCenters(false);\n- 158\n- 159 Weight vWeight = 0.75f;\n- 160 Weight eWeight = 0.125f;\n- 161\n- 162 mask.VertexWeight(0) = vWeight;\n- 163 for (int i = 0; i < valence; ++i) {\n- 164 mask.EdgeWeight(i) = 0.0f;\n- 165 }\n- 166 mask.EdgeWeight(creaseEnds[0]) = eWeight;\n- 167 mask.EdgeWeight(creaseEnds[1]) = eWeight;\n- 168}\n- 169\n- 170template <>\n- 171template \n- 172inline void\n-173Scheme::assignSmoothMaskForVertex(VERTEX const& vertex,\n-MASK& mask) const {\n- 174\n- 175 typedef typename MASK::Weight Weight;\n- 176\n- 177 //\n- 178 // A Smooth vertex must be manifold and interior -- manifold boundary\n-vertices will be\n- 179 // Creases and non-manifold vertices of any kind will be Corners or\n-Creases. If smooth\n- 180 // rules for non-manifold vertices are ever defined, this will need\n-adjusting:\n- 181 //\n- 182 assert(vertex.GetNumFaces() == vertex.GetNumEdges());\n- 183\n- 184 int valence = vertex.GetNumFaces();\n- 185\n- 186 mask.SetNumVertexWeights(1);\n- 187 mask.SetNumEdgeWeights(valence);\n- 188 mask.SetNumFaceWeights(valence);\n- 189 mask.SetFaceWeightsForFaceCenters(true);\n- 190\n- 191 Weight vWeight = (Weight)(valence - 2) / (Weight)valence;\n- 192 Weight fWeight = 1.0f / (Weight)(valence * valence);\n- 193 Weight eWeight = fWeight;\n- 194\n- 195 mask.VertexWeight(0) = vWeight;\n- 196 for (int i = 0; i < valence; ++i) {\n- 197 mask.EdgeWeight(i) = eWeight;\n- 198 mask.FaceWeight(i) = fWeight;\n- 199 }\n- 200}\n- 201\n- 202//\n- 203// Limit masks for position:\n- 204//\n- 205template <>\n- 206template \n- 207inline void\n-208Scheme::assignCornerLimitMask(VERTEX const& /* vertex */,\n-MASK& posMask) const {\n- 209\n- 210 posMask.SetNumVertexWeights(1);\n- 211 posMask.SetNumEdgeWeights(0);\n- 212 posMask.SetNumFaceWeights(0);\n- 213 posMask.SetFaceWeightsForFaceCenters(false);\n- 214\n- 215 posMask.VertexWeight(0) = 1.0f;\n- 216}\n- 217\n- 218template <>\n- 219template \n- 220inline void\n-221Scheme::assignCreaseLimitMask(VERTEX const& vertex, MASK&\n-posMask,\n- 222 int const creaseEnds[2]) const {\n- 223\n- 224 typedef typename MASK::Weight Weight;\n- 225\n- 226 int valence = vertex.GetNumEdges();\n- 227\n- 228 posMask.SetNumVertexWeights(1);\n- 229 posMask.SetNumEdgeWeights(valence);\n- 230 posMask.SetNumFaceWeights(0);\n- 231 posMask.SetFaceWeightsForFaceCenters(false);\n- 232\n- 233 Weight vWeight = (Weight)(2.0 / 3.0);\n- 234 Weight eWeight = (Weight)(1.0 / 6.0);\n- 235\n- 236 posMask.VertexWeight(0) = vWeight;\n- 237 for (int i = 0; i < valence; ++i) {\n- 238 posMask.EdgeWeight(i) = 0.0f;\n- 239 }\n- 240 posMask.EdgeWeight(creaseEnds[0]) = eWeight;\n- 241 posMask.EdgeWeight(creaseEnds[1]) = eWeight;\n- 242}\n- 243\n- 244template <>\n- 245template \n- 246inline void\n-247Scheme::assignSmoothLimitMask(VERTEX const& vertex, MASK&\n-posMask) const {\n- 248\n- 249 typedef typename MASK::Weight Weight;\n- 250\n- 251 int valence = vertex.GetNumFaces();\n- 252 if (valence == 2) {\n- 253 assignCornerLimitMask(vertex, posMask);\n- 254 return;\n- 255 }\n- 256\n- 257 posMask.SetNumVertexWeights(1);\n- 258 posMask.SetNumEdgeWeights(valence);\n- 259 posMask.SetNumFaceWeights(valence);\n- 260 posMask.SetFaceWeightsForFaceCenters(false);\n- 261\n- 262 // Specialize for the regular case:\n- 263 if (valence == 4) {\n- 264 Weight fWeight = (Weight)(1.0 / 36.0);\n- 265 Weight eWeight = (Weight)(1.0 / 9.0);\n- 266 Weight vWeight = (Weight)(4.0 / 9.0);\n- 267\n- 268 posMask.VertexWeight(0) = vWeight;\n- 269\n- 270 posMask.EdgeWeight(0) = eWeight;\n- 271 posMask.EdgeWeight(1) = eWeight;\n- 272 posMask.EdgeWeight(2) = eWeight;\n- 273 posMask.EdgeWeight(3) = eWeight;\n- 274\n- 275 posMask.FaceWeight(0) = fWeight;\n- 276 posMask.FaceWeight(1) = fWeight;\n- 277 posMask.FaceWeight(2) = fWeight;\n- 278 posMask.FaceWeight(3) = fWeight;\n- 279 } else {\n- 280 Weight Valence = (Weight) valence;\n- 281\n- 282 Weight fWeight = 1.0f / (Valence * (Valence + 5.0f));\n- 283 Weight eWeight = 4.0f * fWeight;\n- 284 Weight vWeight = 1.0f - Valence * (eWeight + fWeight);\n- 285\n- 286 posMask.VertexWeight(0) = vWeight;\n- 287 for (int i = 0; i < valence; ++i) {\n- 288 posMask.EdgeWeight(i) = eWeight;\n- 289 posMask.FaceWeight(i) = fWeight;\n- 290 }\n- 291 }\n- 292}\n- 293\n- 294//\n- 295// Limit masks for tangents -- these are stubs for now, or have a temporary\n- 296// implementation\n- 297//\n- 298template <>\n- 299template \n- 300inline void\n-301Scheme::assignCornerLimitTangentMasks(VERTEX const& vertex,\n- 302 MASK& tan1Mask, MASK& tan2Mask) const {\n- 303\n- 304 int valence = vertex.GetNumEdges();\n- 305\n- 306 tan1Mask.SetNumVertexWeights(1);\n- 307 tan1Mask.SetNumEdgeWeights(valence);\n- 308 tan1Mask.SetNumFaceWeights(0);\n- 309 tan1Mask.SetFaceWeightsForFaceCenters(false);\n- 310\n- 311 tan2Mask.SetNumVertexWeights(1);\n- 312 tan2Mask.SetNumEdgeWeights(valence);\n- 313 tan2Mask.SetNumFaceWeights(0);\n- 314 tan2Mask.SetFaceWeightsForFaceCenters(false);\n- 315\n- 316 // Should be at least 2 edges -- be sure to clear weights for any more:\n- 317 tan1Mask.VertexWeight(0) = -1.0f;\n- 318 tan1Mask.EdgeWeight(0) = 1.0f;\n- 319 tan1Mask.EdgeWeight(1) = 0.0f;\n- 320\n- 321 tan2Mask.VertexWeight(0) = -1.0f;\n- 322 tan2Mask.EdgeWeight(0) = 0.0f;\n- 323 tan2Mask.EdgeWeight(1) = 1.0f;\n- 324\n- 325 for (int i = 2; i < valence; ++i) {\n- 326 tan1Mask.EdgeWeight(i) = 0.0f;\n- 327 tan2Mask.EdgeWeight(i) = 0.0f;\n- 328 }\n- 329}\n- 330\n- 331template <>\n- 332template \n- 333inline void\n-334Scheme::assignCreaseLimitTangentMasks(VERTEX const& vertex,\n- 335 MASK& tan1Mask, MASK& tan2Mask, int const creaseEnds[2]) const {\n- 336\n- 337 typedef typename MASK::Weight Weight;\n- 338\n- 339 //\n- 340 // First, the tangent along the crease:\n- 341 // The first crease edge is considered the \"leading\" edge of the span\n- 342 // of surface for which we are evaluating tangents and the second edge the\n- 343 // \"trailing edge\". By convention, the tangent along the crease is\n-oriented\n- 344 // in the direction of the leading edge.\n- 345 //\n- 346 int numEdges = vertex.GetNumEdges();\n- 347 int numFaces = vertex.GetNumFaces();\n- 348\n- 349 tan1Mask.SetNumVertexWeights(1);\n- 350 tan1Mask.SetNumEdgeWeights(numEdges);\n- 351 tan1Mask.SetNumFaceWeights(numFaces);\n- 352 tan1Mask.SetFaceWeightsForFaceCenters(false);\n- 353\n- 354 tan1Mask.VertexWeight(0) = 0.0f;\n- 355 for (int i = 0; i < numEdges; ++i) {\n- 356 tan1Mask.EdgeWeight(i) = 0.0f;\n- 357 }\n- 358 for (int i = 0; i < numFaces; ++i) {\n- 359 tan1Mask.FaceWeight(i) = 0.0f;\n- 360 }\n- 361\n- 362 tan1Mask.EdgeWeight(creaseEnds[0]) = 0.5f;\n- 363 tan1Mask.EdgeWeight(creaseEnds[1]) = -0.5f;\n- 364\n- 365 //\n- 366 // Second, the tangent across the interior faces:\n- 367 // Note this is ambiguous for an interior vertex. We currently return\n- 368 // the tangent for the surface in the counter-clockwise span between the\n- 369 // leading and trailing edges that form the crease. Given the expected\n- 370 // computation of a surface normal as Tan1 X Tan2, this tangent should be\n- 371 // oriented \"inward\" from the crease/boundary -- across the surface rather\n- 372 // than outward and away from it.\n- 373 //\n- 374 tan2Mask.SetNumVertexWeights(1);\n- 375 tan2Mask.SetNumEdgeWeights(numEdges);\n- 376 tan2Mask.SetNumFaceWeights(numFaces);\n- 377 tan2Mask.SetFaceWeightsForFaceCenters(false);\n- 378\n- 379 // Prepend weights of 0 preceding the crease:\n- 380 for (int i = 0; i < creaseEnds[0]; ++i) {\n- 381 tan2Mask.EdgeWeight(i) = 0.0f;\n- 382 tan2Mask.FaceWeight(i) = 0.0f;\n- 383 }\n- 384\n- 385 // Assign weights to crease edge and interior points:\n- 386 int interiorEdgeCount = creaseEnds[1] - creaseEnds[0] - 1;\n- 387 if (interiorEdgeCount == 1) {\n- 388 // The regular case -- uniform B-spline cross-tangent:\n- 389\n- 390 tan2Mask.VertexWeight(0) = (Weight)(-4.0 / 6.0);\n- 391\n- 392 tan2Mask.EdgeWeight(creaseEnds[0]) = (Weight)(-1.0 / 6.0);\n- 393 tan2Mask.EdgeWeight(creaseEnds[0] + 1) = (Weight)( 4.0 / 6.0);\n- 394 tan2Mask.EdgeWeight(creaseEnds[1]) = (Weight)(-1.0 / 6.0);\n- 395\n- 396 tan2Mask.FaceWeight(creaseEnds[0]) = (Weight)(1.0 / 6.0);\n- 397 tan2Mask.FaceWeight(creaseEnds[0] + 1) = (Weight)(1.0 / 6.0);\n- 398 } else if (interiorEdgeCount > 1) {\n- 399 // The irregular case -- formulae from Biermann et al:\n- 400\n- 401 double k = (double) (interiorEdgeCount + 1);\n- 402 double theta = M_PI / k;\n- 403\n- 404 double cosTheta = std::cos(theta);\n- 405 double sinTheta = std::sin(theta);\n- 406\n- 407 // Loop/Schaefer use a different divisor here (3*k + cos(theta)):\n- 408 double commonDenom = 1.0f / (k * (3.0f + cosTheta));\n- 409 double R = (cosTheta + 1.0f) / sinTheta;\n- 410\n- 411 double vertexWeight = 4.0f * R * (cosTheta - 1.0f);\n- 412 double creaseWeight = -R * (1.0f + 2.0f * cosTheta);\n- 413\n- 414 tan2Mask.VertexWeight(0) = (Weight) (vertexWeight * commonDenom);\n- 415\n- 416 tan2Mask.EdgeWeight(creaseEnds[0]) = (Weight) (creaseWeight *\n-commonDenom);\n- 417 tan2Mask.EdgeWeight(creaseEnds[1]) = (Weight) (creaseWeight *\n-commonDenom);\n- 418\n- 419 tan2Mask.FaceWeight(creaseEnds[0]) = (Weight) (sinTheta * commonDenom);\n- 420\n- 421 double sinThetaI = 0.0f;\n- 422 double sinThetaIplus1 = sinTheta;\n- 423 for (int i = 1; i < k; ++i) {\n- 424 sinThetaI = sinThetaIplus1;\n- 425 sinThetaIplus1 = std::sin((i+1)*theta);\n- 426\n- 427 tan2Mask.EdgeWeight(creaseEnds[0] + i) = (Weight) ((4.0f * sinThetaI) *\n-commonDenom);\n- 428 tan2Mask.FaceWeight(creaseEnds[0] + i) = (Weight) ((sinThetaI +\n-sinThetaIplus1) * commonDenom);\n- 429 }\n- 430 } else {\n- 431 // Special case for a single face -- simple average of boundary edges:\n- 432\n- 433 tan2Mask.VertexWeight(0) = -6.0f;\n- 434\n- 435 tan2Mask.EdgeWeight(creaseEnds[0]) = 3.0f;\n- 436 tan2Mask.EdgeWeight(creaseEnds[1]) = 3.0f;\n- 437\n- 438 tan2Mask.FaceWeight(creaseEnds[0]) = 0.0f;\n- 439 }\n- 440\n- 441 // Append weights of 0 following the crease:\n- 442 for (int i = creaseEnds[1]; i < numFaces; ++i) {\n- 443 tan2Mask.FaceWeight(i) = 0.0f;\n- 444 }\n- 445 for (int i = creaseEnds[1] + 1; i < numEdges; ++i) {\n- 446 tan2Mask.EdgeWeight(i) = 0.0f;\n- 447 }\n- 448}\n- 449\n- 450template <>\n- 451template \n- 452inline void\n-453Scheme::assignSmoothLimitTangentMasks(VERTEX const& vertex,\n- 454 MASK& tan1Mask, MASK& tan2Mask) const {\n- 455\n- 456 typedef typename MASK::Weight Weight;\n- 457\n- 458 int valence = vertex.GetNumFaces();\n- 459 if (valence == 2) {\n- 460 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);\n- 461 return;\n- 462 }\n- 463\n- 464 // Compute tan1 initially -- tan2 is simply a rotation:\n- 465 tan1Mask.SetNumVertexWeights(1);\n- 466 tan1Mask.SetNumEdgeWeights(valence);\n- 467 tan1Mask.SetNumFaceWeights(valence);\n- 468 tan1Mask.SetFaceWeightsForFaceCenters(false);\n- 469\n- 470 tan1Mask.VertexWeight(0) = 0.0f;\n- 471\n- 472 if (valence == 4) {\n- 473 tan1Mask.EdgeWeight(0) = 4.0f;\n- 474 tan1Mask.EdgeWeight(1) = 0.0f;\n- 475 tan1Mask.EdgeWeight(2) = -4.0f;\n- 476 tan1Mask.EdgeWeight(3) = 0.0f;\n- 477\n- 478 tan1Mask.FaceWeight(0) = 1.0f;\n- 479 tan1Mask.FaceWeight(1) = -1.0f;\n- 480 tan1Mask.FaceWeight(2) = -1.0f;\n- 481 tan1Mask.FaceWeight(3) = 1.0f;\n- 482 } else {\n- 483 double theta = 2.0f * M_PI / (double)valence;\n- 484\n- 485 double cosTheta = std::cos(theta);\n- 486 double cosHalfTheta = std::cos(theta * 0.5f);\n- 487\n- 488 double lambda = (5.0 / 16.0) + (1.0 / 16.0) *\n- 489 (cosTheta + cosHalfTheta * std::sqrt(2.0f * (9.0f + cosTheta)));\n- 490\n- 491 double edgeWeightScale = 4.0f;\n- 492 double faceWeightScale = 1.0f / (4.0f * lambda - 1.0f);\n- 493\n- 494 for (int i = 0; i < valence; ++i) {\n- 495 double cosThetaI = std::cos( i * theta);\n- 496 double cosThetaIplus1 = std::cos((i+1)* theta);\n- 497\n- 498 tan1Mask.EdgeWeight(i) = (Weight) (edgeWeightScale * cosThetaI);\n- 499 tan1Mask.FaceWeight(i) = (Weight) (faceWeightScale * (cosThetaI +\n-cosThetaIplus1));\n- 500 }\n- 501 }\n- 502\n- 503 // Now rotate/copy tan1 weights to tan2:\n- 504 tan2Mask.SetNumVertexWeights(1);\n- 505 tan2Mask.SetNumEdgeWeights(valence);\n- 506 tan2Mask.SetNumFaceWeights(valence);\n- 507 tan2Mask.SetFaceWeightsForFaceCenters(false);\n- 508\n- 509 tan2Mask.VertexWeight(0) = 0.0f;\n- 510 if (valence == 4) {\n- 511 tan2Mask.EdgeWeight(0) = 0.0f;\n- 512 tan2Mask.EdgeWeight(1) = 4.0f;\n- 513 tan2Mask.EdgeWeight(2) = 0.0f;\n- 514 tan2Mask.EdgeWeight(3) = -4.0f;\n- 515\n- 516 tan2Mask.FaceWeight(0) = 1.0f;\n- 517 tan2Mask.FaceWeight(1) = 1.0f;\n- 518 tan2Mask.FaceWeight(2) = -1.0f;\n- 519 tan2Mask.FaceWeight(3) = -1.0f;\n- 520 } else {\n- 521 tan2Mask.EdgeWeight(0) = tan1Mask.EdgeWeight(valence-1);\n- 522 tan2Mask.FaceWeight(0) = tan1Mask.FaceWeight(valence-1);\n- 523 for (int i = 1; i < valence; ++i) {\n- 524 tan2Mask.EdgeWeight(i) = tan1Mask.EdgeWeight(i-1);\n- 525 tan2Mask.FaceWeight(i) = tan1Mask.FaceWeight(i-1);\n- 526 }\n- 527 }\n- 528}\n- 529\n- 530} // end namespace sdc\n- 531\n- 532} // end namespace OPENSUBDIV_VERSION\n- 533using namespace OPENSUBDIV_VERSION;\n- 534} // end namespace OpenSubdiv\n- 535\n- 536#endif /* OPENSUBDIV3_SDC_CATMARK_SCHEME_H */\n+ 140template <>\n+ 141template \n+ 142inline void\n+143Scheme::assignCreaseLimitTangentMasks(VERTEX const& vertex,\n+ 144 MASK& tan1Mask, MASK& tan2Mask, int const /* creaseEnds */[2]) const {\n+ 145\n+ 146 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);\n+ 147}\n+ 148\n+ 149template <>\n+ 150template \n+ 151inline void\n+152Scheme::assignSmoothLimitTangentMasks(VERTEX const& vertex,\n+ 153 MASK& tan1Mask, MASK& tan2Mask) const {\n+ 154\n+ 155 assignCornerLimitTangentMasks(vertex, tan1Mask, tan2Mask);\n+ 156}\n+ 157\n+ 158} // end namespace sdc\n+ 159\n+ 160} // end namespace OPENSUBDIV_VERSION\n+ 161using namespace OPENSUBDIV_VERSION;\n+ 162} // end namespace OpenSubdiv\n+ 163\n+ 164#endif /* OPENSUBDIV3_SDC_BILINEAR_SCHEME_H */\n OpenSubdiv\n Definition: limits.h:32\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Split\n Split\n Enumerated type for all face splitting schemes.\n Definition: types.h:47\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::SPLIT_TO_QUADS\n @ SPLIT_TO_QUADS\n Used by Catmark and Bilinear.\n Definition: types.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Options::TRI_SUB_SMOOTH\n-@ TRI_SUB_SMOOTH\n-\"smooth triangle\" weights (Catmark scheme only)\n-Definition: options.h:74\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothMaskForEdge\n-void assignSmoothMaskForEdge(EDGE const &edge, MASK &mask) const\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::Rule\n+Rule\n+Definition: crease.h:82\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::GetTopologicalSplitType\n static Split GetTopologicalSplitType()\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseMaskForVertex\n-void assignCreaseMaskForVertex(VERTEX const &edge, MASK &mask, int const\n-creaseEnds[2]) const\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::ComputeVertexVertexMask\n+void ComputeVertexVertexMask(VERTEX const &vertexNeighborhood, MASK\n+&vertexVertexMask, Crease::Rule parentRule=Crease::RULE_UNKNOWN, Crease::Rule\n+childRule=Crease::RULE_UNKNOWN) const\n+Vertex-vertex masks If known, a single Rule or pair of Rules can be specified\n+(indicating a crease tr...\n+Definition: scheme.h:512\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseLimitTangentMasks\n void assignCreaseLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK\n &tan2, int const creaseEnds[2]) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::GetRegularFaceSize\n static int GetRegularFaceSize()\n+OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::ComputeEdgeVertexMask\n+void ComputeEdgeVertexMask(EDGE const &edgeNeighborhood, MASK &edgeVertexMask,\n+Crease::Rule parentRule=Crease::RULE_UNKNOWN, Crease::Rule childRule=Crease::\n+RULE_UNKNOWN) const\n+Edge-vertex masks If known, the Rule for the edge and/or the derived vertex can\n+be specified to accel...\n+Definition: scheme.h:392\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothLimitMask\n void assignSmoothLimitMask(VERTEX const &vertex, MASK &pos) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::GetLocalNeighborhoodSize\n static int GetLocalNeighborhoodSize()\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCreaseLimitMask\n void assignCreaseLimitMask(VERTEX const &vertex, MASK &pos, int const\n creaseEnds[2]) const\n-OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothMaskForVertex\n-void assignSmoothMaskForVertex(VERTEX const &edge, MASK &mask) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::GetRegularVertexValence\n static int GetRegularVertexValence()\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignSmoothLimitTangentMasks\n void assignSmoothLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK\n &tan2) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCornerLimitMask\n void assignCornerLimitMask(VERTEX const &vertex, MASK &pos) const\n OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Scheme::assignCornerLimitTangentMasks\n void assignCornerLimitTangentMasks(VERTEX const &vertex, MASK &tan1, MASK\n &tan2) const\n * opensubdiv\n * sdc\n- * catmarkScheme.h\n+ * bilinearScheme.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00788.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00788.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cudaPatchTable.h File Reference\n+OpenSubdiv: opensubdiv/osd/opencl.h File Reference\n \n \n \n \n \n \n \n@@ -80,48 +80,24 @@\n
\n \n
\n \n
\n-
\n-Classes |\n-Namespaces
\n-
cudaPatchTable.h File Reference
\n+
opencl.h File Reference
\n
\n
\n-
#include "../version.h"
\n-#include "../osd/nonCopyable.h"
\n-#include "../osd/types.h"
\n-#include <vector>
\n+
#include <CL/opencl.h>
\n
\n

Go to the source code of this file.

\n-\n-\n-\n-\n-\n-

\n-Classes

class  CudaPatchTable
 CUDA patch table. More...
 
\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-

\n-Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n
\n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,31 +4,14 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-Classes | Namespaces\n-cudaPatchTable.h File Reference\n-#include \"../version.h\"\n-#include \"../osd/nonCopyable.h\"\n-#include \"../osd/types.h\"\n-#include \n+opencl.h File Reference\n+#include \n Go_to_the_source_code_of_this_file.\n- Classes\n-class \u00a0CudaPatchTable\n-\u00a0 CUDA patch table. More...\n-\u00a0\n- Namespaces\n-namespace \u00a0OpenSubdiv\n-\u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n-\u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n-\u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n-\u00a0\n * opensubdiv\n * osd\n- * cudaPatchTable.h\n+ * opencl.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00788_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00788_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cudaPatchTable.h Source File\n+OpenSubdiv: opensubdiv/osd/opencl.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
cudaPatchTable.h
\n+
opencl.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2015 Pixar
\n+
2// Copyright 2014 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,119 +107,32 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H
\n-
26#define OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H
\n+
25#ifndef OPENSUBDIV3_OSD_OPENCL_H
\n+
26#define OPENSUBDIV3_OSD_OPENCL_H
\n
27
\n-
28#include "../version.h"
\n-
29
\n-
30#include "../osd/nonCopyable.h"
\n-
31#include "../osd/types.h"
\n-
32
\n-
33#include <vector>
\n-
34
\n-
35namespace OpenSubdiv {
\n-
36namespace OPENSUBDIV_VERSION {
\n+
28#if defined(OPENSUBDIV_HAS_CLEW)
\n+
29# include <clew.h>
\n+
30#else
\n+
31# if defined(__APPLE__)
\n+
32# include <OpenCL/opencl.h>
\n+
33# else
\n+
34# include <CL/opencl.h>
\n+
35# endif
\n+
36#endif
\n
37
\n-
38namespace Far{
\n-
39 class PatchTable;
\n-
40};
\n-
41
\n-
42namespace Osd {
\n-
43
\n-
51class CudaPatchTable : private NonCopyable<CudaPatchTable> {
\n-
52public:
\n-
54 static CudaPatchTable *Create(Far::PatchTable const *patchTable,
\n-
55 void *deviceContext = NULL);
\n-
57 ~CudaPatchTable();
\n-
58
\n-
60 void *GetPatchArrayBuffer() const { return _patchArrays; }
\n-
61
\n-
63 void *GetPatchIndexBuffer() const { return _indexBuffer; }
\n-
64
\n-
66 void *GetPatchParamBuffer() const { return _patchParamBuffer; }
\n-
67
\n-
69 void *GetVaryingPatchArrayBuffer() const {
\n-
70 return _varyingPatchArrays;
\n-
71 }
\n-
73 void *GetVaryingPatchIndexBuffer() const {
\n-
74 return _varyingIndexBuffer;
\n-
75 }
\n-
76
\n-
78 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }
\n-
79
\n-
81 void *GetFVarPatchArrayBuffer(int fvarChannel) const {
\n-
82 return _fvarPatchArrays[fvarChannel];
\n-
83 }
\n-
84
\n-
86 void *GetFVarPatchIndexBuffer(int fvarChannel = 0) const {
\n-
87 return _fvarIndexBuffers[fvarChannel];
\n-
88 }
\n-
89
\n-
91 void *GetFVarPatchParamBuffer(int fvarChannel = 0) const {
\n-
92 return _fvarParamBuffers[fvarChannel];
\n-
93 }
\n-
94
\n-
95protected:
\n-
96 CudaPatchTable();
\n-
97
\n-
98 bool allocate(Far::PatchTable const *patchTable);
\n-
99
\n-
100 void *_patchArrays;
\n-
101 void *_indexBuffer;
\n-
102 void *_patchParamBuffer;
\n-
103
\n-
104 void *_varyingPatchArrays;
\n-
105 void *_varyingIndexBuffer;
\n-
106
\n-
107 std::vector<void *> _fvarPatchArrays;
\n-
108 std::vector<void *> _fvarIndexBuffers;
\n-
109 std::vector<void *> _fvarParamBuffers;
\n-
110};
\n-
111
\n-
112} // end namespace Osd
\n-
113
\n-
114} // end namespace OPENSUBDIV_VERSION
\n-
115using namespace OPENSUBDIV_VERSION;
\n-
116
\n-
117} // end namespace OpenSubdiv
\n-
118
\n-
119#endif // OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchArrayBuffer
void * GetFVarPatchArrayBuffer(int fvarChannel) const
Returns the cuda memory of the array of Osd::PatchArray buffer.
Definition: cudaPatchTable.h:81
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::Create
static CudaPatchTable * Create(Far::PatchTable const *patchTable, void *deviceContext=NULL)
Creator. Returns NULL if error.
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetVaryingPatchIndexBuffer
void * GetVaryingPatchIndexBuffer() const
Returns the cuda memory of the array of varying control vertices.
Definition: cudaPatchTable.h:73
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetVaryingPatchArrayBuffer
void * GetVaryingPatchArrayBuffer() const
Returns the cuda memory of the array of Osd::PatchArray buffer.
Definition: cudaPatchTable.h:69
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchArrayBuffer
void * GetPatchArrayBuffer() const
Returns the cuda memory of the array of Osd::PatchArray buffer.
Definition: cudaPatchTable.h:60
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::allocate
bool allocate(Far::PatchTable const *patchTable)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchIndexBuffer
void * GetPatchIndexBuffer() const
Returns the cuda memory of the patch control vertices.
Definition: cudaPatchTable.h:63
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetNumFVarChannels
int GetNumFVarChannels() const
Returns the number of face-varying channels buffers.
Definition: cudaPatchTable.h:78
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchParamBuffer
void * GetPatchParamBuffer() const
Returns the cuda memory of the array of Osd::PatchParam buffer.
Definition: cudaPatchTable.h:66
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchIndexBuffer
void * GetFVarPatchIndexBuffer(int fvarChannel=0) const
Returns the cuda memory of the array of face-varying control vertices.
Definition: cudaPatchTable.h:86
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchParamBuffer
void * GetFVarPatchParamBuffer(int fvarChannel=0) const
Returns the cuda memory of the array of face-varying param.
Definition: cudaPatchTable.h:91
\n-\n-\n-\n+
38#endif // OPENSUBDIV3_OSD_CL_UTIL_H
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-cudaPatchTable.h\n+opencl.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2015 Pixar\n+ 2// Copyright 2014 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,167 +30,25 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H\n- 26#define OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H\n+ 25#ifndef OPENSUBDIV3_OSD_OPENCL_H\n+ 26#define OPENSUBDIV3_OSD_OPENCL_H\n 27\n- 28#include \"../version.h\"\n- 29\n- 30#include \"../osd/nonCopyable.h\"\n- 31#include \"../osd/types.h\"\n- 32\n- 33#include \n- 34\n- 35namespace OpenSubdiv {\n- 36namespace OPENSUBDIV_VERSION {\n+ 28#if defined(OPENSUBDIV_HAS_CLEW)\n+ 29# include \n+ 30#else\n+ 31# if defined(__APPLE__)\n+ 32# include \n+ 33# else\n+ 34# include \n+ 35# endif\n+ 36#endif\n 37\n- 38namespace Far{\n- 39 class PatchTable;\n- 40};\n- 41\n- 42namespace Osd {\n- 43\n-51class CudaPatchTable : private NonCopyable {\n- 52public:\n-54 static CudaPatchTable *Create(Far::PatchTable const *patchTable,\n- 55 void *deviceContext = NULL);\n-57 ~CudaPatchTable();\n- 58\n-60 void *GetPatchArrayBuffer() const { return _patchArrays; }\n- 61\n-63 void *GetPatchIndexBuffer() const { return _indexBuffer; }\n- 64\n-66 void *GetPatchParamBuffer() const { return _patchParamBuffer; }\n- 67\n-69 void *GetVaryingPatchArrayBuffer() const {\n- 70 return _varyingPatchArrays;\n- 71 }\n-73 void *GetVaryingPatchIndexBuffer() const {\n- 74 return _varyingIndexBuffer;\n- 75 }\n- 76\n-78 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }\n- 79\n-81 void *GetFVarPatchArrayBuffer(int fvarChannel) const {\n- 82 return _fvarPatchArrays[fvarChannel];\n- 83 }\n- 84\n-86 void *GetFVarPatchIndexBuffer(int fvarChannel = 0) const {\n- 87 return _fvarIndexBuffers[fvarChannel];\n- 88 }\n- 89\n-91 void *GetFVarPatchParamBuffer(int fvarChannel = 0) const {\n- 92 return _fvarParamBuffers[fvarChannel];\n- 93 }\n- 94\n- 95protected:\n-96 CudaPatchTable();\n- 97\n-98 bool allocate(Far::PatchTable const *patchTable);\n- 99\n-100 void *_patchArrays;\n-101 void *_indexBuffer;\n-102 void *_patchParamBuffer;\n- 103\n-104 void *_varyingPatchArrays;\n-105 void *_varyingIndexBuffer;\n- 106\n-107 std::vector _fvarPatchArrays;\n-108 std::vector _fvarIndexBuffers;\n-109 std::vector _fvarParamBuffers;\n- 110};\n- 111\n- 112} // end namespace Osd\n- 113\n- 114} // end namespace OPENSUBDIV_VERSION\n- 115using namespace OPENSUBDIV_VERSION;\n- 116\n- 117} // end namespace OpenSubdiv\n- 118\n- 119#endif // OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H\n-OpenSubdiv\n-Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n-Container for arrays of parametric patches.\n-Definition: patchTable.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable\n-CUDA patch table.\n-Definition: cudaPatchTable.h:51\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchArrayBuffer\n-void * GetFVarPatchArrayBuffer(int fvarChannel) const\n-Returns the cuda memory of the array of Osd::PatchArray buffer.\n-Definition: cudaPatchTable.h:81\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_fvarParamBuffers\n-std::vector< void * > _fvarParamBuffers\n-Definition: cudaPatchTable.h:109\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::Create\n-static CudaPatchTable * Create(Far::PatchTable const *patchTable, void\n-*deviceContext=NULL)\n-Creator. Returns NULL if error.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_varyingPatchArrays\n-void * _varyingPatchArrays\n-Definition: cudaPatchTable.h:104\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetVaryingPatchIndexBuffer\n-void * GetVaryingPatchIndexBuffer() const\n-Returns the cuda memory of the array of varying control vertices.\n-Definition: cudaPatchTable.h:73\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_fvarIndexBuffers\n-std::vector< void * > _fvarIndexBuffers\n-Definition: cudaPatchTable.h:108\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetVaryingPatchArrayBuffer\n-void * GetVaryingPatchArrayBuffer() const\n-Returns the cuda memory of the array of Osd::PatchArray buffer.\n-Definition: cudaPatchTable.h:69\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_varyingIndexBuffer\n-void * _varyingIndexBuffer\n-Definition: cudaPatchTable.h:105\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::~CudaPatchTable\n-~CudaPatchTable()\n-Destructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchArrayBuffer\n-void * GetPatchArrayBuffer() const\n-Returns the cuda memory of the array of Osd::PatchArray buffer.\n-Definition: cudaPatchTable.h:60\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_fvarPatchArrays\n-std::vector< void * > _fvarPatchArrays\n-Definition: cudaPatchTable.h:107\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::allocate\n-bool allocate(Far::PatchTable const *patchTable)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchIndexBuffer\n-void * GetPatchIndexBuffer() const\n-Returns the cuda memory of the patch control vertices.\n-Definition: cudaPatchTable.h:63\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetNumFVarChannels\n-int GetNumFVarChannels() const\n-Returns the number of face-varying channels buffers.\n-Definition: cudaPatchTable.h:78\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_patchArrays\n-void * _patchArrays\n-Definition: cudaPatchTable.h:100\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchParamBuffer\n-void * GetPatchParamBuffer() const\n-Returns the cuda memory of the array of Osd::PatchParam buffer.\n-Definition: cudaPatchTable.h:66\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchIndexBuffer\n-void * GetFVarPatchIndexBuffer(int fvarChannel=0) const\n-Returns the cuda memory of the array of face-varying control vertices.\n-Definition: cudaPatchTable.h:86\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchParamBuffer\n-void * GetFVarPatchParamBuffer(int fvarChannel=0) const\n-Returns the cuda memory of the array of face-varying param.\n-Definition: cudaPatchTable.h:91\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::CudaPatchTable\n-CudaPatchTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_patchParamBuffer\n-void * _patchParamBuffer\n-Definition: cudaPatchTable.h:102\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_indexBuffer\n-void * _indexBuffer\n-Definition: cudaPatchTable.h:101\n+ 38#endif // OPENSUBDIV3_OSD_CL_UTIL_H\n * opensubdiv\n * osd\n- * cudaPatchTable.h\n+ * opencl.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00791.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00791.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cudaEvaluator.h File Reference\n+OpenSubdiv: opensubdiv/osd/cudaPatchTable.h File Reference\n \n \n \n \n \n \n \n@@ -83,30 +83,28 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
cudaEvaluator.h File Reference
\n+
cudaPatchTable.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include <vector>
\n-#include "../osd/bufferDescriptor.h"
\n+#include "../osd/nonCopyable.h"
\n #include "../osd/types.h"
\n+#include <vector>
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n-\n-\n+\n+\n \n

\n Classes

class  CudaStencilTable
 CUDA stencil table. More...
 
class  CudaEvaluator
class  CudaPatchTable
 CUDA patch table. More...
 
\n \n \n \n \n@@ -117,13 +115,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,32 +5,30 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-cudaEvaluator.h File Reference\n+cudaPatchTable.h File Reference\n #include \"../version.h\"\n-#include \n-#include \"../osd/bufferDescriptor.h\"\n+#include \"../osd/nonCopyable.h\"\n #include \"../osd/types.h\"\n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0CudaStencilTable\n-\u00a0 CUDA stencil table. More...\n-\u00a0\n-class \u00a0CudaEvaluator\n+class \u00a0CudaPatchTable\n+\u00a0 CUDA patch table. More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * cudaEvaluator.h\n+ * cudaPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00791.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00791.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,3 @@\n var a00791 = [\n- [\"CudaStencilTable\", \"a01173.html\", \"a01173\"],\n- [\"CudaEvaluator\", \"a01177.html\", null]\n+ [\"CudaPatchTable\", \"a01181.html\", \"a01181\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00791_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00791_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cudaEvaluator.h Source File\n+OpenSubdiv: opensubdiv/osd/cudaPatchTable.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
cudaEvaluator.h
\n+
cudaPatchTable.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,550 +107,119 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_CUDA_EVALUATOR_H
\n-
26#define OPENSUBDIV3_OSD_CUDA_EVALUATOR_H
\n+
25#ifndef OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H
\n+
26#define OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30#include <vector>
\n-
31#include "../osd/bufferDescriptor.h"
\n-
32#include "../osd/types.h"
\n-
33
\n-
34namespace OpenSubdiv {
\n-
35namespace OPENSUBDIV_VERSION {
\n-
36
\n-
37namespace Far {
\n-
38 class PatchTable;
\n-
39 class StencilTable;
\n-
40 class LimitStencilTable;
\n-
41}
\n-
42
\n-
43namespace Osd {
\n-
44
\n-
52class CudaStencilTable {
\n-
53public:
\n-
54 static CudaStencilTable *Create(Far::StencilTable const *stencilTable,
\n-
55 void *deviceContext = NULL) {
\n-
56 (void)deviceContext; // unused
\n-
57 return new CudaStencilTable(stencilTable);
\n-
58 }
\n-
59 static CudaStencilTable *Create(Far::LimitStencilTable const *limitStencilTable,
\n-
60 void *deviceContext = NULL) {
\n-
61 (void)deviceContext; // unused
\n-
62 return new CudaStencilTable(limitStencilTable);
\n-
63 }
\n+
30#include "../osd/nonCopyable.h"
\n+
31#include "../osd/types.h"
\n+
32
\n+
33#include <vector>
\n+
34
\n+
35namespace OpenSubdiv {
\n+
36namespace OPENSUBDIV_VERSION {
\n+
37
\n+
38namespace Far{
\n+
39 class PatchTable;
\n+
40};
\n+
41
\n+
42namespace Osd {
\n+
43
\n+
51class CudaPatchTable : private NonCopyable<CudaPatchTable> {
\n+
52public:
\n+
54 static CudaPatchTable *Create(Far::PatchTable const *patchTable,
\n+
55 void *deviceContext = NULL);
\n+
57 ~CudaPatchTable();
\n+
58
\n+
60 void *GetPatchArrayBuffer() const { return _patchArrays; }
\n+
61
\n+
63 void *GetPatchIndexBuffer() const { return _indexBuffer; }
\n
64
\n-
65 explicit CudaStencilTable(Far::StencilTable const *stencilTable);
\n-
66 explicit CudaStencilTable(Far::LimitStencilTable const *limitStencilTable);
\n-
67 ~CudaStencilTable();
\n-
68
\n-
69 // interfaces needed for CudaCompute
\n-
70 void *GetSizesBuffer() const { return _sizes; }
\n-
71 void *GetOffsetsBuffer() const { return _offsets; }
\n-
72 void *GetIndicesBuffer() const { return _indices; }
\n-
73 void *GetWeightsBuffer() const { return _weights; }
\n-
74 void *GetDuWeightsBuffer() const { return _duWeights; }
\n-
75 void *GetDvWeightsBuffer() const { return _dvWeights; }
\n-
76 void *GetDuuWeightsBuffer() const { return _duuWeights; }
\n-
77 void *GetDuvWeightsBuffer() const { return _duvWeights; }
\n-
78 void *GetDvvWeightsBuffer() const { return _dvvWeights; }
\n-
79 int GetNumStencils() const { return _numStencils; }
\n-
80
\n-
81private:
\n-
82 void * _sizes,
\n-
83 * _offsets,
\n-
84 * _indices,
\n-
85 * _weights,
\n-
86 * _duWeights,
\n-
87 * _dvWeights,
\n-
88 * _duuWeights,
\n-
89 * _duvWeights,
\n-
90 * _dvvWeights;
\n-
91 int _numStencils;
\n-
92};
\n-
93
\n-
94class CudaEvaluator {
\n-
95public:
\n-
101
\n-
125 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
126 static bool EvalStencils(
\n-
127 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
128 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
129 STENCIL_TABLE const *stencilTable,
\n-
130 const void *instance = NULL,
\n-
131 void * deviceContext = NULL) {
\n-
132
\n-
133 (void)instance; // unused
\n-
134 (void)deviceContext; // unused
\n-
135 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
136 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
137 (int const *)stencilTable->GetSizesBuffer(),
\n-
138 (int const *)stencilTable->GetOffsetsBuffer(),
\n-
139 (int const *)stencilTable->GetIndicesBuffer(),
\n-
140 (float const *)stencilTable->GetWeightsBuffer(),
\n-
141 /*start = */ 0,
\n-
142 /*end = */ stencilTable->GetNumStencils());
\n-
143 }
\n-
144
\n-
171 static bool EvalStencils(
\n-
172 const float *src, BufferDescriptor const &srcDesc,
\n-
173 float *dst, BufferDescriptor const &dstDesc,
\n-
174 const int * sizes,
\n-
175 const int * offsets,
\n-
176 const int * indices,
\n-
177 const float * weights,
\n-
178 int start, int end);
\n-
179
\n-
217 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
218 static bool EvalStencils(
\n-
219 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
220 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
221 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
222 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
223 STENCIL_TABLE const *stencilTable,
\n-
224 const CudaEvaluator *instance = NULL,
\n-
225 void * deviceContext = NULL) {
\n-
226
\n-
227 (void)instance; // unused
\n-
228 (void)deviceContext; // unused
\n-
229
\n-
230 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
231 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
232 duBuffer->BindCudaBuffer(), duDesc,
\n-
233 dvBuffer->BindCudaBuffer(), dvDesc,
\n-
234 (int const *)stencilTable->GetSizesBuffer(),
\n-
235 (int const *)stencilTable->GetOffsetsBuffer(),
\n-
236 (int const *)stencilTable->GetIndicesBuffer(),
\n-
237 (float const *)stencilTable->GetWeightsBuffer(),
\n-
238 (float const *)stencilTable->GetDuWeightsBuffer(),
\n-
239 (float const *)stencilTable->GetDvWeightsBuffer(),
\n-
240 /*start = */ 0,
\n-
241 /*end = */ stencilTable->GetNumStencils());
\n-
242 }
\n-
243
\n-
284 static bool EvalStencils(
\n-
285 const float *src, BufferDescriptor const &srcDesc,
\n-
286 float *dst, BufferDescriptor const &dstDesc,
\n-
287 float *du, BufferDescriptor const &duDesc,
\n-
288 float *dv, BufferDescriptor const &dvDesc,
\n-
289 const int * sizes,
\n-
290 const int * offsets,
\n-
291 const int * indices,
\n-
292 const float * weights,
\n-
293 const float * duWeights,
\n-
294 const float * dvWeights,
\n-
295 int start, int end);
\n-
296
\n-
352 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
353 static bool EvalStencils(
\n-
354 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
355 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
356 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
357 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
358 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
359 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
360 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
361 STENCIL_TABLE const *stencilTable,
\n-
362 const CudaEvaluator *instance = NULL,
\n-
363 void * deviceContext = NULL) {
\n-
364
\n-
365 (void)instance; // unused
\n-
366 (void)deviceContext; // unused
\n-
367
\n-
368 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
369 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
370 duBuffer->BindCudaBuffer(), duDesc,
\n-
371 dvBuffer->BindCudaBuffer(), dvDesc,
\n-
372 duuBuffer->BindCudaBuffer(), duuDesc,
\n-
373 duvBuffer->BindCudaBuffer(), duvDesc,
\n-
374 dvvBuffer->BindCudaBuffer(), dvvDesc,
\n-
375 (int const *)stencilTable->GetSizesBuffer(),
\n-
376 (int const *)stencilTable->GetOffsetsBuffer(),
\n-
377 (int const *)stencilTable->GetIndicesBuffer(),
\n-
378 (float const *)stencilTable->GetWeightsBuffer(),
\n-
379 (float const *)stencilTable->GetDuWeightsBuffer(),
\n-
380 (float const *)stencilTable->GetDvWeightsBuffer(),
\n-
381 (float const *)stencilTable->GetDuuWeightsBuffer(),
\n-
382 (float const *)stencilTable->GetDuvWeightsBuffer(),
\n-
383 (float const *)stencilTable->GetDvvWeightsBuffer(),
\n-
384 /*start = */ 0,
\n-
385 /*end = */ stencilTable->GetNumStencils());
\n-
386 }
\n-
387
\n-
449 static bool EvalStencils(
\n-
450 const float *src, BufferDescriptor const &srcDesc,
\n-
451 float *dst, BufferDescriptor const &dstDesc,
\n-
452 float *du, BufferDescriptor const &duDesc,
\n-
453 float *dv, BufferDescriptor const &dvDesc,
\n-
454 float *duu, BufferDescriptor const &duuDesc,
\n-
455 float *duv, BufferDescriptor const &duvDesc,
\n-
456 float *dvv, BufferDescriptor const &dvvDesc,
\n-
457 const int * sizes,
\n-
458 const int * offsets,
\n-
459 const int * indices,
\n-
460 const float * weights,
\n-
461 const float * duWeights,
\n-
462 const float * dvWeights,
\n-
463 const float * duuWeights,
\n-
464 const float * duvWeights,
\n-
465 const float * dvvWeights,
\n-
466 int start, int end);
\n-
467
\n-
473
\n-
502 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
503 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
504 static bool EvalPatches(
\n-
505 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
506 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
507 int numPatchCoords,
\n-
508 PATCHCOORD_BUFFER *patchCoords,
\n-
509 PATCH_TABLE *patchTable,
\n-
510 CudaEvaluator const *instance,
\n-
511 void * deviceContext = NULL) {
\n-
512
\n-
513 (void)instance; // unused
\n-
514 (void)deviceContext; // unused
\n-
515
\n-
516 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
517 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
518 numPatchCoords,
\n-
519 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n-
520 (const PatchArray *)patchTable->GetPatchArrayBuffer(),
\n-
521 (const int *)patchTable->GetPatchIndexBuffer(),
\n-
522 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n-
523 }
\n-
524
\n-
563 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
564 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
565 static bool EvalPatches(
\n-
566 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
567 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
568 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
569 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
570 int numPatchCoords,
\n-
571 PATCHCOORD_BUFFER *patchCoords,
\n-
572 PATCH_TABLE *patchTable,
\n-
573 CudaEvaluator const *instance,
\n-
574 void * deviceContext = NULL) {
\n-
575
\n-
576 (void)instance; // unused
\n-
577 (void)deviceContext; // unused
\n-
578
\n-
579 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
580 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
581 duBuffer->BindCudaBuffer(), duDesc,
\n-
582 dvBuffer->BindCudaBuffer(), dvDesc,
\n-
583 numPatchCoords,
\n-
584 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n-
585 (const PatchArray *)patchTable->GetPatchArrayBuffer(),
\n-
586 (const int *)patchTable->GetPatchIndexBuffer(),
\n-
587 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n-
588 }
\n-
589
\n-
646 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
647 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
648 static bool EvalPatches(
\n-
649 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
650 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
651 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
652 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
653 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
654 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
655 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
656 int numPatchCoords,
\n-
657 PATCHCOORD_BUFFER *patchCoords,
\n-
658 PATCH_TABLE *patchTable,
\n-
659 CudaEvaluator const *instance,
\n-
660 void * deviceContext = NULL) {
\n-
661
\n-
662 (void)instance; // unused
\n-
663 (void)deviceContext; // unused
\n-
664
\n-
665 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
666 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
667 duBuffer->BindCudaBuffer(), duDesc,
\n-
668 dvBuffer->BindCudaBuffer(), dvDesc,
\n-
669 duuBuffer->BindCudaBuffer(), duuDesc,
\n-
670 duvBuffer->BindCudaBuffer(), duvDesc,
\n-
671 dvvBuffer->BindCudaBuffer(), dvvDesc,
\n-
672 numPatchCoords,
\n-
673 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n-
674 (const PatchArray *)patchTable->GetPatchArrayBuffer(),
\n-
675 (const int *)patchTable->GetPatchIndexBuffer(),
\n-
676 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n-
677 }
\n-
678
\n-
706 static bool EvalPatches(
\n-
707 const float *src, BufferDescriptor const &srcDesc,
\n-
708 float *dst, BufferDescriptor const &dstDesc,
\n-
709 int numPatchCoords,
\n-
710 const PatchCoord *patchCoords,
\n-
711 const PatchArray *patchArrays,
\n-
712 const int *patchIndices,
\n-
713 const PatchParam *patchParams);
\n-
714
\n-
752 static bool EvalPatches(
\n-
753 const float *src, BufferDescriptor const &srcDesc,
\n-
754 float *dst, BufferDescriptor const &dstDesc,
\n-
755 float *du, BufferDescriptor const &duDesc,
\n-
756 float *dv, BufferDescriptor const &dvDesc,
\n-
757 int numPatchCoords,
\n-
758 PatchCoord const *patchCoords,
\n-
759 PatchArray const *patchArrays,
\n-
760 const int *patchIndices,
\n-
761 PatchParam const *patchParams);
\n-
762
\n-
815 static bool EvalPatches(
\n-
816 const float *src, BufferDescriptor const &srcDesc,
\n-
817 float *dst, BufferDescriptor const &dstDesc,
\n-
818 float *du, BufferDescriptor const &duDesc,
\n-
819 float *dv, BufferDescriptor const &dvDesc,
\n-
820 float *duu, BufferDescriptor const &duuDesc,
\n-
821 float *duv, BufferDescriptor const &duvDesc,
\n-
822 float *dvv, BufferDescriptor const &dvvDesc,
\n-
823 int numPatchCoords,
\n-
824 PatchCoord const *patchCoords,
\n-
825 PatchArray const *patchArrays,
\n-
826 const int *patchIndices,
\n-
827 PatchParam const *patchParams);
\n-
828
\n-
857 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
858 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
859 static bool EvalPatchesVarying(
\n-
860 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
861 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
862 int numPatchCoords,
\n-
863 PATCHCOORD_BUFFER *patchCoords,
\n-
864 PATCH_TABLE *patchTable,
\n-
865 CudaEvaluator const *instance,
\n-
866 void * deviceContext = NULL) {
\n-
867
\n-
868 (void)instance; // unused
\n-
869 (void)deviceContext; // unused
\n-
870
\n-
871 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
872 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
873 numPatchCoords,
\n-
874 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n-
875 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),
\n-
876 (const int *)patchTable->GetVaryingPatchIndexBuffer(),
\n-
877 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n-
878 }
\n-
879
\n-
920 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
921 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
922 static bool EvalPatchesVarying(
\n-
923 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
924 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
925 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
926 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
927 int numPatchCoords,
\n-
928 PATCHCOORD_BUFFER *patchCoords,
\n-
929 PATCH_TABLE *patchTable,
\n-
930 CudaEvaluator const *instance,
\n-
931 void * deviceContext = NULL) {
\n-
932
\n-
933 (void)instance; // unused
\n-
934 (void)deviceContext; // unused
\n-
935
\n-
936 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
937 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
938 duBuffer->BindCudaBuffer(), duDesc,
\n-
939 dvBuffer->BindCudaBuffer(), dvDesc,
\n-
940 numPatchCoords,
\n-
941 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n-
942 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),
\n-
943 (const int *)patchTable->GetVaryingPatchIndexBuffer(),
\n-
944 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n-
945 }
\n-
946
\n-
1005 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1006 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1007 static bool EvalPatchesVarying(
\n-
1008 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1009 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1010 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1011 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1012 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1013 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1014 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1015 int numPatchCoords,
\n-
1016 PATCHCOORD_BUFFER *patchCoords,
\n-
1017 PATCH_TABLE *patchTable,
\n-
1018 CudaEvaluator const *instance,
\n-
1019 void * deviceContext = NULL) {
\n-
1020
\n-
1021 (void)instance; // unused
\n-
1022 (void)deviceContext; // unused
\n-
1023
\n-
1024 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
1025 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
1026 duBuffer->BindCudaBuffer(), duDesc,
\n-
1027 dvBuffer->BindCudaBuffer(), dvDesc,
\n-
1028 duuBuffer->BindCudaBuffer(), duuDesc,
\n-
1029 duvBuffer->BindCudaBuffer(), duvDesc,
\n-
1030 dvvBuffer->BindCudaBuffer(), dvvDesc,
\n-
1031 numPatchCoords,
\n-
1032 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n-
1033 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),
\n-
1034 (const int *)patchTable->GetVaryingPatchIndexBuffer(),
\n-
1035 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n-
1036 }
\n-
1037
\n-
1068 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1069 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1070 static bool EvalPatchesFaceVarying(
\n-
1071 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1072 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1073 int numPatchCoords,
\n-
1074 PATCHCOORD_BUFFER *patchCoords,
\n-
1075 PATCH_TABLE *patchTable,
\n-
1076 int fvarChannel,
\n-
1077 CudaEvaluator const *instance,
\n-
1078 void * deviceContext = NULL) {
\n-
1079
\n-
1080 (void)instance; // unused
\n-
1081 (void)deviceContext; // unused
\n-
1082
\n-
1083 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
1084 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
1085 numPatchCoords,
\n-
1086 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n-
1087 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
1088 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1089 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
1090 }
\n-
1091
\n-
1134 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1135 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1136 static bool EvalPatchesFaceVarying(
\n-
1137 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1138 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1139 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1140 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1141 int numPatchCoords,
\n-
1142 PATCHCOORD_BUFFER *patchCoords,
\n-
1143 PATCH_TABLE *patchTable,
\n-
1144 int fvarChannel,
\n-
1145 CudaEvaluator const *instance,
\n-
1146 void * deviceContext = NULL) {
\n-
1147
\n-
1148 (void)instance; // unused
\n-
1149 (void)deviceContext; // unused
\n-
1150
\n-
1151 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
1152 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
1153 duBuffer->BindCudaBuffer(), duDesc,
\n-
1154 dvBuffer->BindCudaBuffer(), dvDesc,
\n-
1155 numPatchCoords,
\n-
1156 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n-
1157 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
1158 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1159 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
1160 }
\n-
1161
\n-
1222 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1223 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1224 static bool EvalPatchesFaceVarying(
\n-
1225 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1226 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1227 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1228 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1229 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1230 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1231 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1232 int numPatchCoords,
\n-
1233 PATCHCOORD_BUFFER *patchCoords,
\n-
1234 PATCH_TABLE *patchTable,
\n-
1235 int fvarChannel,
\n-
1236 CudaEvaluator const *instance,
\n-
1237 void * deviceContext = NULL) {
\n-
1238
\n-
1239 (void)instance; // unused
\n-
1240 (void)deviceContext; // unused
\n-
1241
\n-
1242 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n-
1243 dstBuffer->BindCudaBuffer(), dstDesc,
\n-
1244 duBuffer->BindCudaBuffer(), duDesc,
\n-
1245 dvBuffer->BindCudaBuffer(), dvDesc,
\n-
1246 duuBuffer->BindCudaBuffer(), duuDesc,
\n-
1247 duvBuffer->BindCudaBuffer(), duvDesc,
\n-
1248 dvvBuffer->BindCudaBuffer(), dvvDesc,
\n-
1249 numPatchCoords,
\n-
1250 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n-
1251 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
1252 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1253 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
1254 }
\n-
1255
\n-
1261 static void Synchronize(void *deviceContext = NULL);
\n-
1262};
\n-
1263
\n-
1264
\n-
1265} // end namespace Osd
\n-
1266
\n-
1267} // end namespace OPENSUBDIV_VERSION
\n-
1268using namespace OPENSUBDIV_VERSION;
\n-
1269
\n-
1270} // end namespace OpenSubdiv
\n-
1271
\n-
1272
\n-
1273#endif // OPENSUBDIV3_OSD_CUDA_EVALUATOR_H
\n+
66 void *GetPatchParamBuffer() const { return _patchParamBuffer; }
\n+
67
\n+
69 void *GetVaryingPatchArrayBuffer() const {
\n+
70 return _varyingPatchArrays;
\n+
71 }
\n+
73 void *GetVaryingPatchIndexBuffer() const {
\n+
74 return _varyingIndexBuffer;
\n+
75 }
\n+
76
\n+
78 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }
\n+
79
\n+
81 void *GetFVarPatchArrayBuffer(int fvarChannel) const {
\n+
82 return _fvarPatchArrays[fvarChannel];
\n+
83 }
\n+
84
\n+
86 void *GetFVarPatchIndexBuffer(int fvarChannel = 0) const {
\n+
87 return _fvarIndexBuffers[fvarChannel];
\n+
88 }
\n+
89
\n+
91 void *GetFVarPatchParamBuffer(int fvarChannel = 0) const {
\n+
92 return _fvarParamBuffers[fvarChannel];
\n+
93 }
\n+
94
\n+
95protected:
\n+
96 CudaPatchTable();
\n+
97
\n+
98 bool allocate(Far::PatchTable const *patchTable);
\n+
99
\n+
100 void *_patchArrays;
\n+
101 void *_indexBuffer;
\n+
102 void *_patchParamBuffer;
\n+
103
\n+
104 void *_varyingPatchArrays;
\n+
105 void *_varyingIndexBuffer;
\n+
106
\n+
107 std::vector<void *> _fvarPatchArrays;
\n+
108 std::vector<void *> _fvarIndexBuffers;
\n+
109 std::vector<void *> _fvarParamBuffers;
\n+
110};
\n+
111
\n+
112} // end namespace Osd
\n+
113
\n+
114} // end namespace OPENSUBDIV_VERSION
\n+
115using namespace OPENSUBDIV_VERSION;
\n+
116
\n+
117} // end namespace OpenSubdiv
\n+
118
\n+
119#endif // OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::CudaStencilTable
CudaStencilTable(Far::LimitStencilTable const *limitStencilTable)
\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::Create
static CudaStencilTable * Create(Far::LimitStencilTable const *limitStencilTable, void *deviceContext=NULL)
Definition: cudaEvaluator.h:59
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::CudaStencilTable
CudaStencilTable(Far::StencilTable const *stencilTable)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::Create
static CudaStencilTable * Create(Far::StencilTable const *stencilTable, void *deviceContext=NULL)
Definition: cudaEvaluator.h:54
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrays, const int *patchIndices, const PatchParam *patchParams)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:922
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: cudaEvaluator.h:648
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, const CudaEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: cudaEvaluator.h:353
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, int start, int end)
Static eval stencils function which takes raw cuda buffers for input and output.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:859
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const CudaEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: cudaEvaluator.h:218
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndices, PatchParam const *patchParams)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:1136
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, const void *instance=NULL, void *deviceContext=NULL)
Generic static compute function. This function has a same signature as other device kernels have so t...
Definition: cudaEvaluator.h:126
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw cuda pointers for input and output.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, const float *duuWeights, const float *duvWeights, const float *dvvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw cuda pointers for input and output.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::Synchronize
static void Synchronize(void *deviceContext=NULL)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndices, PatchParam const *patchParams)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:1007
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:504
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:1224
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: cudaEvaluator.h:565
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:1070
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord
Coordinates set on a patch table.
Definition: types.h:42
\n-\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchArrayBuffer
void * GetFVarPatchArrayBuffer(int fvarChannel) const
Returns the cuda memory of the array of Osd::PatchArray buffer.
Definition: cudaPatchTable.h:81
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::Create
static CudaPatchTable * Create(Far::PatchTable const *patchTable, void *deviceContext=NULL)
Creator. Returns NULL if error.
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetVaryingPatchIndexBuffer
void * GetVaryingPatchIndexBuffer() const
Returns the cuda memory of the array of varying control vertices.
Definition: cudaPatchTable.h:73
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetVaryingPatchArrayBuffer
void * GetVaryingPatchArrayBuffer() const
Returns the cuda memory of the array of Osd::PatchArray buffer.
Definition: cudaPatchTable.h:69
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchArrayBuffer
void * GetPatchArrayBuffer() const
Returns the cuda memory of the array of Osd::PatchArray buffer.
Definition: cudaPatchTable.h:60
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::allocate
bool allocate(Far::PatchTable const *patchTable)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchIndexBuffer
void * GetPatchIndexBuffer() const
Returns the cuda memory of the patch control vertices.
Definition: cudaPatchTable.h:63
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetNumFVarChannels
int GetNumFVarChannels() const
Returns the number of face-varying channels buffers.
Definition: cudaPatchTable.h:78
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchParamBuffer
void * GetPatchParamBuffer() const
Returns the cuda memory of the array of Osd::PatchParam buffer.
Definition: cudaPatchTable.h:66
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchIndexBuffer
void * GetFVarPatchIndexBuffer(int fvarChannel=0) const
Returns the cuda memory of the array of face-varying control vertices.
Definition: cudaPatchTable.h:86
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchParamBuffer
void * GetFVarPatchParamBuffer(int fvarChannel=0) const
Returns the cuda memory of the array of face-varying param.
Definition: cudaPatchTable.h:91
\n+\n+\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-cudaEvaluator.h\n+cudaPatchTable.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,740 +30,167 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_CUDA_EVALUATOR_H\n- 26#define OPENSUBDIV3_OSD_CUDA_EVALUATOR_H\n+ 25#ifndef OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H\n+ 26#define OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H\n 27\n 28#include \"../version.h\"\n 29\n- 30#include \n- 31#include \"../osd/bufferDescriptor.h\"\n- 32#include \"../osd/types.h\"\n- 33\n- 34namespace OpenSubdiv {\n- 35namespace OPENSUBDIV_VERSION {\n- 36\n- 37namespace Far {\n- 38 class PatchTable;\n- 39 class StencilTable;\n- 40 class LimitStencilTable;\n- 41}\n- 42\n- 43namespace Osd {\n- 44\n-52class CudaStencilTable {\n- 53public:\n-54 static CudaStencilTable *Create(Far::StencilTable const *stencilTable,\n- 55 void *deviceContext = NULL) {\n- 56 (void)deviceContext; // unused\n- 57 return new CudaStencilTable(stencilTable);\n- 58 }\n-59 static CudaStencilTable *Create(Far::LimitStencilTable const\n-*limitStencilTable,\n- 60 void *deviceContext = NULL) {\n- 61 (void)deviceContext; // unused\n- 62 return new CudaStencilTable(limitStencilTable);\n- 63 }\n+ 30#include \"../osd/nonCopyable.h\"\n+ 31#include \"../osd/types.h\"\n+ 32\n+ 33#include \n+ 34\n+ 35namespace OpenSubdiv {\n+ 36namespace OPENSUBDIV_VERSION {\n+ 37\n+ 38namespace Far{\n+ 39 class PatchTable;\n+ 40};\n+ 41\n+ 42namespace Osd {\n+ 43\n+51class CudaPatchTable : private NonCopyable {\n+ 52public:\n+54 static CudaPatchTable *Create(Far::PatchTable const *patchTable,\n+ 55 void *deviceContext = NULL);\n+57 ~CudaPatchTable();\n+ 58\n+60 void *GetPatchArrayBuffer() const { return _patchArrays; }\n+ 61\n+63 void *GetPatchIndexBuffer() const { return _indexBuffer; }\n 64\n-65 explicit CudaStencilTable(Far::StencilTable const *stencilTable);\n-66 explicit CudaStencilTable(Far::LimitStencilTable const *limitStencilTable);\n-67 ~CudaStencilTable();\n- 68\n- 69 // interfaces needed for CudaCompute\n-70 void *GetSizesBuffer() const { return _sizes; }\n-71 void *GetOffsetsBuffer() const { return _offsets; }\n-72 void *GetIndicesBuffer() const { return _indices; }\n-73 void *GetWeightsBuffer() const { return _weights; }\n-74 void *GetDuWeightsBuffer() const { return _duWeights; }\n-75 void *GetDvWeightsBuffer() const { return _dvWeights; }\n-76 void *GetDuuWeightsBuffer() const { return _duuWeights; }\n-77 void *GetDuvWeightsBuffer() const { return _duvWeights; }\n-78 void *GetDvvWeightsBuffer() const { return _dvvWeights; }\n-79 int GetNumStencils() const { return _numStencils; }\n- 80\n- 81private:\n- 82 void * _sizes,\n- 83 * _offsets,\n- 84 * _indices,\n- 85 * _weights,\n- 86 * _duWeights,\n- 87 * _dvWeights,\n- 88 * _duuWeights,\n- 89 * _duvWeights,\n- 90 * _dvvWeights;\n- 91 int _numStencils;\n- 92};\n- 93\n-94class CudaEvaluator {\n- 95public:\n- 101\n- 125 template \n-126 static bool EvalStencils(\n- 127 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 128 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 129 STENCIL_TABLE const *stencilTable,\n- 130 const void *instance = NULL,\n- 131 void * deviceContext = NULL) {\n- 132\n- 133 (void)instance; // unused\n- 134 (void)deviceContext; // unused\n- 135 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,\n- 136 dstBuffer->BindCudaBuffer(), dstDesc,\n- 137 (int const *)stencilTable->GetSizesBuffer(),\n- 138 (int const *)stencilTable->GetOffsetsBuffer(),\n- 139 (int const *)stencilTable->GetIndicesBuffer(),\n- 140 (float const *)stencilTable->GetWeightsBuffer(),\n- 141 /*start = */ 0,\n- 142 /*end = */ stencilTable->GetNumStencils());\n- 143 }\n- 144\n-171 static bool EvalStencils(\n- 172 const float *src, BufferDescriptor const &srcDesc,\n- 173 float *dst, BufferDescriptor const &dstDesc,\n- 174 const int * sizes,\n- 175 const int * offsets,\n- 176 const int * indices,\n- 177 const float * weights,\n- 178 int start, int end);\n- 179\n- 217 template \n-218 static bool EvalStencils(\n- 219 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 220 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 221 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 222 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 223 STENCIL_TABLE const *stencilTable,\n- 224 const CudaEvaluator *instance = NULL,\n- 225 void * deviceContext = NULL) {\n- 226\n- 227 (void)instance; // unused\n- 228 (void)deviceContext; // unused\n- 229\n- 230 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,\n- 231 dstBuffer->BindCudaBuffer(), dstDesc,\n- 232 duBuffer->BindCudaBuffer(), duDesc,\n- 233 dvBuffer->BindCudaBuffer(), dvDesc,\n- 234 (int const *)stencilTable->GetSizesBuffer(),\n- 235 (int const *)stencilTable->GetOffsetsBuffer(),\n- 236 (int const *)stencilTable->GetIndicesBuffer(),\n- 237 (float const *)stencilTable->GetWeightsBuffer(),\n- 238 (float const *)stencilTable->GetDuWeightsBuffer(),\n- 239 (float const *)stencilTable->GetDvWeightsBuffer(),\n- 240 /*start = */ 0,\n- 241 /*end = */ stencilTable->GetNumStencils());\n- 242 }\n- 243\n-284 static bool EvalStencils(\n- 285 const float *src, BufferDescriptor const &srcDesc,\n- 286 float *dst, BufferDescriptor const &dstDesc,\n- 287 float *du, BufferDescriptor const &duDesc,\n- 288 float *dv, BufferDescriptor const &dvDesc,\n- 289 const int * sizes,\n- 290 const int * offsets,\n- 291 const int * indices,\n- 292 const float * weights,\n- 293 const float * duWeights,\n- 294 const float * dvWeights,\n- 295 int start, int end);\n- 296\n- 352 template \n-353 static bool EvalStencils(\n- 354 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 355 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 356 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 357 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 358 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 359 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 360 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 361 STENCIL_TABLE const *stencilTable,\n- 362 const CudaEvaluator *instance = NULL,\n- 363 void * deviceContext = NULL) {\n- 364\n- 365 (void)instance; // unused\n- 366 (void)deviceContext; // unused\n- 367\n- 368 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,\n- 369 dstBuffer->BindCudaBuffer(), dstDesc,\n- 370 duBuffer->BindCudaBuffer(), duDesc,\n- 371 dvBuffer->BindCudaBuffer(), dvDesc,\n- 372 duuBuffer->BindCudaBuffer(), duuDesc,\n- 373 duvBuffer->BindCudaBuffer(), duvDesc,\n- 374 dvvBuffer->BindCudaBuffer(), dvvDesc,\n- 375 (int const *)stencilTable->GetSizesBuffer(),\n- 376 (int const *)stencilTable->GetOffsetsBuffer(),\n- 377 (int const *)stencilTable->GetIndicesBuffer(),\n- 378 (float const *)stencilTable->GetWeightsBuffer(),\n- 379 (float const *)stencilTable->GetDuWeightsBuffer(),\n- 380 (float const *)stencilTable->GetDvWeightsBuffer(),\n- 381 (float const *)stencilTable->GetDuuWeightsBuffer(),\n- 382 (float const *)stencilTable->GetDuvWeightsBuffer(),\n- 383 (float const *)stencilTable->GetDvvWeightsBuffer(),\n- 384 /*start = */ 0,\n- 385 /*end = */ stencilTable->GetNumStencils());\n- 386 }\n- 387\n-449 static bool EvalStencils(\n- 450 const float *src, BufferDescriptor const &srcDesc,\n- 451 float *dst, BufferDescriptor const &dstDesc,\n- 452 float *du, BufferDescriptor const &duDesc,\n- 453 float *dv, BufferDescriptor const &dvDesc,\n- 454 float *duu, BufferDescriptor const &duuDesc,\n- 455 float *duv, BufferDescriptor const &duvDesc,\n- 456 float *dvv, BufferDescriptor const &dvvDesc,\n- 457 const int * sizes,\n- 458 const int * offsets,\n- 459 const int * indices,\n- 460 const float * weights,\n- 461 const float * duWeights,\n- 462 const float * dvWeights,\n- 463 const float * duuWeights,\n- 464 const float * duvWeights,\n- 465 const float * dvvWeights,\n- 466 int start, int end);\n- 467\n- 473\n- 502 template \n-504 static bool EvalPatches(\n- 505 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 506 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 507 int numPatchCoords,\n- 508 PATCHCOORD_BUFFER *patchCoords,\n- 509 PATCH_TABLE *patchTable,\n- 510 CudaEvaluator const *instance,\n- 511 void * deviceContext = NULL) {\n- 512\n- 513 (void)instance; // unused\n- 514 (void)deviceContext; // unused\n- 515\n- 516 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n- 517 dstBuffer->BindCudaBuffer(), dstDesc,\n- 518 numPatchCoords,\n- 519 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n- 520 (const PatchArray *)patchTable->GetPatchArrayBuffer(),\n- 521 (const int *)patchTable->GetPatchIndexBuffer(),\n- 522 (const PatchParam *)patchTable->GetPatchParamBuffer());\n- 523 }\n- 524\n- 563 template \n-565 static bool EvalPatches(\n- 566 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 567 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 568 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 569 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 570 int numPatchCoords,\n- 571 PATCHCOORD_BUFFER *patchCoords,\n- 572 PATCH_TABLE *patchTable,\n- 573 CudaEvaluator const *instance,\n- 574 void * deviceContext = NULL) {\n- 575\n- 576 (void)instance; // unused\n- 577 (void)deviceContext; // unused\n- 578\n- 579 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n- 580 dstBuffer->BindCudaBuffer(), dstDesc,\n- 581 duBuffer->BindCudaBuffer(), duDesc,\n- 582 dvBuffer->BindCudaBuffer(), dvDesc,\n- 583 numPatchCoords,\n- 584 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n- 585 (const PatchArray *)patchTable->GetPatchArrayBuffer(),\n- 586 (const int *)patchTable->GetPatchIndexBuffer(),\n- 587 (const PatchParam *)patchTable->GetPatchParamBuffer());\n- 588 }\n- 589\n- 646 template \n-648 static bool EvalPatches(\n- 649 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 650 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 651 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 652 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 653 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 654 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 655 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 656 int numPatchCoords,\n- 657 PATCHCOORD_BUFFER *patchCoords,\n- 658 PATCH_TABLE *patchTable,\n- 659 CudaEvaluator const *instance,\n- 660 void * deviceContext = NULL) {\n- 661\n- 662 (void)instance; // unused\n- 663 (void)deviceContext; // unused\n- 664\n- 665 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n- 666 dstBuffer->BindCudaBuffer(), dstDesc,\n- 667 duBuffer->BindCudaBuffer(), duDesc,\n- 668 dvBuffer->BindCudaBuffer(), dvDesc,\n- 669 duuBuffer->BindCudaBuffer(), duuDesc,\n- 670 duvBuffer->BindCudaBuffer(), duvDesc,\n- 671 dvvBuffer->BindCudaBuffer(), dvvDesc,\n- 672 numPatchCoords,\n- 673 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n- 674 (const PatchArray *)patchTable->GetPatchArrayBuffer(),\n- 675 (const int *)patchTable->GetPatchIndexBuffer(),\n- 676 (const PatchParam *)patchTable->GetPatchParamBuffer());\n- 677 }\n- 678\n-706 static bool EvalPatches(\n- 707 const float *src, BufferDescriptor const &srcDesc,\n- 708 float *dst, BufferDescriptor const &dstDesc,\n- 709 int numPatchCoords,\n- 710 const PatchCoord *patchCoords,\n- 711 const PatchArray *patchArrays,\n- 712 const int *patchIndices,\n- 713 const PatchParam *patchParams);\n- 714\n-752 static bool EvalPatches(\n- 753 const float *src, BufferDescriptor const &srcDesc,\n- 754 float *dst, BufferDescriptor const &dstDesc,\n- 755 float *du, BufferDescriptor const &duDesc,\n- 756 float *dv, BufferDescriptor const &dvDesc,\n- 757 int numPatchCoords,\n- 758 PatchCoord const *patchCoords,\n- 759 PatchArray const *patchArrays,\n- 760 const int *patchIndices,\n- 761 PatchParam const *patchParams);\n- 762\n-815 static bool EvalPatches(\n- 816 const float *src, BufferDescriptor const &srcDesc,\n- 817 float *dst, BufferDescriptor const &dstDesc,\n- 818 float *du, BufferDescriptor const &duDesc,\n- 819 float *dv, BufferDescriptor const &dvDesc,\n- 820 float *duu, BufferDescriptor const &duuDesc,\n- 821 float *duv, BufferDescriptor const &duvDesc,\n- 822 float *dvv, BufferDescriptor const &dvvDesc,\n- 823 int numPatchCoords,\n- 824 PatchCoord const *patchCoords,\n- 825 PatchArray const *patchArrays,\n- 826 const int *patchIndices,\n- 827 PatchParam const *patchParams);\n- 828\n- 857 template \n-859 static bool EvalPatchesVarying(\n- 860 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 861 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 862 int numPatchCoords,\n- 863 PATCHCOORD_BUFFER *patchCoords,\n- 864 PATCH_TABLE *patchTable,\n- 865 CudaEvaluator const *instance,\n- 866 void * deviceContext = NULL) {\n- 867\n- 868 (void)instance; // unused\n- 869 (void)deviceContext; // unused\n- 870\n- 871 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n- 872 dstBuffer->BindCudaBuffer(), dstDesc,\n- 873 numPatchCoords,\n- 874 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n- 875 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),\n- 876 (const int *)patchTable->GetVaryingPatchIndexBuffer(),\n- 877 (const PatchParam *)patchTable->GetPatchParamBuffer());\n- 878 }\n- 879\n- 920 template \n-922 static bool EvalPatchesVarying(\n- 923 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 924 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 925 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 926 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 927 int numPatchCoords,\n- 928 PATCHCOORD_BUFFER *patchCoords,\n- 929 PATCH_TABLE *patchTable,\n- 930 CudaEvaluator const *instance,\n- 931 void * deviceContext = NULL) {\n- 932\n- 933 (void)instance; // unused\n- 934 (void)deviceContext; // unused\n- 935\n- 936 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n- 937 dstBuffer->BindCudaBuffer(), dstDesc,\n- 938 duBuffer->BindCudaBuffer(), duDesc,\n- 939 dvBuffer->BindCudaBuffer(), dvDesc,\n- 940 numPatchCoords,\n- 941 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n- 942 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),\n- 943 (const int *)patchTable->GetVaryingPatchIndexBuffer(),\n- 944 (const PatchParam *)patchTable->GetPatchParamBuffer());\n- 945 }\n- 946\n- 1005 template \n-1007 static bool EvalPatchesVarying(\n- 1008 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1009 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1010 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1011 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1012 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1013 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1014 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1015 int numPatchCoords,\n- 1016 PATCHCOORD_BUFFER *patchCoords,\n- 1017 PATCH_TABLE *patchTable,\n- 1018 CudaEvaluator const *instance,\n- 1019 void * deviceContext = NULL) {\n- 1020\n- 1021 (void)instance; // unused\n- 1022 (void)deviceContext; // unused\n- 1023\n- 1024 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n- 1025 dstBuffer->BindCudaBuffer(), dstDesc,\n- 1026 duBuffer->BindCudaBuffer(), duDesc,\n- 1027 dvBuffer->BindCudaBuffer(), dvDesc,\n- 1028 duuBuffer->BindCudaBuffer(), duuDesc,\n- 1029 duvBuffer->BindCudaBuffer(), duvDesc,\n- 1030 dvvBuffer->BindCudaBuffer(), dvvDesc,\n- 1031 numPatchCoords,\n- 1032 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n- 1033 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),\n- 1034 (const int *)patchTable->GetVaryingPatchIndexBuffer(),\n- 1035 (const PatchParam *)patchTable->GetPatchParamBuffer());\n- 1036 }\n- 1037\n- 1068 template \n-1070 static bool EvalPatchesFaceVarying(\n- 1071 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1072 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1073 int numPatchCoords,\n- 1074 PATCHCOORD_BUFFER *patchCoords,\n- 1075 PATCH_TABLE *patchTable,\n- 1076 int fvarChannel,\n- 1077 CudaEvaluator const *instance,\n- 1078 void * deviceContext = NULL) {\n- 1079\n- 1080 (void)instance; // unused\n- 1081 (void)deviceContext; // unused\n- 1082\n- 1083 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n- 1084 dstBuffer->BindCudaBuffer(), dstDesc,\n- 1085 numPatchCoords,\n- 1086 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n- 1087 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 1088 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1089 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 1090 }\n- 1091\n- 1134 template \n-1136 static bool EvalPatchesFaceVarying(\n- 1137 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1138 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1139 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1140 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1141 int numPatchCoords,\n- 1142 PATCHCOORD_BUFFER *patchCoords,\n- 1143 PATCH_TABLE *patchTable,\n- 1144 int fvarChannel,\n- 1145 CudaEvaluator const *instance,\n- 1146 void * deviceContext = NULL) {\n- 1147\n- 1148 (void)instance; // unused\n- 1149 (void)deviceContext; // unused\n- 1150\n- 1151 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n- 1152 dstBuffer->BindCudaBuffer(), dstDesc,\n- 1153 duBuffer->BindCudaBuffer(), duDesc,\n- 1154 dvBuffer->BindCudaBuffer(), dvDesc,\n- 1155 numPatchCoords,\n- 1156 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n- 1157 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 1158 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1159 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 1160 }\n- 1161\n- 1222 template \n-1224 static bool EvalPatchesFaceVarying(\n- 1225 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1226 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1227 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1228 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1229 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1230 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1231 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1232 int numPatchCoords,\n- 1233 PATCHCOORD_BUFFER *patchCoords,\n- 1234 PATCH_TABLE *patchTable,\n- 1235 int fvarChannel,\n- 1236 CudaEvaluator const *instance,\n- 1237 void * deviceContext = NULL) {\n- 1238\n- 1239 (void)instance; // unused\n- 1240 (void)deviceContext; // unused\n- 1241\n- 1242 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n- 1243 dstBuffer->BindCudaBuffer(), dstDesc,\n- 1244 duBuffer->BindCudaBuffer(), duDesc,\n- 1245 dvBuffer->BindCudaBuffer(), dvDesc,\n- 1246 duuBuffer->BindCudaBuffer(), duuDesc,\n- 1247 duvBuffer->BindCudaBuffer(), duvDesc,\n- 1248 dvvBuffer->BindCudaBuffer(), dvvDesc,\n- 1249 numPatchCoords,\n- 1250 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n- 1251 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 1252 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1253 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 1254 }\n- 1255\n-1261 static void Synchronize(void *deviceContext = NULL);\n- 1262};\n- 1263\n- 1264\n- 1265} // end namespace Osd\n- 1266\n- 1267} // end namespace OPENSUBDIV_VERSION\n- 1268using namespace OPENSUBDIV_VERSION;\n- 1269\n- 1270} // end namespace OpenSubdiv\n- 1271\n- 1272\n- 1273#endif // OPENSUBDIV3_OSD_CUDA_EVALUATOR_H\n+66 void *GetPatchParamBuffer() const { return _patchParamBuffer; }\n+ 67\n+69 void *GetVaryingPatchArrayBuffer() const {\n+ 70 return _varyingPatchArrays;\n+ 71 }\n+73 void *GetVaryingPatchIndexBuffer() const {\n+ 74 return _varyingIndexBuffer;\n+ 75 }\n+ 76\n+78 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }\n+ 79\n+81 void *GetFVarPatchArrayBuffer(int fvarChannel) const {\n+ 82 return _fvarPatchArrays[fvarChannel];\n+ 83 }\n+ 84\n+86 void *GetFVarPatchIndexBuffer(int fvarChannel = 0) const {\n+ 87 return _fvarIndexBuffers[fvarChannel];\n+ 88 }\n+ 89\n+91 void *GetFVarPatchParamBuffer(int fvarChannel = 0) const {\n+ 92 return _fvarParamBuffers[fvarChannel];\n+ 93 }\n+ 94\n+ 95protected:\n+96 CudaPatchTable();\n+ 97\n+98 bool allocate(Far::PatchTable const *patchTable);\n+ 99\n+100 void *_patchArrays;\n+101 void *_indexBuffer;\n+102 void *_patchParamBuffer;\n+ 103\n+104 void *_varyingPatchArrays;\n+105 void *_varyingIndexBuffer;\n+ 106\n+107 std::vector _fvarPatchArrays;\n+108 std::vector _fvarIndexBuffers;\n+109 std::vector _fvarParamBuffers;\n+ 110};\n+ 111\n+ 112} // end namespace Osd\n+ 113\n+ 114} // end namespace OPENSUBDIV_VERSION\n+ 115using namespace OPENSUBDIV_VERSION;\n+ 116\n+ 117} // end namespace OpenSubdiv\n+ 118\n+ 119#endif // OPENSUBDIV3_OSD_CUDA_PATCH_TABLE_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n-Stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:273\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n-Limit stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:583\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n-BufferDescriptor is a struct which describes buffer elements in interleaved\n-data buffers....\n-Definition: bufferDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable\n-CUDA stencil table.\n-Definition: cudaEvaluator.h:52\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetDuvWeightsBuffer\n-void * GetDuvWeightsBuffer() const\n-Definition: cudaEvaluator.h:77\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::~CudaStencilTable\n-~CudaStencilTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetDuWeightsBuffer\n-void * GetDuWeightsBuffer() const\n-Definition: cudaEvaluator.h:74\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetDvvWeightsBuffer\n-void * GetDvvWeightsBuffer() const\n-Definition: cudaEvaluator.h:78\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetOffsetsBuffer\n-void * GetOffsetsBuffer() const\n-Definition: cudaEvaluator.h:71\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::CudaStencilTable\n-CudaStencilTable(Far::LimitStencilTable const *limitStencilTable)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetDvWeightsBuffer\n-void * GetDvWeightsBuffer() const\n-Definition: cudaEvaluator.h:75\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetNumStencils\n-int GetNumStencils() const\n-Definition: cudaEvaluator.h:79\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetDuuWeightsBuffer\n-void * GetDuuWeightsBuffer() const\n-Definition: cudaEvaluator.h:76\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetWeightsBuffer\n-void * GetWeightsBuffer() const\n-Definition: cudaEvaluator.h:73\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::Create\n-static CudaStencilTable * Create(Far::LimitStencilTable const\n-*limitStencilTable, void *deviceContext=NULL)\n-Definition: cudaEvaluator.h:59\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetSizesBuffer\n-void * GetSizesBuffer() const\n-Definition: cudaEvaluator.h:70\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetIndicesBuffer\n-void * GetIndicesBuffer() const\n-Definition: cudaEvaluator.h:72\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::CudaStencilTable\n-CudaStencilTable(Far::StencilTable const *stencilTable)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::Create\n-static CudaStencilTable * Create(Far::StencilTable const *stencilTable, void\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n+Container for arrays of parametric patches.\n+Definition: patchTable.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable\n+CUDA patch table.\n+Definition: cudaPatchTable.h:51\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchArrayBuffer\n+void * GetFVarPatchArrayBuffer(int fvarChannel) const\n+Returns the cuda memory of the array of Osd::PatchArray buffer.\n+Definition: cudaPatchTable.h:81\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_fvarParamBuffers\n+std::vector< void * > _fvarParamBuffers\n+Definition: cudaPatchTable.h:109\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::Create\n+static CudaPatchTable * Create(Far::PatchTable const *patchTable, void\n *deviceContext=NULL)\n-Definition: cudaEvaluator.h:54\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator\n-Definition: cudaEvaluator.h:94\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n-static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const\n-PatchCoord *patchCoords, const PatchArray *patchArrays, const int\n-*patchIndices, const PatchParam *patchParams)\n-Static limit eval function. It takes an array of PatchCoord and evaluate limit\n-values on given PatchT...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void\n-*deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cudaEvaluator.h:922\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void\n-*deviceContext=NULL)\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: cudaEvaluator.h:648\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n-const CudaEvaluator *instance=NULL, void *deviceContext=NULL)\n-Generic static eval stencils function with derivatives. This function has a\n-same signature as other d...\n-Definition: cudaEvaluator.h:353\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n-static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int\n-*offsets, const int *indices, const float *weights, int start, int end)\n-Static eval stencils function which takes raw cuda buffers for input and\n-output.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n-CudaEvaluator const *instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cudaEvaluator.h:859\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const\n-CudaEvaluator *instance=NULL, void *deviceContext=NULL)\n-Generic static eval stencils function with derivatives. This function has a\n-same signature as other d...\n-Definition: cudaEvaluator.h:218\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n-static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n-BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n-float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord\n-const *patchCoords, PatchArray const *patchArrays, const int *patchIndices,\n-PatchParam const *patchParams)\n-Static limit eval function. It takes an array of PatchCoord and evaluate limit\n-values on given PatchT...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CudaEvaluator const\n-*instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cudaEvaluator.h:1136\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n-const *stencilTable, const void *instance=NULL, void *deviceContext=NULL)\n-Generic static compute function. This function has a same signature as other\n-device kernels have so t...\n-Definition: cudaEvaluator.h:126\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n-static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int\n-*offsets, const int *indices, const float *weights, const float *duWeights,\n-const float *dvWeights, int start, int end)\n-Static eval stencils function with derivatives, which takes raw cuda pointers\n-for input and output.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n-static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n-BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n-float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int\n-*offsets, const int *indices, const float *weights, const float *duWeights,\n-const float *dvWeights, const float *duuWeights, const float *duvWeights, const\n-float *dvvWeights, int start, int end)\n-Static eval stencils function with derivatives, which takes raw cuda pointers\n-for input and output.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::Synchronize\n-static void Synchronize(void *deviceContext=NULL)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n-static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords,\n-PatchCoord const *patchCoords, PatchArray const *patchArrays, const int\n-*patchIndices, PatchParam const *patchParams)\n-Static limit eval function. It takes an array of PatchCoord and evaluate limit\n-values on given PatchT...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const\n-*instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cudaEvaluator.h:1007\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const\n-*instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cudaEvaluator.h:504\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n-CudaEvaluator const *instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cudaEvaluator.h:1224\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: cudaEvaluator.h:565\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n-fvarChannel, CudaEvaluator const *instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cudaEvaluator.h:1070\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord\n-Coordinates set on a patch table.\n-Definition: types.h:42\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArray\n-Definition: types.h:66\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParam\n-Definition: types.h:127\n+Creator. Returns NULL if error.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_varyingPatchArrays\n+void * _varyingPatchArrays\n+Definition: cudaPatchTable.h:104\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetVaryingPatchIndexBuffer\n+void * GetVaryingPatchIndexBuffer() const\n+Returns the cuda memory of the array of varying control vertices.\n+Definition: cudaPatchTable.h:73\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_fvarIndexBuffers\n+std::vector< void * > _fvarIndexBuffers\n+Definition: cudaPatchTable.h:108\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetVaryingPatchArrayBuffer\n+void * GetVaryingPatchArrayBuffer() const\n+Returns the cuda memory of the array of Osd::PatchArray buffer.\n+Definition: cudaPatchTable.h:69\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_varyingIndexBuffer\n+void * _varyingIndexBuffer\n+Definition: cudaPatchTable.h:105\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::~CudaPatchTable\n+~CudaPatchTable()\n+Destructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchArrayBuffer\n+void * GetPatchArrayBuffer() const\n+Returns the cuda memory of the array of Osd::PatchArray buffer.\n+Definition: cudaPatchTable.h:60\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_fvarPatchArrays\n+std::vector< void * > _fvarPatchArrays\n+Definition: cudaPatchTable.h:107\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::allocate\n+bool allocate(Far::PatchTable const *patchTable)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchIndexBuffer\n+void * GetPatchIndexBuffer() const\n+Returns the cuda memory of the patch control vertices.\n+Definition: cudaPatchTable.h:63\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetNumFVarChannels\n+int GetNumFVarChannels() const\n+Returns the number of face-varying channels buffers.\n+Definition: cudaPatchTable.h:78\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_patchArrays\n+void * _patchArrays\n+Definition: cudaPatchTable.h:100\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetPatchParamBuffer\n+void * GetPatchParamBuffer() const\n+Returns the cuda memory of the array of Osd::PatchParam buffer.\n+Definition: cudaPatchTable.h:66\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchIndexBuffer\n+void * GetFVarPatchIndexBuffer(int fvarChannel=0) const\n+Returns the cuda memory of the array of face-varying control vertices.\n+Definition: cudaPatchTable.h:86\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::GetFVarPatchParamBuffer\n+void * GetFVarPatchParamBuffer(int fvarChannel=0) const\n+Returns the cuda memory of the array of face-varying param.\n+Definition: cudaPatchTable.h:91\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::CudaPatchTable\n+CudaPatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_patchParamBuffer\n+void * _patchParamBuffer\n+Definition: cudaPatchTable.h:102\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::_indexBuffer\n+void * _indexBuffer\n+Definition: cudaPatchTable.h:101\n * opensubdiv\n * osd\n- * cudaEvaluator.h\n+ * cudaPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00794.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00794.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/opencl.h File Reference\n+OpenSubdiv: opensubdiv/osd/cudaEvaluator.h File Reference\n \n \n \n \n \n \n \n@@ -80,24 +80,50 @@\n
\n \n
\n \n
\n-
opencl.h File Reference
\n+
\n+Classes |\n+Namespaces
\n+
cudaEvaluator.h File Reference
\n
\n
\n-
#include <CL/opencl.h>
\n+
#include "../version.h"
\n+#include <vector>
\n+#include "../osd/bufferDescriptor.h"
\n+#include "../osd/types.h"
\n
\n

Go to the source code of this file.

\n+\n+\n+\n+\n+\n+\n+\n+

\n+Classes

class  CudaStencilTable
 CUDA stencil table. More...
 
class  CudaEvaluator
 
\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+

\n+Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n
\n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,14 +4,33 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-opencl.h File Reference\n-#include \n+Classes | Namespaces\n+cudaEvaluator.h File Reference\n+#include \"../version.h\"\n+#include \n+#include \"../osd/bufferDescriptor.h\"\n+#include \"../osd/types.h\"\n Go_to_the_source_code_of_this_file.\n+ Classes\n+class \u00a0CudaStencilTable\n+\u00a0 CUDA stencil table. More...\n+\u00a0\n+class \u00a0CudaEvaluator\n+\u00a0\n+ Namespaces\n+namespace \u00a0OpenSubdiv\n+\u00a0\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n+\u00a0\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n+\u00a0\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n+\u00a0\n * opensubdiv\n * osd\n- * opencl.h\n+ * cudaEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00794_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00794_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/opencl.h Source File\n+OpenSubdiv: opensubdiv/osd/cudaEvaluator.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
opencl.h
\n+
cudaEvaluator.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2014 Pixar
\n+
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,32 +107,550 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_OPENCL_H
\n-
26#define OPENSUBDIV3_OSD_OPENCL_H
\n+
25#ifndef OPENSUBDIV3_OSD_CUDA_EVALUATOR_H
\n+
26#define OPENSUBDIV3_OSD_CUDA_EVALUATOR_H
\n
27
\n-
28#if defined(OPENSUBDIV_HAS_CLEW)
\n-
29# include <clew.h>
\n-
30#else
\n-
31# if defined(__APPLE__)
\n-
32# include <OpenCL/opencl.h>
\n-
33# else
\n-
34# include <CL/opencl.h>
\n-
35# endif
\n-
36#endif
\n-
37
\n-
38#endif // OPENSUBDIV3_OSD_CL_UTIL_H
\n+
28#include "../version.h"
\n+
29
\n+
30#include <vector>
\n+
31#include "../osd/bufferDescriptor.h"
\n+
32#include "../osd/types.h"
\n+
33
\n+
34namespace OpenSubdiv {
\n+
35namespace OPENSUBDIV_VERSION {
\n+
36
\n+
37namespace Far {
\n+
38 class PatchTable;
\n+
39 class StencilTable;
\n+
40 class LimitStencilTable;
\n+
41}
\n+
42
\n+
43namespace Osd {
\n+
44
\n+
52class CudaStencilTable {
\n+
53public:
\n+
54 static CudaStencilTable *Create(Far::StencilTable const *stencilTable,
\n+
55 void *deviceContext = NULL) {
\n+
56 (void)deviceContext; // unused
\n+
57 return new CudaStencilTable(stencilTable);
\n+
58 }
\n+
59 static CudaStencilTable *Create(Far::LimitStencilTable const *limitStencilTable,
\n+
60 void *deviceContext = NULL) {
\n+
61 (void)deviceContext; // unused
\n+
62 return new CudaStencilTable(limitStencilTable);
\n+
63 }
\n+
64
\n+
65 explicit CudaStencilTable(Far::StencilTable const *stencilTable);
\n+
66 explicit CudaStencilTable(Far::LimitStencilTable const *limitStencilTable);
\n+
67 ~CudaStencilTable();
\n+
68
\n+
69 // interfaces needed for CudaCompute
\n+
70 void *GetSizesBuffer() const { return _sizes; }
\n+
71 void *GetOffsetsBuffer() const { return _offsets; }
\n+
72 void *GetIndicesBuffer() const { return _indices; }
\n+
73 void *GetWeightsBuffer() const { return _weights; }
\n+
74 void *GetDuWeightsBuffer() const { return _duWeights; }
\n+
75 void *GetDvWeightsBuffer() const { return _dvWeights; }
\n+
76 void *GetDuuWeightsBuffer() const { return _duuWeights; }
\n+
77 void *GetDuvWeightsBuffer() const { return _duvWeights; }
\n+
78 void *GetDvvWeightsBuffer() const { return _dvvWeights; }
\n+
79 int GetNumStencils() const { return _numStencils; }
\n+
80
\n+
81private:
\n+
82 void * _sizes,
\n+
83 * _offsets,
\n+
84 * _indices,
\n+
85 * _weights,
\n+
86 * _duWeights,
\n+
87 * _dvWeights,
\n+
88 * _duuWeights,
\n+
89 * _duvWeights,
\n+
90 * _dvvWeights;
\n+
91 int _numStencils;
\n+
92};
\n+
93
\n+
94class CudaEvaluator {
\n+
95public:
\n+
101
\n+
125 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
126 static bool EvalStencils(
\n+
127 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
128 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
129 STENCIL_TABLE const *stencilTable,
\n+
130 const void *instance = NULL,
\n+
131 void * deviceContext = NULL) {
\n+
132
\n+
133 (void)instance; // unused
\n+
134 (void)deviceContext; // unused
\n+
135 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
136 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
137 (int const *)stencilTable->GetSizesBuffer(),
\n+
138 (int const *)stencilTable->GetOffsetsBuffer(),
\n+
139 (int const *)stencilTable->GetIndicesBuffer(),
\n+
140 (float const *)stencilTable->GetWeightsBuffer(),
\n+
141 /*start = */ 0,
\n+
142 /*end = */ stencilTable->GetNumStencils());
\n+
143 }
\n+
144
\n+
171 static bool EvalStencils(
\n+
172 const float *src, BufferDescriptor const &srcDesc,
\n+
173 float *dst, BufferDescriptor const &dstDesc,
\n+
174 const int * sizes,
\n+
175 const int * offsets,
\n+
176 const int * indices,
\n+
177 const float * weights,
\n+
178 int start, int end);
\n+
179
\n+
217 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
218 static bool EvalStencils(
\n+
219 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
220 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
221 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
222 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
223 STENCIL_TABLE const *stencilTable,
\n+
224 const CudaEvaluator *instance = NULL,
\n+
225 void * deviceContext = NULL) {
\n+
226
\n+
227 (void)instance; // unused
\n+
228 (void)deviceContext; // unused
\n+
229
\n+
230 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
231 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
232 duBuffer->BindCudaBuffer(), duDesc,
\n+
233 dvBuffer->BindCudaBuffer(), dvDesc,
\n+
234 (int const *)stencilTable->GetSizesBuffer(),
\n+
235 (int const *)stencilTable->GetOffsetsBuffer(),
\n+
236 (int const *)stencilTable->GetIndicesBuffer(),
\n+
237 (float const *)stencilTable->GetWeightsBuffer(),
\n+
238 (float const *)stencilTable->GetDuWeightsBuffer(),
\n+
239 (float const *)stencilTable->GetDvWeightsBuffer(),
\n+
240 /*start = */ 0,
\n+
241 /*end = */ stencilTable->GetNumStencils());
\n+
242 }
\n+
243
\n+
284 static bool EvalStencils(
\n+
285 const float *src, BufferDescriptor const &srcDesc,
\n+
286 float *dst, BufferDescriptor const &dstDesc,
\n+
287 float *du, BufferDescriptor const &duDesc,
\n+
288 float *dv, BufferDescriptor const &dvDesc,
\n+
289 const int * sizes,
\n+
290 const int * offsets,
\n+
291 const int * indices,
\n+
292 const float * weights,
\n+
293 const float * duWeights,
\n+
294 const float * dvWeights,
\n+
295 int start, int end);
\n+
296
\n+
352 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
353 static bool EvalStencils(
\n+
354 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
355 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
356 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
357 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
358 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
359 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
360 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
361 STENCIL_TABLE const *stencilTable,
\n+
362 const CudaEvaluator *instance = NULL,
\n+
363 void * deviceContext = NULL) {
\n+
364
\n+
365 (void)instance; // unused
\n+
366 (void)deviceContext; // unused
\n+
367
\n+
368 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
369 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
370 duBuffer->BindCudaBuffer(), duDesc,
\n+
371 dvBuffer->BindCudaBuffer(), dvDesc,
\n+
372 duuBuffer->BindCudaBuffer(), duuDesc,
\n+
373 duvBuffer->BindCudaBuffer(), duvDesc,
\n+
374 dvvBuffer->BindCudaBuffer(), dvvDesc,
\n+
375 (int const *)stencilTable->GetSizesBuffer(),
\n+
376 (int const *)stencilTable->GetOffsetsBuffer(),
\n+
377 (int const *)stencilTable->GetIndicesBuffer(),
\n+
378 (float const *)stencilTable->GetWeightsBuffer(),
\n+
379 (float const *)stencilTable->GetDuWeightsBuffer(),
\n+
380 (float const *)stencilTable->GetDvWeightsBuffer(),
\n+
381 (float const *)stencilTable->GetDuuWeightsBuffer(),
\n+
382 (float const *)stencilTable->GetDuvWeightsBuffer(),
\n+
383 (float const *)stencilTable->GetDvvWeightsBuffer(),
\n+
384 /*start = */ 0,
\n+
385 /*end = */ stencilTable->GetNumStencils());
\n+
386 }
\n+
387
\n+
449 static bool EvalStencils(
\n+
450 const float *src, BufferDescriptor const &srcDesc,
\n+
451 float *dst, BufferDescriptor const &dstDesc,
\n+
452 float *du, BufferDescriptor const &duDesc,
\n+
453 float *dv, BufferDescriptor const &dvDesc,
\n+
454 float *duu, BufferDescriptor const &duuDesc,
\n+
455 float *duv, BufferDescriptor const &duvDesc,
\n+
456 float *dvv, BufferDescriptor const &dvvDesc,
\n+
457 const int * sizes,
\n+
458 const int * offsets,
\n+
459 const int * indices,
\n+
460 const float * weights,
\n+
461 const float * duWeights,
\n+
462 const float * dvWeights,
\n+
463 const float * duuWeights,
\n+
464 const float * duvWeights,
\n+
465 const float * dvvWeights,
\n+
466 int start, int end);
\n+
467
\n+
473
\n+
502 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
503 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
504 static bool EvalPatches(
\n+
505 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
506 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
507 int numPatchCoords,
\n+
508 PATCHCOORD_BUFFER *patchCoords,
\n+
509 PATCH_TABLE *patchTable,
\n+
510 CudaEvaluator const *instance,
\n+
511 void * deviceContext = NULL) {
\n+
512
\n+
513 (void)instance; // unused
\n+
514 (void)deviceContext; // unused
\n+
515
\n+
516 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
517 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
518 numPatchCoords,
\n+
519 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n+
520 (const PatchArray *)patchTable->GetPatchArrayBuffer(),
\n+
521 (const int *)patchTable->GetPatchIndexBuffer(),
\n+
522 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n+
523 }
\n+
524
\n+
563 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
564 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
565 static bool EvalPatches(
\n+
566 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
567 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
568 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
569 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
570 int numPatchCoords,
\n+
571 PATCHCOORD_BUFFER *patchCoords,
\n+
572 PATCH_TABLE *patchTable,
\n+
573 CudaEvaluator const *instance,
\n+
574 void * deviceContext = NULL) {
\n+
575
\n+
576 (void)instance; // unused
\n+
577 (void)deviceContext; // unused
\n+
578
\n+
579 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
580 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
581 duBuffer->BindCudaBuffer(), duDesc,
\n+
582 dvBuffer->BindCudaBuffer(), dvDesc,
\n+
583 numPatchCoords,
\n+
584 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n+
585 (const PatchArray *)patchTable->GetPatchArrayBuffer(),
\n+
586 (const int *)patchTable->GetPatchIndexBuffer(),
\n+
587 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n+
588 }
\n+
589
\n+
646 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
647 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
648 static bool EvalPatches(
\n+
649 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
650 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
651 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
652 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
653 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
654 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
655 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
656 int numPatchCoords,
\n+
657 PATCHCOORD_BUFFER *patchCoords,
\n+
658 PATCH_TABLE *patchTable,
\n+
659 CudaEvaluator const *instance,
\n+
660 void * deviceContext = NULL) {
\n+
661
\n+
662 (void)instance; // unused
\n+
663 (void)deviceContext; // unused
\n+
664
\n+
665 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
666 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
667 duBuffer->BindCudaBuffer(), duDesc,
\n+
668 dvBuffer->BindCudaBuffer(), dvDesc,
\n+
669 duuBuffer->BindCudaBuffer(), duuDesc,
\n+
670 duvBuffer->BindCudaBuffer(), duvDesc,
\n+
671 dvvBuffer->BindCudaBuffer(), dvvDesc,
\n+
672 numPatchCoords,
\n+
673 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n+
674 (const PatchArray *)patchTable->GetPatchArrayBuffer(),
\n+
675 (const int *)patchTable->GetPatchIndexBuffer(),
\n+
676 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n+
677 }
\n+
678
\n+
706 static bool EvalPatches(
\n+
707 const float *src, BufferDescriptor const &srcDesc,
\n+
708 float *dst, BufferDescriptor const &dstDesc,
\n+
709 int numPatchCoords,
\n+
710 const PatchCoord *patchCoords,
\n+
711 const PatchArray *patchArrays,
\n+
712 const int *patchIndices,
\n+
713 const PatchParam *patchParams);
\n+
714
\n+
752 static bool EvalPatches(
\n+
753 const float *src, BufferDescriptor const &srcDesc,
\n+
754 float *dst, BufferDescriptor const &dstDesc,
\n+
755 float *du, BufferDescriptor const &duDesc,
\n+
756 float *dv, BufferDescriptor const &dvDesc,
\n+
757 int numPatchCoords,
\n+
758 PatchCoord const *patchCoords,
\n+
759 PatchArray const *patchArrays,
\n+
760 const int *patchIndices,
\n+
761 PatchParam const *patchParams);
\n+
762
\n+
815 static bool EvalPatches(
\n+
816 const float *src, BufferDescriptor const &srcDesc,
\n+
817 float *dst, BufferDescriptor const &dstDesc,
\n+
818 float *du, BufferDescriptor const &duDesc,
\n+
819 float *dv, BufferDescriptor const &dvDesc,
\n+
820 float *duu, BufferDescriptor const &duuDesc,
\n+
821 float *duv, BufferDescriptor const &duvDesc,
\n+
822 float *dvv, BufferDescriptor const &dvvDesc,
\n+
823 int numPatchCoords,
\n+
824 PatchCoord const *patchCoords,
\n+
825 PatchArray const *patchArrays,
\n+
826 const int *patchIndices,
\n+
827 PatchParam const *patchParams);
\n+
828
\n+
857 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
858 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
859 static bool EvalPatchesVarying(
\n+
860 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
861 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
862 int numPatchCoords,
\n+
863 PATCHCOORD_BUFFER *patchCoords,
\n+
864 PATCH_TABLE *patchTable,
\n+
865 CudaEvaluator const *instance,
\n+
866 void * deviceContext = NULL) {
\n+
867
\n+
868 (void)instance; // unused
\n+
869 (void)deviceContext; // unused
\n+
870
\n+
871 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
872 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
873 numPatchCoords,
\n+
874 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n+
875 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),
\n+
876 (const int *)patchTable->GetVaryingPatchIndexBuffer(),
\n+
877 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n+
878 }
\n+
879
\n+
920 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
921 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
922 static bool EvalPatchesVarying(
\n+
923 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
924 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
925 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
926 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
927 int numPatchCoords,
\n+
928 PATCHCOORD_BUFFER *patchCoords,
\n+
929 PATCH_TABLE *patchTable,
\n+
930 CudaEvaluator const *instance,
\n+
931 void * deviceContext = NULL) {
\n+
932
\n+
933 (void)instance; // unused
\n+
934 (void)deviceContext; // unused
\n+
935
\n+
936 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
937 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
938 duBuffer->BindCudaBuffer(), duDesc,
\n+
939 dvBuffer->BindCudaBuffer(), dvDesc,
\n+
940 numPatchCoords,
\n+
941 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n+
942 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),
\n+
943 (const int *)patchTable->GetVaryingPatchIndexBuffer(),
\n+
944 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n+
945 }
\n+
946
\n+
1005 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1006 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1007 static bool EvalPatchesVarying(
\n+
1008 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1009 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1010 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1011 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1012 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1013 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1014 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1015 int numPatchCoords,
\n+
1016 PATCHCOORD_BUFFER *patchCoords,
\n+
1017 PATCH_TABLE *patchTable,
\n+
1018 CudaEvaluator const *instance,
\n+
1019 void * deviceContext = NULL) {
\n+
1020
\n+
1021 (void)instance; // unused
\n+
1022 (void)deviceContext; // unused
\n+
1023
\n+
1024 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
1025 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
1026 duBuffer->BindCudaBuffer(), duDesc,
\n+
1027 dvBuffer->BindCudaBuffer(), dvDesc,
\n+
1028 duuBuffer->BindCudaBuffer(), duuDesc,
\n+
1029 duvBuffer->BindCudaBuffer(), duvDesc,
\n+
1030 dvvBuffer->BindCudaBuffer(), dvvDesc,
\n+
1031 numPatchCoords,
\n+
1032 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n+
1033 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),
\n+
1034 (const int *)patchTable->GetVaryingPatchIndexBuffer(),
\n+
1035 (const PatchParam *)patchTable->GetPatchParamBuffer());
\n+
1036 }
\n+
1037
\n+
1068 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1069 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1070 static bool EvalPatchesFaceVarying(
\n+
1071 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1072 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1073 int numPatchCoords,
\n+
1074 PATCHCOORD_BUFFER *patchCoords,
\n+
1075 PATCH_TABLE *patchTable,
\n+
1076 int fvarChannel,
\n+
1077 CudaEvaluator const *instance,
\n+
1078 void * deviceContext = NULL) {
\n+
1079
\n+
1080 (void)instance; // unused
\n+
1081 (void)deviceContext; // unused
\n+
1082
\n+
1083 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
1084 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
1085 numPatchCoords,
\n+
1086 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n+
1087 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
1088 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1089 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
1090 }
\n+
1091
\n+
1134 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1135 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1136 static bool EvalPatchesFaceVarying(
\n+
1137 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1138 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1139 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1140 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1141 int numPatchCoords,
\n+
1142 PATCHCOORD_BUFFER *patchCoords,
\n+
1143 PATCH_TABLE *patchTable,
\n+
1144 int fvarChannel,
\n+
1145 CudaEvaluator const *instance,
\n+
1146 void * deviceContext = NULL) {
\n+
1147
\n+
1148 (void)instance; // unused
\n+
1149 (void)deviceContext; // unused
\n+
1150
\n+
1151 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
1152 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
1153 duBuffer->BindCudaBuffer(), duDesc,
\n+
1154 dvBuffer->BindCudaBuffer(), dvDesc,
\n+
1155 numPatchCoords,
\n+
1156 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n+
1157 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
1158 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1159 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
1160 }
\n+
1161
\n+
1222 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1223 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1224 static bool EvalPatchesFaceVarying(
\n+
1225 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1226 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1227 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1228 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1229 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1230 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1231 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1232 int numPatchCoords,
\n+
1233 PATCHCOORD_BUFFER *patchCoords,
\n+
1234 PATCH_TABLE *patchTable,
\n+
1235 int fvarChannel,
\n+
1236 CudaEvaluator const *instance,
\n+
1237 void * deviceContext = NULL) {
\n+
1238
\n+
1239 (void)instance; // unused
\n+
1240 (void)deviceContext; // unused
\n+
1241
\n+
1242 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,
\n+
1243 dstBuffer->BindCudaBuffer(), dstDesc,
\n+
1244 duBuffer->BindCudaBuffer(), duDesc,
\n+
1245 dvBuffer->BindCudaBuffer(), dvDesc,
\n+
1246 duuBuffer->BindCudaBuffer(), duuDesc,
\n+
1247 duvBuffer->BindCudaBuffer(), duvDesc,
\n+
1248 dvvBuffer->BindCudaBuffer(), dvvDesc,
\n+
1249 numPatchCoords,
\n+
1250 (const PatchCoord *)patchCoords->BindCudaBuffer(),
\n+
1251 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
1252 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1253 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
1254 }
\n+
1255
\n+
1261 static void Synchronize(void *deviceContext = NULL);
\n+
1262};
\n+
1263
\n+
1264
\n+
1265} // end namespace Osd
\n+
1266
\n+
1267} // end namespace OPENSUBDIV_VERSION
\n+
1268using namespace OPENSUBDIV_VERSION;
\n+
1269
\n+
1270} // end namespace OpenSubdiv
\n+
1271
\n+
1272
\n+
1273#endif // OPENSUBDIV3_OSD_CUDA_EVALUATOR_H
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::CudaStencilTable
CudaStencilTable(Far::LimitStencilTable const *limitStencilTable)
\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::Create
static CudaStencilTable * Create(Far::LimitStencilTable const *limitStencilTable, void *deviceContext=NULL)
Definition: cudaEvaluator.h:59
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::CudaStencilTable
CudaStencilTable(Far::StencilTable const *stencilTable)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::Create
static CudaStencilTable * Create(Far::StencilTable const *stencilTable, void *deviceContext=NULL)
Definition: cudaEvaluator.h:54
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrays, const int *patchIndices, const PatchParam *patchParams)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:922
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: cudaEvaluator.h:648
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, const CudaEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: cudaEvaluator.h:353
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, int start, int end)
Static eval stencils function which takes raw cuda buffers for input and output.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:859
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const CudaEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: cudaEvaluator.h:218
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndices, PatchParam const *patchParams)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:1136
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, const void *instance=NULL, void *deviceContext=NULL)
Generic static compute function. This function has a same signature as other device kernels have so t...
Definition: cudaEvaluator.h:126
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw cuda pointers for input and output.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, const float *duuWeights, const float *duvWeights, const float *dvvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw cuda pointers for input and output.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::Synchronize
static void Synchronize(void *deviceContext=NULL)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndices, PatchParam const *patchParams)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:1007
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:504
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:1224
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: cudaEvaluator.h:565
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CudaEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cudaEvaluator.h:1070
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord
Coordinates set on a patch table.
Definition: types.h:42
\n+\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-opencl.h\n+cudaEvaluator.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2014 Pixar\n+ 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,25 +30,740 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_OPENCL_H\n- 26#define OPENSUBDIV3_OSD_OPENCL_H\n+ 25#ifndef OPENSUBDIV3_OSD_CUDA_EVALUATOR_H\n+ 26#define OPENSUBDIV3_OSD_CUDA_EVALUATOR_H\n 27\n- 28#if defined(OPENSUBDIV_HAS_CLEW)\n- 29# include \n- 30#else\n- 31# if defined(__APPLE__)\n- 32# include \n- 33# else\n- 34# include \n- 35# endif\n- 36#endif\n- 37\n- 38#endif // OPENSUBDIV3_OSD_CL_UTIL_H\n+ 28#include \"../version.h\"\n+ 29\n+ 30#include \n+ 31#include \"../osd/bufferDescriptor.h\"\n+ 32#include \"../osd/types.h\"\n+ 33\n+ 34namespace OpenSubdiv {\n+ 35namespace OPENSUBDIV_VERSION {\n+ 36\n+ 37namespace Far {\n+ 38 class PatchTable;\n+ 39 class StencilTable;\n+ 40 class LimitStencilTable;\n+ 41}\n+ 42\n+ 43namespace Osd {\n+ 44\n+52class CudaStencilTable {\n+ 53public:\n+54 static CudaStencilTable *Create(Far::StencilTable const *stencilTable,\n+ 55 void *deviceContext = NULL) {\n+ 56 (void)deviceContext; // unused\n+ 57 return new CudaStencilTable(stencilTable);\n+ 58 }\n+59 static CudaStencilTable *Create(Far::LimitStencilTable const\n+*limitStencilTable,\n+ 60 void *deviceContext = NULL) {\n+ 61 (void)deviceContext; // unused\n+ 62 return new CudaStencilTable(limitStencilTable);\n+ 63 }\n+ 64\n+65 explicit CudaStencilTable(Far::StencilTable const *stencilTable);\n+66 explicit CudaStencilTable(Far::LimitStencilTable const *limitStencilTable);\n+67 ~CudaStencilTable();\n+ 68\n+ 69 // interfaces needed for CudaCompute\n+70 void *GetSizesBuffer() const { return _sizes; }\n+71 void *GetOffsetsBuffer() const { return _offsets; }\n+72 void *GetIndicesBuffer() const { return _indices; }\n+73 void *GetWeightsBuffer() const { return _weights; }\n+74 void *GetDuWeightsBuffer() const { return _duWeights; }\n+75 void *GetDvWeightsBuffer() const { return _dvWeights; }\n+76 void *GetDuuWeightsBuffer() const { return _duuWeights; }\n+77 void *GetDuvWeightsBuffer() const { return _duvWeights; }\n+78 void *GetDvvWeightsBuffer() const { return _dvvWeights; }\n+79 int GetNumStencils() const { return _numStencils; }\n+ 80\n+ 81private:\n+ 82 void * _sizes,\n+ 83 * _offsets,\n+ 84 * _indices,\n+ 85 * _weights,\n+ 86 * _duWeights,\n+ 87 * _dvWeights,\n+ 88 * _duuWeights,\n+ 89 * _duvWeights,\n+ 90 * _dvvWeights;\n+ 91 int _numStencils;\n+ 92};\n+ 93\n+94class CudaEvaluator {\n+ 95public:\n+ 101\n+ 125 template \n+126 static bool EvalStencils(\n+ 127 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 128 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 129 STENCIL_TABLE const *stencilTable,\n+ 130 const void *instance = NULL,\n+ 131 void * deviceContext = NULL) {\n+ 132\n+ 133 (void)instance; // unused\n+ 134 (void)deviceContext; // unused\n+ 135 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 136 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 137 (int const *)stencilTable->GetSizesBuffer(),\n+ 138 (int const *)stencilTable->GetOffsetsBuffer(),\n+ 139 (int const *)stencilTable->GetIndicesBuffer(),\n+ 140 (float const *)stencilTable->GetWeightsBuffer(),\n+ 141 /*start = */ 0,\n+ 142 /*end = */ stencilTable->GetNumStencils());\n+ 143 }\n+ 144\n+171 static bool EvalStencils(\n+ 172 const float *src, BufferDescriptor const &srcDesc,\n+ 173 float *dst, BufferDescriptor const &dstDesc,\n+ 174 const int * sizes,\n+ 175 const int * offsets,\n+ 176 const int * indices,\n+ 177 const float * weights,\n+ 178 int start, int end);\n+ 179\n+ 217 template \n+218 static bool EvalStencils(\n+ 219 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 220 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 221 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 222 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 223 STENCIL_TABLE const *stencilTable,\n+ 224 const CudaEvaluator *instance = NULL,\n+ 225 void * deviceContext = NULL) {\n+ 226\n+ 227 (void)instance; // unused\n+ 228 (void)deviceContext; // unused\n+ 229\n+ 230 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 231 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 232 duBuffer->BindCudaBuffer(), duDesc,\n+ 233 dvBuffer->BindCudaBuffer(), dvDesc,\n+ 234 (int const *)stencilTable->GetSizesBuffer(),\n+ 235 (int const *)stencilTable->GetOffsetsBuffer(),\n+ 236 (int const *)stencilTable->GetIndicesBuffer(),\n+ 237 (float const *)stencilTable->GetWeightsBuffer(),\n+ 238 (float const *)stencilTable->GetDuWeightsBuffer(),\n+ 239 (float const *)stencilTable->GetDvWeightsBuffer(),\n+ 240 /*start = */ 0,\n+ 241 /*end = */ stencilTable->GetNumStencils());\n+ 242 }\n+ 243\n+284 static bool EvalStencils(\n+ 285 const float *src, BufferDescriptor const &srcDesc,\n+ 286 float *dst, BufferDescriptor const &dstDesc,\n+ 287 float *du, BufferDescriptor const &duDesc,\n+ 288 float *dv, BufferDescriptor const &dvDesc,\n+ 289 const int * sizes,\n+ 290 const int * offsets,\n+ 291 const int * indices,\n+ 292 const float * weights,\n+ 293 const float * duWeights,\n+ 294 const float * dvWeights,\n+ 295 int start, int end);\n+ 296\n+ 352 template \n+353 static bool EvalStencils(\n+ 354 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 355 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 356 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 357 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 358 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 359 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 360 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 361 STENCIL_TABLE const *stencilTable,\n+ 362 const CudaEvaluator *instance = NULL,\n+ 363 void * deviceContext = NULL) {\n+ 364\n+ 365 (void)instance; // unused\n+ 366 (void)deviceContext; // unused\n+ 367\n+ 368 return EvalStencils(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 369 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 370 duBuffer->BindCudaBuffer(), duDesc,\n+ 371 dvBuffer->BindCudaBuffer(), dvDesc,\n+ 372 duuBuffer->BindCudaBuffer(), duuDesc,\n+ 373 duvBuffer->BindCudaBuffer(), duvDesc,\n+ 374 dvvBuffer->BindCudaBuffer(), dvvDesc,\n+ 375 (int const *)stencilTable->GetSizesBuffer(),\n+ 376 (int const *)stencilTable->GetOffsetsBuffer(),\n+ 377 (int const *)stencilTable->GetIndicesBuffer(),\n+ 378 (float const *)stencilTable->GetWeightsBuffer(),\n+ 379 (float const *)stencilTable->GetDuWeightsBuffer(),\n+ 380 (float const *)stencilTable->GetDvWeightsBuffer(),\n+ 381 (float const *)stencilTable->GetDuuWeightsBuffer(),\n+ 382 (float const *)stencilTable->GetDuvWeightsBuffer(),\n+ 383 (float const *)stencilTable->GetDvvWeightsBuffer(),\n+ 384 /*start = */ 0,\n+ 385 /*end = */ stencilTable->GetNumStencils());\n+ 386 }\n+ 387\n+449 static bool EvalStencils(\n+ 450 const float *src, BufferDescriptor const &srcDesc,\n+ 451 float *dst, BufferDescriptor const &dstDesc,\n+ 452 float *du, BufferDescriptor const &duDesc,\n+ 453 float *dv, BufferDescriptor const &dvDesc,\n+ 454 float *duu, BufferDescriptor const &duuDesc,\n+ 455 float *duv, BufferDescriptor const &duvDesc,\n+ 456 float *dvv, BufferDescriptor const &dvvDesc,\n+ 457 const int * sizes,\n+ 458 const int * offsets,\n+ 459 const int * indices,\n+ 460 const float * weights,\n+ 461 const float * duWeights,\n+ 462 const float * dvWeights,\n+ 463 const float * duuWeights,\n+ 464 const float * duvWeights,\n+ 465 const float * dvvWeights,\n+ 466 int start, int end);\n+ 467\n+ 473\n+ 502 template \n+504 static bool EvalPatches(\n+ 505 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 506 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 507 int numPatchCoords,\n+ 508 PATCHCOORD_BUFFER *patchCoords,\n+ 509 PATCH_TABLE *patchTable,\n+ 510 CudaEvaluator const *instance,\n+ 511 void * deviceContext = NULL) {\n+ 512\n+ 513 (void)instance; // unused\n+ 514 (void)deviceContext; // unused\n+ 515\n+ 516 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 517 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 518 numPatchCoords,\n+ 519 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n+ 520 (const PatchArray *)patchTable->GetPatchArrayBuffer(),\n+ 521 (const int *)patchTable->GetPatchIndexBuffer(),\n+ 522 (const PatchParam *)patchTable->GetPatchParamBuffer());\n+ 523 }\n+ 524\n+ 563 template \n+565 static bool EvalPatches(\n+ 566 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 567 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 568 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 569 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 570 int numPatchCoords,\n+ 571 PATCHCOORD_BUFFER *patchCoords,\n+ 572 PATCH_TABLE *patchTable,\n+ 573 CudaEvaluator const *instance,\n+ 574 void * deviceContext = NULL) {\n+ 575\n+ 576 (void)instance; // unused\n+ 577 (void)deviceContext; // unused\n+ 578\n+ 579 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 580 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 581 duBuffer->BindCudaBuffer(), duDesc,\n+ 582 dvBuffer->BindCudaBuffer(), dvDesc,\n+ 583 numPatchCoords,\n+ 584 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n+ 585 (const PatchArray *)patchTable->GetPatchArrayBuffer(),\n+ 586 (const int *)patchTable->GetPatchIndexBuffer(),\n+ 587 (const PatchParam *)patchTable->GetPatchParamBuffer());\n+ 588 }\n+ 589\n+ 646 template \n+648 static bool EvalPatches(\n+ 649 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 650 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 651 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 652 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 653 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 654 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 655 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 656 int numPatchCoords,\n+ 657 PATCHCOORD_BUFFER *patchCoords,\n+ 658 PATCH_TABLE *patchTable,\n+ 659 CudaEvaluator const *instance,\n+ 660 void * deviceContext = NULL) {\n+ 661\n+ 662 (void)instance; // unused\n+ 663 (void)deviceContext; // unused\n+ 664\n+ 665 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 666 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 667 duBuffer->BindCudaBuffer(), duDesc,\n+ 668 dvBuffer->BindCudaBuffer(), dvDesc,\n+ 669 duuBuffer->BindCudaBuffer(), duuDesc,\n+ 670 duvBuffer->BindCudaBuffer(), duvDesc,\n+ 671 dvvBuffer->BindCudaBuffer(), dvvDesc,\n+ 672 numPatchCoords,\n+ 673 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n+ 674 (const PatchArray *)patchTable->GetPatchArrayBuffer(),\n+ 675 (const int *)patchTable->GetPatchIndexBuffer(),\n+ 676 (const PatchParam *)patchTable->GetPatchParamBuffer());\n+ 677 }\n+ 678\n+706 static bool EvalPatches(\n+ 707 const float *src, BufferDescriptor const &srcDesc,\n+ 708 float *dst, BufferDescriptor const &dstDesc,\n+ 709 int numPatchCoords,\n+ 710 const PatchCoord *patchCoords,\n+ 711 const PatchArray *patchArrays,\n+ 712 const int *patchIndices,\n+ 713 const PatchParam *patchParams);\n+ 714\n+752 static bool EvalPatches(\n+ 753 const float *src, BufferDescriptor const &srcDesc,\n+ 754 float *dst, BufferDescriptor const &dstDesc,\n+ 755 float *du, BufferDescriptor const &duDesc,\n+ 756 float *dv, BufferDescriptor const &dvDesc,\n+ 757 int numPatchCoords,\n+ 758 PatchCoord const *patchCoords,\n+ 759 PatchArray const *patchArrays,\n+ 760 const int *patchIndices,\n+ 761 PatchParam const *patchParams);\n+ 762\n+815 static bool EvalPatches(\n+ 816 const float *src, BufferDescriptor const &srcDesc,\n+ 817 float *dst, BufferDescriptor const &dstDesc,\n+ 818 float *du, BufferDescriptor const &duDesc,\n+ 819 float *dv, BufferDescriptor const &dvDesc,\n+ 820 float *duu, BufferDescriptor const &duuDesc,\n+ 821 float *duv, BufferDescriptor const &duvDesc,\n+ 822 float *dvv, BufferDescriptor const &dvvDesc,\n+ 823 int numPatchCoords,\n+ 824 PatchCoord const *patchCoords,\n+ 825 PatchArray const *patchArrays,\n+ 826 const int *patchIndices,\n+ 827 PatchParam const *patchParams);\n+ 828\n+ 857 template \n+859 static bool EvalPatchesVarying(\n+ 860 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 861 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 862 int numPatchCoords,\n+ 863 PATCHCOORD_BUFFER *patchCoords,\n+ 864 PATCH_TABLE *patchTable,\n+ 865 CudaEvaluator const *instance,\n+ 866 void * deviceContext = NULL) {\n+ 867\n+ 868 (void)instance; // unused\n+ 869 (void)deviceContext; // unused\n+ 870\n+ 871 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 872 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 873 numPatchCoords,\n+ 874 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n+ 875 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),\n+ 876 (const int *)patchTable->GetVaryingPatchIndexBuffer(),\n+ 877 (const PatchParam *)patchTable->GetPatchParamBuffer());\n+ 878 }\n+ 879\n+ 920 template \n+922 static bool EvalPatchesVarying(\n+ 923 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 924 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 925 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 926 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 927 int numPatchCoords,\n+ 928 PATCHCOORD_BUFFER *patchCoords,\n+ 929 PATCH_TABLE *patchTable,\n+ 930 CudaEvaluator const *instance,\n+ 931 void * deviceContext = NULL) {\n+ 932\n+ 933 (void)instance; // unused\n+ 934 (void)deviceContext; // unused\n+ 935\n+ 936 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 937 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 938 duBuffer->BindCudaBuffer(), duDesc,\n+ 939 dvBuffer->BindCudaBuffer(), dvDesc,\n+ 940 numPatchCoords,\n+ 941 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n+ 942 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),\n+ 943 (const int *)patchTable->GetVaryingPatchIndexBuffer(),\n+ 944 (const PatchParam *)patchTable->GetPatchParamBuffer());\n+ 945 }\n+ 946\n+ 1005 template \n+1007 static bool EvalPatchesVarying(\n+ 1008 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1009 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1010 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1011 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1012 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1013 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1014 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1015 int numPatchCoords,\n+ 1016 PATCHCOORD_BUFFER *patchCoords,\n+ 1017 PATCH_TABLE *patchTable,\n+ 1018 CudaEvaluator const *instance,\n+ 1019 void * deviceContext = NULL) {\n+ 1020\n+ 1021 (void)instance; // unused\n+ 1022 (void)deviceContext; // unused\n+ 1023\n+ 1024 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 1025 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 1026 duBuffer->BindCudaBuffer(), duDesc,\n+ 1027 dvBuffer->BindCudaBuffer(), dvDesc,\n+ 1028 duuBuffer->BindCudaBuffer(), duuDesc,\n+ 1029 duvBuffer->BindCudaBuffer(), duvDesc,\n+ 1030 dvvBuffer->BindCudaBuffer(), dvvDesc,\n+ 1031 numPatchCoords,\n+ 1032 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n+ 1033 (const PatchArray *)patchTable->GetVaryingPatchArrayBuffer(),\n+ 1034 (const int *)patchTable->GetVaryingPatchIndexBuffer(),\n+ 1035 (const PatchParam *)patchTable->GetPatchParamBuffer());\n+ 1036 }\n+ 1037\n+ 1068 template \n+1070 static bool EvalPatchesFaceVarying(\n+ 1071 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1072 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1073 int numPatchCoords,\n+ 1074 PATCHCOORD_BUFFER *patchCoords,\n+ 1075 PATCH_TABLE *patchTable,\n+ 1076 int fvarChannel,\n+ 1077 CudaEvaluator const *instance,\n+ 1078 void * deviceContext = NULL) {\n+ 1079\n+ 1080 (void)instance; // unused\n+ 1081 (void)deviceContext; // unused\n+ 1082\n+ 1083 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 1084 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 1085 numPatchCoords,\n+ 1086 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n+ 1087 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 1088 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1089 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 1090 }\n+ 1091\n+ 1134 template \n+1136 static bool EvalPatchesFaceVarying(\n+ 1137 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1138 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1139 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1140 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1141 int numPatchCoords,\n+ 1142 PATCHCOORD_BUFFER *patchCoords,\n+ 1143 PATCH_TABLE *patchTable,\n+ 1144 int fvarChannel,\n+ 1145 CudaEvaluator const *instance,\n+ 1146 void * deviceContext = NULL) {\n+ 1147\n+ 1148 (void)instance; // unused\n+ 1149 (void)deviceContext; // unused\n+ 1150\n+ 1151 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 1152 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 1153 duBuffer->BindCudaBuffer(), duDesc,\n+ 1154 dvBuffer->BindCudaBuffer(), dvDesc,\n+ 1155 numPatchCoords,\n+ 1156 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n+ 1157 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 1158 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1159 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 1160 }\n+ 1161\n+ 1222 template \n+1224 static bool EvalPatchesFaceVarying(\n+ 1225 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1226 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1227 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1228 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1229 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1230 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1231 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1232 int numPatchCoords,\n+ 1233 PATCHCOORD_BUFFER *patchCoords,\n+ 1234 PATCH_TABLE *patchTable,\n+ 1235 int fvarChannel,\n+ 1236 CudaEvaluator const *instance,\n+ 1237 void * deviceContext = NULL) {\n+ 1238\n+ 1239 (void)instance; // unused\n+ 1240 (void)deviceContext; // unused\n+ 1241\n+ 1242 return EvalPatches(srcBuffer->BindCudaBuffer(), srcDesc,\n+ 1243 dstBuffer->BindCudaBuffer(), dstDesc,\n+ 1244 duBuffer->BindCudaBuffer(), duDesc,\n+ 1245 dvBuffer->BindCudaBuffer(), dvDesc,\n+ 1246 duuBuffer->BindCudaBuffer(), duuDesc,\n+ 1247 duvBuffer->BindCudaBuffer(), duvDesc,\n+ 1248 dvvBuffer->BindCudaBuffer(), dvvDesc,\n+ 1249 numPatchCoords,\n+ 1250 (const PatchCoord *)patchCoords->BindCudaBuffer(),\n+ 1251 (const PatchArray *)patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 1252 (const int *)patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1253 (const PatchParam *)patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 1254 }\n+ 1255\n+1261 static void Synchronize(void *deviceContext = NULL);\n+ 1262};\n+ 1263\n+ 1264\n+ 1265} // end namespace Osd\n+ 1266\n+ 1267} // end namespace OPENSUBDIV_VERSION\n+ 1268using namespace OPENSUBDIV_VERSION;\n+ 1269\n+ 1270} // end namespace OpenSubdiv\n+ 1271\n+ 1272\n+ 1273#endif // OPENSUBDIV3_OSD_CUDA_EVALUATOR_H\n+OpenSubdiv\n+Definition: limits.h:32\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n+Stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:273\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n+Limit stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:583\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n+BufferDescriptor is a struct which describes buffer elements in interleaved\n+data buffers....\n+Definition: bufferDescriptor.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable\n+CUDA stencil table.\n+Definition: cudaEvaluator.h:52\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetDuvWeightsBuffer\n+void * GetDuvWeightsBuffer() const\n+Definition: cudaEvaluator.h:77\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::~CudaStencilTable\n+~CudaStencilTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetDuWeightsBuffer\n+void * GetDuWeightsBuffer() const\n+Definition: cudaEvaluator.h:74\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetDvvWeightsBuffer\n+void * GetDvvWeightsBuffer() const\n+Definition: cudaEvaluator.h:78\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetOffsetsBuffer\n+void * GetOffsetsBuffer() const\n+Definition: cudaEvaluator.h:71\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::CudaStencilTable\n+CudaStencilTable(Far::LimitStencilTable const *limitStencilTable)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetDvWeightsBuffer\n+void * GetDvWeightsBuffer() const\n+Definition: cudaEvaluator.h:75\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetNumStencils\n+int GetNumStencils() const\n+Definition: cudaEvaluator.h:79\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetDuuWeightsBuffer\n+void * GetDuuWeightsBuffer() const\n+Definition: cudaEvaluator.h:76\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetWeightsBuffer\n+void * GetWeightsBuffer() const\n+Definition: cudaEvaluator.h:73\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::Create\n+static CudaStencilTable * Create(Far::LimitStencilTable const\n+*limitStencilTable, void *deviceContext=NULL)\n+Definition: cudaEvaluator.h:59\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetSizesBuffer\n+void * GetSizesBuffer() const\n+Definition: cudaEvaluator.h:70\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::GetIndicesBuffer\n+void * GetIndicesBuffer() const\n+Definition: cudaEvaluator.h:72\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::CudaStencilTable\n+CudaStencilTable(Far::StencilTable const *stencilTable)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::Create\n+static CudaStencilTable * Create(Far::StencilTable const *stencilTable, void\n+*deviceContext=NULL)\n+Definition: cudaEvaluator.h:54\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator\n+Definition: cudaEvaluator.h:94\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n+static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const\n+PatchCoord *patchCoords, const PatchArray *patchArrays, const int\n+*patchIndices, const PatchParam *patchParams)\n+Static limit eval function. It takes an array of PatchCoord and evaluate limit\n+values on given PatchT...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void\n+*deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cudaEvaluator.h:922\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const *instance, void\n+*deviceContext=NULL)\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: cudaEvaluator.h:648\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n+const CudaEvaluator *instance=NULL, void *deviceContext=NULL)\n+Generic static eval stencils function with derivatives. This function has a\n+same signature as other d...\n+Definition: cudaEvaluator.h:353\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n+static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int\n+*offsets, const int *indices, const float *weights, int start, int end)\n+Static eval stencils function which takes raw cuda buffers for input and\n+output.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n+CudaEvaluator const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cudaEvaluator.h:859\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const\n+CudaEvaluator *instance=NULL, void *deviceContext=NULL)\n+Generic static eval stencils function with derivatives. This function has a\n+same signature as other d...\n+Definition: cudaEvaluator.h:218\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n+static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n+BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n+float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord\n+const *patchCoords, PatchArray const *patchArrays, const int *patchIndices,\n+PatchParam const *patchParams)\n+Static limit eval function. It takes an array of PatchCoord and evaluate limit\n+values on given PatchT...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CudaEvaluator const\n+*instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cudaEvaluator.h:1136\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n+const *stencilTable, const void *instance=NULL, void *deviceContext=NULL)\n+Generic static compute function. This function has a same signature as other\n+device kernels have so t...\n+Definition: cudaEvaluator.h:126\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n+static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int\n+*offsets, const int *indices, const float *weights, const float *duWeights,\n+const float *dvWeights, int start, int end)\n+Static eval stencils function with derivatives, which takes raw cuda pointers\n+for input and output.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalStencils\n+static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n+BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n+float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int\n+*offsets, const int *indices, const float *weights, const float *duWeights,\n+const float *dvWeights, const float *duuWeights, const float *duvWeights, const\n+float *dvvWeights, int start, int end)\n+Static eval stencils function with derivatives, which takes raw cuda pointers\n+for input and output.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::Synchronize\n+static void Synchronize(void *deviceContext=NULL)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n+static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords,\n+PatchCoord const *patchCoords, PatchArray const *patchArrays, const int\n+*patchIndices, PatchParam const *patchParams)\n+Static limit eval function. It takes an array of PatchCoord and evaluate limit\n+values on given PatchT...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const\n+*instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cudaEvaluator.h:1007\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CudaEvaluator const\n+*instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cudaEvaluator.h:504\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n+CudaEvaluator const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cudaEvaluator.h:1224\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable, CudaEvaluator const *instance, void *deviceContext=NULL)\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: cudaEvaluator.h:565\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n+fvarChannel, CudaEvaluator const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cudaEvaluator.h:1070\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord\n+Coordinates set on a patch table.\n+Definition: types.h:42\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArray\n+Definition: types.h:66\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParam\n+Definition: types.h:127\n * opensubdiv\n * osd\n- * opencl.h\n+ * cudaEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00797.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00797.html", "unified_diff": "@@ -87,15 +87,15 @@\n
\n Classes |\n Namespaces
\n
clVertexBuffer.h File Reference
\n \n
\n
#include "../version.h"
\n-#include "../osd/opencl.h"
\n+#include "../osd/opencl.h"
\n
\n

Go to the source code of this file.

\n \n \n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00800.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00800.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/clEvaluator.h File Reference\n+OpenSubdiv: opensubdiv/osd/clPatchTable.h File Reference\n \n \n \n \n \n \n \n@@ -83,30 +83,29 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
clEvaluator.h File Reference
\n+
clPatchTable.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../osd/opencl.h"
\n+#include "../osd/opencl.h"
\n+#include "../osd/nonCopyable.h"
\n #include "../osd/types.h"
\n-#include "../osd/bufferDescriptor.h"
\n+#include <vector>
\n
\n

Go to the source code of this file.

\n

\n Classes

class  CLVertexBuffer
 Concrete vertex buffer class for OpenCL subdivision. More...
\n \n-\n-\n-\n-\n+\n+\n \n

\n Classes

class  CLStencilTable
 OpenCL stencil table. More...
 
class  CLEvaluator
class  CLPatchTable
 CL patch table. More...
 
\n \n \n \n \n@@ -117,13 +116,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,32 +5,31 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-clEvaluator.h File Reference\n+clPatchTable.h File Reference\n #include \"../version.h\"\n #include \"../osd/opencl.h\"\n+#include \"../osd/nonCopyable.h\"\n #include \"../osd/types.h\"\n-#include \"../osd/bufferDescriptor.h\"\n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0CLStencilTable\n-\u00a0 OpenCL stencil table. More...\n-\u00a0\n-class \u00a0CLEvaluator\n+class \u00a0CLPatchTable\n+\u00a0 CL patch table. More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * clEvaluator.h\n+ * clPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00800.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00800.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,3 @@\n var a00800 = [\n- [\"CLStencilTable\", \"a01137.html\", \"a01137\"],\n- [\"CLEvaluator\", \"a01141.html\", \"a01141\"]\n+ [\"CLPatchTable\", \"a01145.html\", \"a01145\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00800_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00800_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/clEvaluator.h Source File\n+OpenSubdiv: opensubdiv/osd/clPatchTable.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
clEvaluator.h
\n+
clPatchTable.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,1217 +107,120 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_CL_EVALUATOR_H
\n-
26#define OPENSUBDIV3_OSD_CL_EVALUATOR_H
\n+
25#ifndef OPENSUBDIV3_OSD_CL_PATCH_TABLE_H
\n+
26#define OPENSUBDIV3_OSD_CL_PATCH_TABLE_H
\n
27
\n
28#include "../version.h"
\n
29
\n
30#include "../osd/opencl.h"
\n-
31#include "../osd/types.h"
\n-
32#include "../osd/bufferDescriptor.h"
\n+
31#include "../osd/nonCopyable.h"
\n+
32#include "../osd/types.h"
\n
33
\n-
34namespace OpenSubdiv {
\n-
35namespace OPENSUBDIV_VERSION {
\n-
36
\n-
37namespace Far {
\n-
38 class PatchTable;
\n-
39 class StencilTable;
\n-
40 class LimitStencilTable;
\n-
41}
\n+
34#include <vector>
\n+
35
\n+
36namespace OpenSubdiv {
\n+
37namespace OPENSUBDIV_VERSION {
\n+
38
\n+
39namespace Far{
\n+
40 class PatchTable;
\n+
41};
\n
42
\n
43namespace Osd {
\n
44
\n-
52class CLStencilTable {
\n+
52class CLPatchTable : private NonCopyable<CLPatchTable> {
\n
53public:
\n-
54 template <typename DEVICE_CONTEXT>
\n-
55 static CLStencilTable *Create(Far::StencilTable const *stencilTable,
\n-
56 DEVICE_CONTEXT context) {
\n-
57 return new CLStencilTable(stencilTable, context->GetContext());
\n-
58 }
\n-
59
\n-
60 template <typename DEVICE_CONTEXT>
\n-
61 static CLStencilTable *Create(
\n-
62 Far::LimitStencilTable const *limitStencilTable,
\n-
63 DEVICE_CONTEXT context) {
\n-
64 return new CLStencilTable(limitStencilTable, context->GetContext());
\n-
65 }
\n+
55 static CLPatchTable *Create(Far::PatchTable const *patchTable,
\n+
56 cl_context clContext);
\n+
57
\n+
58 template <typename DEVICE_CONTEXT>
\n+
59 static CLPatchTable * Create(Far::PatchTable const *patchTable,
\n+
60 DEVICE_CONTEXT context) {
\n+
61 return Create(patchTable, context->GetContext());
\n+
62 }
\n+
63
\n+
65 ~CLPatchTable();
\n
66
\n-
67 CLStencilTable(Far::StencilTable const *stencilTable,
\n-
68 cl_context clContext);
\n-
69 CLStencilTable(Far::LimitStencilTable const *limitStencilTable,
\n-
70 cl_context clContext);
\n-
71 ~CLStencilTable();
\n+
68 cl_mem GetPatchArrayBuffer() const { return _patchArrays; }
\n+
69
\n+
71 cl_mem GetPatchIndexBuffer() const { return _indexBuffer; }
\n
72
\n-
73 // interfaces needed for CLComputeKernel
\n-
74 cl_mem GetSizesBuffer() const { return _sizes; }
\n-
75 cl_mem GetOffsetsBuffer() const { return _offsets; }
\n-
76 cl_mem GetIndicesBuffer() const { return _indices; }
\n-
77 cl_mem GetWeightsBuffer() const { return _weights; }
\n-
78 cl_mem GetDuWeightsBuffer() const { return _duWeights; }
\n-
79 cl_mem GetDvWeightsBuffer() const { return _dvWeights; }
\n-
80 cl_mem GetDuuWeightsBuffer() const { return _duuWeights; }
\n-
81 cl_mem GetDuvWeightsBuffer() const { return _duvWeights; }
\n-
82 cl_mem GetDvvWeightsBuffer() const { return _dvvWeights; }
\n-
83 int GetNumStencils() const { return _numStencils; }
\n+
74 cl_mem GetPatchParamBuffer() const { return _patchParamBuffer; }
\n+
75
\n+
77 cl_mem GetVaryingPatchArrayBuffer() const { return _varyingPatchArrays; }
\n+
78
\n+
80 cl_mem GetVaryingPatchIndexBuffer() const { return _varyingIndexBuffer; }
\n+
81
\n+
83 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }
\n
84
\n-
85private:
\n-
86 cl_mem _sizes;
\n-
87 cl_mem _offsets;
\n-
88 cl_mem _indices;
\n-
89 cl_mem _weights;
\n-
90 cl_mem _duWeights;
\n-
91 cl_mem _dvWeights;
\n-
92 cl_mem _duuWeights;
\n-
93 cl_mem _duvWeights;
\n-
94 cl_mem _dvvWeights;
\n-
95 int _numStencils;
\n-
96};
\n-
97
\n-
98// ---------------------------------------------------------------------------
\n-
99
\n-
100class CLEvaluator {
\n-
101public:
\n-
102 typedef bool Instantiatable;
\n-
103
\n-
105 template <typename DEVICE_CONTEXT>
\n-
106 static CLEvaluator *Create(BufferDescriptor const &srcDesc,
\n-
107 BufferDescriptor const &dstDesc,
\n-
108 BufferDescriptor const &duDesc,
\n-
109 BufferDescriptor const &dvDesc,
\n-
110 DEVICE_CONTEXT deviceContext) {
\n-
111 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n-
112 deviceContext->GetContext(),
\n-
113 deviceContext->GetCommandQueue());
\n-
114 }
\n-
115
\n-
116 static CLEvaluator * Create(BufferDescriptor const &srcDesc,
\n-
117 BufferDescriptor const &dstDesc,
\n-
118 BufferDescriptor const &duDesc,
\n-
119 BufferDescriptor const &dvDesc,
\n-
120 cl_context clContext,
\n-
121 cl_command_queue clCommandQueue) {
\n-
122 CLEvaluator *instance = new CLEvaluator(clContext, clCommandQueue);
\n-
123 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc))
\n-
124 return instance;
\n-
125 delete instance;
\n-
126 return NULL;
\n-
127 }
\n-
128
\n-
130 template <typename DEVICE_CONTEXT>
\n-
131 static CLEvaluator *Create(BufferDescriptor const &srcDesc,
\n-
132 BufferDescriptor const &dstDesc,
\n-
133 BufferDescriptor const &duDesc,
\n-
134 BufferDescriptor const &dvDesc,
\n-
135 BufferDescriptor const &duuDesc,
\n-
136 BufferDescriptor const &duvDesc,
\n-
137 BufferDescriptor const &dvvDesc,
\n-
138 DEVICE_CONTEXT deviceContext) {
\n-
139 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n-
140 duuDesc, duvDesc, dvvDesc,
\n-
141 deviceContext->GetContext(),
\n-
142 deviceContext->GetCommandQueue());
\n-
143 }
\n-
144
\n-
145 static CLEvaluator * Create(BufferDescriptor const &srcDesc,
\n-
146 BufferDescriptor const &dstDesc,
\n-
147 BufferDescriptor const &duDesc,
\n-
148 BufferDescriptor const &dvDesc,
\n-
149 BufferDescriptor const &duuDesc,
\n-
150 BufferDescriptor const &duvDesc,
\n-
151 BufferDescriptor const &dvvDesc,
\n-
152 cl_context clContext,
\n-
153 cl_command_queue clCommandQueue) {
\n-
154 CLEvaluator *instance = new CLEvaluator(clContext, clCommandQueue);
\n-
155 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc,
\n-
156 duuDesc, duvDesc, dvvDesc))
\n-
157 return instance;
\n-
158 delete instance;
\n-
159 return NULL;
\n-
160 }
\n-
161
\n-
163 CLEvaluator(cl_context context, cl_command_queue queue);
\n-
164
\n-
166 ~CLEvaluator();
\n-
167
\n-
173
\n-
217 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
218 typename STENCIL_TABLE, typename DEVICE_CONTEXT>
\n-
219 static bool EvalStencils(
\n-
220 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
221 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
222 STENCIL_TABLE const *stencilTable,
\n-
223 CLEvaluator const *instance,
\n-
224 DEVICE_CONTEXT deviceContext,
\n-
225 unsigned int numStartEvents=0,
\n-
226 const cl_event* startEvents=NULL,
\n-
227 cl_event* endEvent=NULL) {
\n-
228
\n-
229 if (instance) {
\n-
230 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
231 dstBuffer, dstDesc,
\n-
232 stencilTable,
\n-
233 numStartEvents, startEvents, endEvent);
\n-
234 } else {
\n-
235 // Create an instance on demand (slow)
\n-
236 instance = Create(srcDesc, dstDesc,
\n-
237 BufferDescriptor(),
\n-
238 BufferDescriptor(),
\n-
239 deviceContext);
\n-
240 if (instance) {
\n-
241 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
242 dstBuffer, dstDesc,
\n-
243 stencilTable,
\n-
244 numStartEvents, startEvents, endEvent);
\n-
245 delete instance;
\n-
246 return r;
\n-
247 }
\n-
248 return false;
\n-
249 }
\n-
250 }
\n-
251
\n-
307 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
308 typename STENCIL_TABLE, typename DEVICE_CONTEXT>
\n-
309 static bool EvalStencils(
\n-
310 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
311 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
312 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
313 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
314 STENCIL_TABLE const *stencilTable,
\n-
315 CLEvaluator const *instance,
\n-
316 DEVICE_CONTEXT deviceContext,
\n-
317 unsigned int numStartEvents=0,
\n-
318 const cl_event* startEvents=NULL,
\n-
319 cl_event* endEvent=NULL) {
\n-
320
\n-
321 if (instance) {
\n-
322 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
323 dstBuffer, dstDesc,
\n-
324 duBuffer, duDesc,
\n-
325 dvBuffer, dvDesc,
\n-
326 stencilTable,
\n-
327 numStartEvents, startEvents, endEvent);
\n-
328 } else {
\n-
329 // Create an instance on demand (slow)
\n-
330 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,
\n-
331 deviceContext);
\n-
332 if (instance) {
\n-
333 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
334 dstBuffer, dstDesc,
\n-
335 duBuffer, duDesc,
\n-
336 dvBuffer, dvDesc,
\n-
337 stencilTable,
\n-
338 numStartEvents, startEvents, endEvent);
\n-
339 delete instance;
\n-
340 return r;
\n-
341 }
\n-
342 return false;
\n-
343 }
\n-
344 }
\n-
345
\n-
419 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
420 typename STENCIL_TABLE, typename DEVICE_CONTEXT>
\n-
421 static bool EvalStencils(
\n-
422 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
423 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
424 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
425 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
426 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
427 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
428 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
429 STENCIL_TABLE const *stencilTable,
\n-
430 CLEvaluator const *instance,
\n-
431 DEVICE_CONTEXT deviceContext,
\n-
432 unsigned int numStartEvents=0,
\n-
433 const cl_event* startEvents=NULL,
\n-
434 cl_event* endEvent=NULL) {
\n-
435
\n-
436 if (instance) {
\n-
437 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
438 dstBuffer, dstDesc,
\n-
439 duBuffer, duDesc,
\n-
440 dvBuffer, dvDesc,
\n-
441 duuBuffer, duuDesc,
\n-
442 duvBuffer, duvDesc,
\n-
443 dvvBuffer, dvvDesc,
\n-
444 stencilTable,
\n-
445 numStartEvents, startEvents, endEvent);
\n-
446 } else {
\n-
447 // Create an instance on demand (slow)
\n-
448 instance = Create(srcDesc, dstDesc,
\n-
449 duDesc, dvDesc,
\n-
450 duuDesc, duvDesc, dvvDesc,
\n-
451 deviceContext);
\n-
452 if (instance) {
\n-
453 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
454 dstBuffer, dstDesc,
\n-
455 duBuffer, duDesc,
\n-
456 dvBuffer, dvDesc,
\n-
457 duuBuffer, duuDesc,
\n-
458 duvBuffer, duvDesc,
\n-
459 dvvBuffer, dvvDesc,
\n-
460 stencilTable,
\n-
461 numStartEvents, startEvents, endEvent);
\n-
462 delete instance;
\n-
463 return r;
\n-
464 }
\n-
465 return false;
\n-
466 }
\n-
467 }
\n-
468
\n-
499 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
500 bool EvalStencils(
\n-
501 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
502 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
503 STENCIL_TABLE const *stencilTable,
\n-
504 unsigned int numStartEvents=0,
\n-
505 const cl_event* startEvents=NULL,
\n-
506 cl_event* endEvent=NULL) const {
\n-
507 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
508 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
509 stencilTable->GetSizesBuffer(),
\n-
510 stencilTable->GetOffsetsBuffer(),
\n-
511 stencilTable->GetIndicesBuffer(),
\n-
512 stencilTable->GetWeightsBuffer(),
\n-
513 0,
\n-
514 stencilTable->GetNumStencils(),
\n-
515 numStartEvents, startEvents, endEvent);
\n-
516 }
\n-
517
\n-
560 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
561 bool EvalStencils(
\n-
562 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
563 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
564 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
565 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
566 STENCIL_TABLE const *stencilTable,
\n-
567 unsigned int numStartEvents=0,
\n-
568 const cl_event* startEvents=NULL,
\n-
569 cl_event* endEvent=NULL) const {
\n-
570 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
571 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
572 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n-
573 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n-
574 stencilTable->GetSizesBuffer(),
\n-
575 stencilTable->GetOffsetsBuffer(),
\n-
576 stencilTable->GetIndicesBuffer(),
\n-
577 stencilTable->GetWeightsBuffer(),
\n-
578 stencilTable->GetDuWeightsBuffer(),
\n-
579 stencilTable->GetDvWeightsBuffer(),
\n-
580 0,
\n-
581 stencilTable->GetNumStencils(),
\n-
582 numStartEvents, startEvents, endEvent);
\n-
583 }
\n-
584
\n-
645 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
646 bool EvalStencils(
\n-
647 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
648 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
649 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
650 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
651 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
652 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
653 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
654 STENCIL_TABLE const *stencilTable,
\n-
655 unsigned int numStartEvents=0,
\n-
656 const cl_event* startEvents=NULL,
\n-
657 cl_event* endEvent=NULL) const {
\n-
658 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
659 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
660 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n-
661 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n-
662 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,
\n-
663 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,
\n-
664 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,
\n-
665 stencilTable->GetSizesBuffer(),
\n-
666 stencilTable->GetOffsetsBuffer(),
\n-
667 stencilTable->GetIndicesBuffer(),
\n-
668 stencilTable->GetWeightsBuffer(),
\n-
669 stencilTable->GetDuWeightsBuffer(),
\n-
670 stencilTable->GetDvWeightsBuffer(),
\n-
671 stencilTable->GetDuuWeightsBuffer(),
\n-
672 stencilTable->GetDuvWeightsBuffer(),
\n-
673 stencilTable->GetDvvWeightsBuffer(),
\n-
674 0,
\n-
675 stencilTable->GetNumStencils(),
\n-
676 numStartEvents, startEvents, endEvent);
\n-
677 }
\n-
678
\n-
681 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,
\n-
682 cl_mem dst, BufferDescriptor const &dstDesc,
\n-
683 cl_mem sizes,
\n-
684 cl_mem offsets,
\n-
685 cl_mem indices,
\n-
686 cl_mem weights,
\n-
687 int start,
\n-
688 int end,
\n-
689 unsigned int numStartEvents=0,
\n-
690 const cl_event* startEvents=NULL,
\n-
691 cl_event* endEvent=NULL) const;
\n-
692
\n-
741 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,
\n-
742 cl_mem dst, BufferDescriptor const &dstDesc,
\n-
743 cl_mem du, BufferDescriptor const &duDesc,
\n-
744 cl_mem dv, BufferDescriptor const &dvDesc,
\n-
745 cl_mem sizes,
\n-
746 cl_mem offsets,
\n-
747 cl_mem indices,
\n-
748 cl_mem weights,
\n-
749 cl_mem duWeights,
\n-
750 cl_mem dvWeights,
\n-
751 int start,
\n-
752 int end,
\n-
753 unsigned int numStartEvents=0,
\n-
754 const cl_event* startEvents=NULL,
\n-
755 cl_event* endEvent=NULL) const;
\n-
756
\n-
823 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,
\n-
824 cl_mem dst, BufferDescriptor const &dstDesc,
\n-
825 cl_mem du, BufferDescriptor const &duDesc,
\n-
826 cl_mem dv, BufferDescriptor const &dvDesc,
\n-
827 cl_mem duu, BufferDescriptor const &duuDesc,
\n-
828 cl_mem duv, BufferDescriptor const &duvDesc,
\n-
829 cl_mem dvv, BufferDescriptor const &dvvDesc,
\n-
830 cl_mem sizes,
\n-
831 cl_mem offsets,
\n-
832 cl_mem indices,
\n-
833 cl_mem weights,
\n-
834 cl_mem duWeights,
\n-
835 cl_mem dvWeights,
\n-
836 cl_mem duuWeights,
\n-
837 cl_mem duvWeights,
\n-
838 cl_mem dvvWeights,
\n-
839 int start,
\n-
840 int end,
\n-
841 unsigned int numStartEvents=0,
\n-
842 const cl_event* startEvents=NULL,
\n-
843 cl_event* endEvent=NULL) const;
\n-
844
\n-
850
\n-
899 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
900 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n-
901 typename DEVICE_CONTEXT>
\n-
902 static bool EvalPatches(
\n-
903 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
904 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
905 int numPatchCoords,
\n-
906 PATCHCOORD_BUFFER *patchCoords,
\n-
907 PATCH_TABLE *patchTable,
\n-
908 CLEvaluator const *instance,
\n-
909 DEVICE_CONTEXT deviceContext,
\n-
910 unsigned int numStartEvents=0,
\n-
911 const cl_event* startEvents=NULL,
\n-
912 cl_event* endEvent=NULL) {
\n-
913
\n-
914 if (instance) {
\n-
915 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
916 dstBuffer, dstDesc,
\n-
917 numPatchCoords, patchCoords,
\n-
918 patchTable,
\n-
919 numStartEvents, startEvents, endEvent);
\n-
920 } else {
\n-
921 // Create an instance on demand (slow)
\n-
922 (void)deviceContext; // unused
\n-
923 instance = Create(srcDesc, dstDesc,
\n-
924 BufferDescriptor(),
\n-
925 BufferDescriptor(),
\n-
926 deviceContext);
\n-
927 if (instance) {
\n-
928 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
929 dstBuffer, dstDesc,
\n-
930 numPatchCoords, patchCoords,
\n-
931 patchTable,
\n-
932 numStartEvents, startEvents, endEvent);
\n-
933 delete instance;
\n-
934 return r;
\n-
935 }
\n-
936 return false;
\n-
937 }
\n-
938 }
\n-
939
\n-
1000 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1001 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n-
1002 typename DEVICE_CONTEXT>
\n-
1003 static bool EvalPatches(
\n-
1004 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1005 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1006 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1007 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1008 int numPatchCoords,
\n-
1009 PATCHCOORD_BUFFER *patchCoords,
\n-
1010 PATCH_TABLE *patchTable,
\n-
1011 CLEvaluator const *instance,
\n-
1012 DEVICE_CONTEXT deviceContext,
\n-
1013 unsigned int numStartEvents=0,
\n-
1014 const cl_event* startEvents=NULL,
\n-
1015 cl_event* endEvent=NULL) {
\n-
1016
\n-
1017 if (instance) {
\n-
1018 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
1019 dstBuffer, dstDesc,
\n-
1020 duBuffer, duDesc,
\n-
1021 dvBuffer, dvDesc,
\n-
1022 numPatchCoords, patchCoords,
\n-
1023 patchTable,
\n-
1024 numStartEvents, startEvents, endEvent);
\n-
1025 } else {
\n-
1026 // Create an instance on demand (slow)
\n-
1027 (void)deviceContext; // unused
\n-
1028 instance = Create(srcDesc, dstDesc, duDesc, dvDesc, deviceContext);
\n-
1029 if (instance) {
\n-
1030 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
1031 dstBuffer, dstDesc,
\n-
1032 duBuffer, duDesc,
\n-
1033 dvBuffer, dvDesc,
\n-
1034 numPatchCoords, patchCoords,
\n-
1035 patchTable,
\n-
1036 numStartEvents, startEvents, endEvent);
\n-
1037 delete instance;
\n-
1038 return r;
\n-
1039 }
\n-
1040 return false;
\n-
1041 }
\n-
1042 }
\n-
1043
\n-
1122 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1123 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n-
1124 typename DEVICE_CONTEXT>
\n-
1125 static bool EvalPatches(
\n-
1126 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1127 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1128 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1129 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1130 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1131 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1132 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1133 int numPatchCoords,
\n-
1134 PATCHCOORD_BUFFER *patchCoords,
\n-
1135 PATCH_TABLE *patchTable,
\n-
1136 CLEvaluator const *instance,
\n-
1137 DEVICE_CONTEXT deviceContext,
\n-
1138 unsigned int numStartEvents=0,
\n-
1139 const cl_event* startEvents=NULL,
\n-
1140 cl_event* endEvent=NULL) {
\n-
1141
\n-
1142 if (instance) {
\n-
1143 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
1144 dstBuffer, dstDesc,
\n-
1145 duBuffer, duDesc,
\n-
1146 dvBuffer, dvDesc,
\n-
1147 duuBuffer, duuDesc,
\n-
1148 duvBuffer, duvDesc,
\n-
1149 dvvBuffer, dvvDesc,
\n-
1150 numPatchCoords, patchCoords,
\n-
1151 patchTable,
\n-
1152 numStartEvents, startEvents, endEvent);
\n-
1153 } else {
\n-
1154 // Create an instance on demand (slow)
\n-
1155 (void)deviceContext; // unused
\n-
1156 instance = Create(srcDesc, dstDesc,
\n-
1157 duDesc, dvDesc,
\n-
1158 duuDesc, duvDesc, dvvDesc,
\n-
1159 deviceContext);
\n-
1160 if (instance) {
\n-
1161 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
1162 dstBuffer, dstDesc,
\n-
1163 duBuffer, duDesc,
\n-
1164 dvBuffer, dvDesc,
\n-
1165 duuBuffer, duuDesc,
\n-
1166 duvBuffer, duvDesc,
\n-
1167 dvvBuffer, dvvDesc,
\n-
1168 numPatchCoords, patchCoords,
\n-
1169 patchTable,
\n-
1170 numStartEvents, startEvents, endEvent);
\n-
1171 delete instance;
\n-
1172 return r;
\n-
1173 }
\n-
1174 return false;
\n-
1175 }
\n-
1176 }
\n-
1177
\n-
1215 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1216 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1217 bool EvalPatches(
\n-
1218 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1219 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1220 int numPatchCoords,
\n-
1221 PATCHCOORD_BUFFER *patchCoords,
\n-
1222 PATCH_TABLE *patchTable,
\n-
1223 unsigned int numStartEvents=0,
\n-
1224 const cl_event* startEvents=NULL,
\n-
1225 cl_event* endEvent=NULL) const {
\n-
1226
\n-
1227 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
1228 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
1229 0, BufferDescriptor(),
\n-
1230 0, BufferDescriptor(),
\n-
1231 numPatchCoords,
\n-
1232 patchCoords->BindCLBuffer(_clCommandQueue),
\n-
1233 patchTable->GetPatchArrayBuffer(),
\n-
1234 patchTable->GetPatchIndexBuffer(),
\n-
1235 patchTable->GetPatchParamBuffer(),
\n-
1236 numStartEvents, startEvents, endEvent);
\n-
1237 }
\n-
1238
\n-
1286 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1287 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1288 bool EvalPatches(
\n-
1289 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1290 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1291 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1292 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1293 int numPatchCoords,
\n-
1294 PATCHCOORD_BUFFER *patchCoords,
\n-
1295 PATCH_TABLE *patchTable,
\n-
1296 unsigned int numStartEvents=0,
\n-
1297 const cl_event* startEvents=NULL,
\n-
1298 cl_event* endEvent=NULL) const {
\n-
1299
\n-
1300 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
1301 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
1302 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n-
1303 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n-
1304 numPatchCoords,
\n-
1305 patchCoords->BindCLBuffer(_clCommandQueue),
\n-
1306 patchTable->GetPatchArrayBuffer(),
\n-
1307 patchTable->GetPatchIndexBuffer(),
\n-
1308 patchTable->GetPatchParamBuffer(),
\n-
1309 numStartEvents, startEvents, endEvent);
\n-
1310 }
\n-
1311
\n-
1377 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1378 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1379 bool EvalPatches(
\n-
1380 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1381 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1382 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1383 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1384 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1385 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1386 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1387 int numPatchCoords,
\n-
1388 PATCHCOORD_BUFFER *patchCoords,
\n-
1389 PATCH_TABLE *patchTable,
\n-
1390 unsigned int numStartEvents=0,
\n-
1391 const cl_event* startEvents=NULL,
\n-
1392 cl_event* endEvent=NULL) const {
\n-
1393
\n-
1394 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
1395 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
1396 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n-
1397 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n-
1398 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,
\n-
1399 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,
\n-
1400 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,
\n-
1401 numPatchCoords,
\n-
1402 patchCoords->BindCLBuffer(_clCommandQueue),
\n-
1403 patchTable->GetPatchArrayBuffer(),
\n-
1404 patchTable->GetPatchIndexBuffer(),
\n-
1405 patchTable->GetPatchParamBuffer(),
\n-
1406 numStartEvents, startEvents, endEvent);
\n-
1407 }
\n-
1408
\n-
1409 bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc,
\n-
1410 cl_mem dst, BufferDescriptor const &dstDesc,
\n-
1411 cl_mem du, BufferDescriptor const &duDesc,
\n-
1412 cl_mem dv, BufferDescriptor const &dvDesc,
\n-
1413 int numPatchCoords,
\n-
1414 cl_mem patchCoordsBuffer,
\n-
1415 cl_mem patchArrayBuffer,
\n-
1416 cl_mem patchIndexBuffer,
\n-
1417 cl_mem patchParamsBuffer,
\n-
1418 unsigned int numStartEvents=0,
\n-
1419 const cl_event* startEvents=NULL,
\n-
1420 cl_event* endEvent=NULL) const;
\n-
1421
\n-
1422 bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc,
\n-
1423 cl_mem dst, BufferDescriptor const &dstDesc,
\n-
1424 cl_mem du, BufferDescriptor const &duDesc,
\n-
1425 cl_mem dv, BufferDescriptor const &dvDesc,
\n-
1426 cl_mem duu, BufferDescriptor const &duuDesc,
\n-
1427 cl_mem duv, BufferDescriptor const &duvDesc,
\n-
1428 cl_mem dvv, BufferDescriptor const &dvvDesc,
\n-
1429 int numPatchCoords,
\n-
1430 cl_mem patchCoordsBuffer,
\n-
1431 cl_mem patchArrayBuffer,
\n-
1432 cl_mem patchIndexBuffer,
\n-
1433 cl_mem patchParamsBuffer,
\n-
1434 unsigned int numStartEvents=0,
\n-
1435 const cl_event* startEvents=NULL,
\n-
1436 cl_event* endEvent=NULL) const;
\n-
1437
\n-
1486 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1487 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n-
1488 typename DEVICE_CONTEXT>
\n-
1489 static bool EvalPatchesVarying(
\n-
1490 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1491 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1492 int numPatchCoords,
\n-
1493 PATCHCOORD_BUFFER *patchCoords,
\n-
1494 PATCH_TABLE *patchTable,
\n-
1495 CLEvaluator const *instance,
\n-
1496 DEVICE_CONTEXT deviceContext,
\n-
1497 unsigned int numStartEvents=0,
\n-
1498 const cl_event* startEvents=NULL,
\n-
1499 cl_event* endEvent=NULL) {
\n-
1500
\n-
1501 if (instance) {
\n-
1502 return instance->EvalPatchesVarying(
\n-
1503 srcBuffer, srcDesc,
\n-
1504 dstBuffer, dstDesc,
\n-
1505 numPatchCoords, patchCoords,
\n-
1506 patchTable,
\n-
1507 numStartEvents, startEvents, endEvent);
\n-
1508 } else {
\n-
1509 // Create an instance on demand (slow)
\n-
1510 (void)deviceContext; // unused
\n-
1511 instance = Create(srcDesc, dstDesc,
\n-
1512 BufferDescriptor(),
\n-
1513 BufferDescriptor(),
\n-
1514 deviceContext);
\n-
1515 if (instance) {
\n-
1516 bool r = instance->EvalPatchesVarying(
\n-
1517 srcBuffer, srcDesc,
\n-
1518 dstBuffer, dstDesc,
\n-
1519 numPatchCoords, patchCoords,
\n-
1520 patchTable,
\n-
1521 numStartEvents, startEvents, endEvent);
\n-
1522 delete instance;
\n-
1523 return r;
\n-
1524 }
\n-
1525 return false;
\n-
1526 }
\n-
1527 }
\n-
1528
\n-
1566 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1567 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1568 bool EvalPatchesVarying(
\n-
1569 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1570 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1571 int numPatchCoords,
\n-
1572 PATCHCOORD_BUFFER *patchCoords,
\n-
1573 PATCH_TABLE *patchTable,
\n-
1574 unsigned int numStartEvents=0,
\n-
1575 const cl_event* startEvents=NULL,
\n-
1576 cl_event* endEvent=NULL) const {
\n-
1577
\n-
1578 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
1579 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
1580 0, BufferDescriptor(),
\n-
1581 0, BufferDescriptor(),
\n-
1582 numPatchCoords,
\n-
1583 patchCoords->BindCLBuffer(_clCommandQueue),
\n-
1584 patchTable->GetVaryingPatchArrayBuffer(),
\n-
1585 patchTable->GetVaryingPatchIndexBuffer(),
\n-
1586 patchTable->GetPatchParamBuffer(),
\n-
1587 numStartEvents, startEvents, endEvent);
\n-
1588 }
\n-
1589
\n-
1650 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1651 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n-
1652 typename DEVICE_CONTEXT>
\n-
1653 static bool EvalPatchesVarying(
\n-
1654 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1655 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1656 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1657 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1658 int numPatchCoords,
\n-
1659 PATCHCOORD_BUFFER *patchCoords,
\n-
1660 PATCH_TABLE *patchTable,
\n-
1661 CLEvaluator const *instance,
\n-
1662 DEVICE_CONTEXT deviceContext,
\n-
1663 unsigned int numStartEvents=0,
\n-
1664 const cl_event* startEvents=NULL,
\n-
1665 cl_event* endEvent=NULL) {
\n-
1666
\n-
1667 if (instance) {
\n-
1668 return instance->EvalPatchesVarying(
\n-
1669 srcBuffer, srcDesc,
\n-
1670 dstBuffer, dstDesc,
\n-
1671 duBuffer, duDesc,
\n-
1672 dvBuffer, dvDesc,
\n-
1673 numPatchCoords, patchCoords,
\n-
1674 patchTable,
\n-
1675 numStartEvents, startEvents, endEvent);
\n-
1676 } else {
\n-
1677 // Create an instance on demand (slow)
\n-
1678 (void)deviceContext; // unused
\n-
1679 instance = Create(srcDesc, dstDesc,
\n-
1680 duDesc, dvDesc,
\n-
1681 deviceContext);
\n-
1682 if (instance) {
\n-
1683 bool r = instance->EvalPatchesVarying(
\n-
1684 srcBuffer, srcDesc,
\n-
1685 dstBuffer, dstDesc,
\n-
1686 duBuffer, duDesc,
\n-
1687 dvBuffer, dvDesc,
\n-
1688 numPatchCoords, patchCoords,
\n-
1689 patchTable,
\n-
1690 numStartEvents, startEvents, endEvent);
\n-
1691 delete instance;
\n-
1692 return r;
\n-
1693 }
\n-
1694 return false;
\n-
1695 }
\n-
1696 }
\n-
1697
\n-
1747 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1748 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1749 bool EvalPatchesVarying(
\n-
1750 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1751 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1752 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1753 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1754 int numPatchCoords,
\n-
1755 PATCHCOORD_BUFFER *patchCoords,
\n-
1756 PATCH_TABLE *patchTable,
\n-
1757 unsigned int numStartEvents=0,
\n-
1758 const cl_event* startEvents=NULL,
\n-
1759 cl_event* endEvent=NULL) const {
\n-
1760
\n-
1761 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
1762 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
1763 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n-
1764 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n-
1765 numPatchCoords,
\n-
1766 patchCoords->BindCLBuffer(_clCommandQueue),
\n-
1767 patchTable->GetVaryingPatchArrayBuffer(),
\n-
1768 patchTable->GetVaryingPatchIndexBuffer(),
\n-
1769 patchTable->GetPatchParamBuffer(),
\n-
1770 numStartEvents, startEvents, endEvent);
\n-
1771 }
\n-
1772
\n-
1851 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1852 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n-
1853 typename DEVICE_CONTEXT>
\n-
1854 static bool EvalPatchesVarying(
\n-
1855 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1856 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1857 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1858 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1859 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1860 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1861 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1862 int numPatchCoords,
\n-
1863 PATCHCOORD_BUFFER *patchCoords,
\n-
1864 PATCH_TABLE *patchTable,
\n-
1865 CLEvaluator const *instance,
\n-
1866 DEVICE_CONTEXT deviceContext,
\n-
1867 unsigned int numStartEvents=0,
\n-
1868 const cl_event* startEvents=NULL,
\n-
1869 cl_event* endEvent=NULL) {
\n-
1870
\n-
1871 if (instance) {
\n-
1872 return instance->EvalPatchesVarying(
\n-
1873 srcBuffer, srcDesc,
\n-
1874 dstBuffer, dstDesc,
\n-
1875 duBuffer, duDesc,
\n-
1876 dvBuffer, dvDesc,
\n-
1877 duuBuffer, duuDesc,
\n-
1878 duvBuffer, duvDesc,
\n-
1879 dvvBuffer, dvvDesc,
\n-
1880 numPatchCoords, patchCoords,
\n-
1881 patchTable,
\n-
1882 numStartEvents, startEvents, endEvent);
\n-
1883 } else {
\n-
1884 // Create an instance on demand (slow)
\n-
1885 (void)deviceContext; // unused
\n-
1886 instance = Create(srcDesc, dstDesc,
\n-
1887 duDesc, dvDesc,
\n-
1888 duuDesc, duvDesc, dvvDesc,
\n-
1889 deviceContext);
\n-
1890 if (instance) {
\n-
1891 bool r = instance->EvalPatchesVarying(
\n-
1892 srcBuffer, srcDesc,
\n-
1893 dstBuffer, dstDesc,
\n-
1894 duBuffer, duDesc,
\n-
1895 dvBuffer, dvDesc,
\n-
1896 duuBuffer, duuDesc,
\n-
1897 duvBuffer, duvDesc,
\n-
1898 dvvBuffer, dvvDesc,
\n-
1899 numPatchCoords, patchCoords,
\n-
1900 patchTable,
\n-
1901 numStartEvents, startEvents, endEvent);
\n-
1902 delete instance;
\n-
1903 return r;
\n-
1904 }
\n-
1905 return false;
\n-
1906 }
\n-
1907 }
\n-
1908
\n-
1976 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1977 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1978 bool EvalPatchesVarying(
\n-
1979 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1980 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1981 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1982 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1983 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1984 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1985 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1986 int numPatchCoords,
\n-
1987 PATCHCOORD_BUFFER *patchCoords,
\n-
1988 PATCH_TABLE *patchTable,
\n-
1989 unsigned int numStartEvents=0,
\n-
1990 const cl_event* startEvents=NULL,
\n-
1991 cl_event* endEvent=NULL) const {
\n-
1992
\n-
1993 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
1994 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
1995 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n-
1996 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n-
1997 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,
\n-
1998 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,
\n-
1999 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,
\n-
2000 numPatchCoords,
\n-
2001 patchCoords->BindCLBuffer(_clCommandQueue),
\n-
2002 patchTable->GetVaryingPatchArrayBuffer(),
\n-
2003 patchTable->GetVaryingPatchIndexBuffer(),
\n-
2004 patchTable->GetPatchParamBuffer(),
\n-
2005 numStartEvents, startEvents, endEvent);
\n-
2006 }
\n-
2007
\n-
2058 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
2059 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n-
2060 typename DEVICE_CONTEXT>
\n-
2061 static bool EvalPatchesFaceVarying(
\n-
2062 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
2063 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
2064 int numPatchCoords,
\n-
2065 PATCHCOORD_BUFFER *patchCoords,
\n-
2066 PATCH_TABLE *patchTable,
\n-
2067 int fvarChannel,
\n-
2068 CLEvaluator const *instance,
\n-
2069 DEVICE_CONTEXT deviceContext,
\n-
2070 unsigned int numStartEvents=0,
\n-
2071 const cl_event* startEvents=NULL,
\n-
2072 cl_event* endEvent=NULL) {
\n-
2073
\n-
2074 if (instance) {
\n-
2075 return instance->EvalPatchesFaceVarying(
\n-
2076 srcBuffer, srcDesc,
\n-
2077 dstBuffer, dstDesc,
\n-
2078 numPatchCoords, patchCoords,
\n-
2079 patchTable, fvarChannel,
\n-
2080 numStartEvents, startEvents, endEvent);
\n-
2081 } else {
\n-
2082 // Create an instance on demand (slow)
\n-
2083 (void)deviceContext; // unused
\n-
2084 instance = Create(srcDesc, dstDesc,
\n-
2085 BufferDescriptor(),
\n-
2086 BufferDescriptor(),
\n-
2087 deviceContext);
\n-
2088 if (instance) {
\n-
2089 bool r = instance->EvalPatchesFaceVarying(
\n-
2090 srcBuffer, srcDesc,
\n-
2091 dstBuffer, dstDesc,
\n-
2092 numPatchCoords, patchCoords,
\n-
2093 patchTable, fvarChannel,
\n-
2094 numStartEvents, startEvents, endEvent);
\n-
2095 delete instance;
\n-
2096 return r;
\n-
2097 }
\n-
2098 return false;
\n-
2099 }
\n-
2100 }
\n-
2101
\n-
2141 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
2142 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
2143 bool EvalPatchesFaceVarying(
\n-
2144 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
2145 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
2146 int numPatchCoords,
\n-
2147 PATCHCOORD_BUFFER *patchCoords,
\n-
2148 PATCH_TABLE *patchTable,
\n-
2149 int fvarChannel = 0,
\n-
2150 unsigned int numStartEvents=0,
\n-
2151 const cl_event* startEvents=NULL,
\n-
2152 cl_event* endEvent=NULL) const {
\n-
2153
\n-
2154 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
2155 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
2156 0, BufferDescriptor(),
\n-
2157 0, BufferDescriptor(),
\n-
2158 numPatchCoords,
\n-
2159 patchCoords->BindCLBuffer(_clCommandQueue),
\n-
2160 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
2161 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
2162 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n-
2163 numStartEvents, startEvents, endEvent);
\n-
2164 }
\n-
2165
\n-
2228 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
2229 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n-
2230 typename DEVICE_CONTEXT>
\n-
2231 static bool EvalPatchesFaceVarying(
\n-
2232 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
2233 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
2234 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
2235 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
2236 int numPatchCoords,
\n-
2237 PATCHCOORD_BUFFER *patchCoords,
\n-
2238 PATCH_TABLE *patchTable,
\n-
2239 int fvarChannel,
\n-
2240 CLEvaluator const *instance,
\n-
2241 DEVICE_CONTEXT deviceContext,
\n-
2242 unsigned int numStartEvents=0,
\n-
2243 const cl_event* startEvents=NULL,
\n-
2244 cl_event* endEvent=NULL) {
\n-
2245
\n-
2246 if (instance) {
\n-
2247 return instance->EvalPatchesFaceVarying(
\n-
2248 srcBuffer, srcDesc,
\n-
2249 dstBuffer, dstDesc,
\n-
2250 duBuffer, duDesc,
\n-
2251 dvBuffer, dvDesc,
\n-
2252 numPatchCoords, patchCoords,
\n-
2253 patchTable, fvarChannel,
\n-
2254 numStartEvents, startEvents, endEvent);
\n-
2255 } else {
\n-
2256 // Create an instance on demand (slow)
\n-
2257 (void)deviceContext; // unused
\n-
2258 instance = Create(srcDesc, dstDesc,
\n-
2259 duDesc, dvDesc, deviceContext);
\n-
2260 if (instance) {
\n-
2261 bool r = instance->EvalPatchesFaceVarying(
\n-
2262 srcBuffer, srcDesc,
\n-
2263 dstBuffer, dstDesc,
\n-
2264 duBuffer, duDesc,
\n-
2265 dvBuffer, dvDesc,
\n-
2266 numPatchCoords, patchCoords,
\n-
2267 patchTable, fvarChannel,
\n-
2268 numStartEvents, startEvents, endEvent);
\n-
2269 delete instance;
\n-
2270 return r;
\n-
2271 }
\n-
2272 return false;
\n-
2273 }
\n-
2274 }
\n-
2275
\n-
2327 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
2328 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
2329 bool EvalPatchesFaceVarying(
\n-
2330 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
2331 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
2332 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
2333 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
2334 int numPatchCoords,
\n-
2335 PATCHCOORD_BUFFER *patchCoords,
\n-
2336 PATCH_TABLE *patchTable,
\n-
2337 int fvarChannel = 0,
\n-
2338 unsigned int numStartEvents=0,
\n-
2339 const cl_event* startEvents=NULL,
\n-
2340 cl_event* endEvent=NULL) const {
\n-
2341
\n-
2342 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
2343 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
2344 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n-
2345 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n-
2346 numPatchCoords,
\n-
2347 patchCoords->BindCLBuffer(_clCommandQueue),
\n-
2348 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
2349 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
2350 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n-
2351 numStartEvents, startEvents, endEvent);
\n-
2352 }
\n-
2353
\n-
2434 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
2435 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n-
2436 typename DEVICE_CONTEXT>
\n-
2437 static bool EvalPatchesFaceVarying(
\n-
2438 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
2439 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
2440 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
2441 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
2442 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
2443 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
2444 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
2445 int numPatchCoords,
\n-
2446 PATCHCOORD_BUFFER *patchCoords,
\n-
2447 PATCH_TABLE *patchTable,
\n-
2448 int fvarChannel,
\n-
2449 CLEvaluator const *instance,
\n-
2450 DEVICE_CONTEXT deviceContext,
\n-
2451 unsigned int numStartEvents=0,
\n-
2452 const cl_event* startEvents=NULL,
\n-
2453 cl_event* endEvent=NULL) {
\n-
2454
\n-
2455 if (instance) {
\n-
2456 return instance->EvalPatchesFaceVarying(
\n-
2457 srcBuffer, srcDesc,
\n-
2458 dstBuffer, dstDesc,
\n-
2459 duBuffer, duDesc,
\n-
2460 dvBuffer, dvDesc,
\n-
2461 duuBuffer, duuDesc,
\n-
2462 duvBuffer, duvDesc,
\n-
2463 dvvBuffer, dvvDesc,
\n-
2464 numPatchCoords, patchCoords,
\n-
2465 patchTable, fvarChannel,
\n-
2466 numStartEvents, startEvents, endEvent);
\n-
2467 } else {
\n-
2468 // Create an instance on demand (slow)
\n-
2469 (void)deviceContext; // unused
\n-
2470 instance = Create(srcDesc, dstDesc,
\n-
2471 duDesc, dvDesc,
\n-
2472 duuDesc, duvDesc, dvvDesc,
\n-
2473 deviceContext);
\n-
2474 if (instance) {
\n-
2475 bool r = instance->EvalPatchesFaceVarying(
\n-
2476 srcBuffer, srcDesc,
\n-
2477 dstBuffer, dstDesc,
\n-
2478 duBuffer, duDesc,
\n-
2479 dvBuffer, dvDesc,
\n-
2480 duuBuffer, duuDesc,
\n-
2481 duvBuffer, duvDesc,
\n-
2482 dvvBuffer, dvvDesc,
\n-
2483 numPatchCoords, patchCoords,
\n-
2484 patchTable, fvarChannel,
\n-
2485 numStartEvents, startEvents, endEvent);
\n-
2486 delete instance;
\n-
2487 return r;
\n-
2488 }
\n-
2489 return false;
\n-
2490 }
\n-
2491 }
\n-
2492
\n-
2562 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
2563 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
2564 bool EvalPatchesFaceVarying(
\n-
2565 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
2566 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
2567 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
2568 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
2569 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
2570 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
2571 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
2572 int numPatchCoords,
\n-
2573 PATCHCOORD_BUFFER *patchCoords,
\n-
2574 PATCH_TABLE *patchTable,
\n-
2575 int fvarChannel = 0,
\n-
2576 unsigned int numStartEvents=0,
\n-
2577 const cl_event* startEvents=NULL,
\n-
2578 cl_event* endEvent=NULL) const {
\n-
2579
\n-
2580 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n-
2581 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n-
2582 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n-
2583 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n-
2584 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,
\n-
2585 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,
\n-
2586 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,
\n-
2587 numPatchCoords,
\n-
2588 patchCoords->BindCLBuffer(_clCommandQueue),
\n-
2589 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
2590 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
2591 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n-
2592 numStartEvents, startEvents, endEvent);
\n-
2593 }
\n-
2594
\n-
2600
\n-
2603 bool Compile(BufferDescriptor const &srcDesc,
\n-
2604 BufferDescriptor const &dstDesc,
\n-
2605 BufferDescriptor const &duDesc = BufferDescriptor(),
\n-
2606 BufferDescriptor const &dvDesc = BufferDescriptor(),
\n-
2607 BufferDescriptor const &duuDesc = BufferDescriptor(),
\n-
2608 BufferDescriptor const &duvDesc = BufferDescriptor(),
\n-
2609 BufferDescriptor const &dvvDesc = BufferDescriptor());
\n-
2610
\n-
2612 template <typename DEVICE_CONTEXT>
\n-
2613 static void Synchronize(DEVICE_CONTEXT deviceContext) {
\n-
2614 Synchronize(deviceContext->GetCommandQueue());
\n-
2615 }
\n-
2616
\n-
2617 static void Synchronize(cl_command_queue queue);
\n-
2618
\n-
2619private:
\n-
2620 cl_context _clContext;
\n-
2621 cl_command_queue _clCommandQueue;
\n-
2622 cl_program _program;
\n-
2623 cl_kernel _stencilKernel;
\n-
2624 cl_kernel _stencilDerivKernel;
\n-
2625 cl_kernel _patchKernel;
\n-
2626};
\n-
2627
\n-
2628} // end namespace Osd
\n-
2629
\n-
2630} // end namespace OPENSUBDIV_VERSION
\n-
2631using namespace OPENSUBDIV_VERSION;
\n-
2632
\n-
2633} // end namespace OpenSubdiv
\n-
2634
\n-
2635
\n-
2636#endif // OPENSUBDIV3_OSD_CL_EVALUATOR_H
\n+
86 cl_mem GetFVarPatchArrayBuffer(int fvarChannel = 0) const { return _fvarPatchArrays[fvarChannel]; }
\n+
87
\n+
89 cl_mem GetFVarPatchIndexBuffer(int fvarChannel = 0) const { return _fvarIndexBuffers[fvarChannel]; }
\n+
90
\n+
92 cl_mem GetFVarPatchParamBuffer(int fvarChannel = 0) const { return _fvarParamBuffers[fvarChannel]; }
\n+
93
\n+
94protected:
\n+
95 CLPatchTable();
\n+
96
\n+
97 bool allocate(Far::PatchTable const *patchTable, cl_context clContext);
\n+
98
\n+
99 cl_mem _patchArrays;
\n+
100 cl_mem _indexBuffer;
\n+
101 cl_mem _patchParamBuffer;
\n+
102
\n+
103 cl_mem _varyingPatchArrays;
\n+
104 cl_mem _varyingIndexBuffer;
\n+
105
\n+
106 std::vector<cl_mem> _fvarPatchArrays;
\n+
107 std::vector<cl_mem> _fvarIndexBuffers;
\n+
108 std::vector<cl_mem> _fvarParamBuffers;
\n+
109
\n+
110};
\n+
111
\n+
112} // end namespace Osd
\n+
113
\n+
114} // end namespace OPENSUBDIV_VERSION
\n+
115using namespace OPENSUBDIV_VERSION;
\n+
116
\n+
117} // end namespace OpenSubdiv
\n+
118
\n+
119#endif // OPENSUBDIV3_OSD_CL_PATCH_TABLE_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::CLStencilTable
CLStencilTable(Far::StencilTable const *stencilTable, cl_context clContext)
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::Create
static CLStencilTable * Create(Far::LimitStencilTable const *limitStencilTable, DEVICE_CONTEXT context)
Definition: clEvaluator.h:61
\n-\n-\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::Create
static CLStencilTable * Create(Far::StencilTable const *stencilTable, DEVICE_CONTEXT context)
Definition: clEvaluator.h:55
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::CLStencilTable
CLStencilTable(Far::LimitStencilTable const *limitStencilTable, cl_context clContext)
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: clEvaluator.h:309
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Synchronize
static void Synchronize(cl_command_queue queue)
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst, BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc, cl_mem dv, BufferDescriptor const &dvDesc, cl_mem sizes, cl_mem offsets, cl_mem indices, cl_mem weights, cl_mem duWeights, cl_mem dvWeights, int start, int end, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Dispatch the CL compute kernel asynchronously. returns false if the kernel hasn't been compiled yet.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: clEvaluator.h:219
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create
static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, cl_context clContext, cl_command_queue clCommandQueue)
Definition: clEvaluator.h:116
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2231
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst, BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc, cl_mem dv, BufferDescriptor const &dvDesc, int numPatchCoords, cl_mem patchCoordsBuffer, cl_mem patchArrayBuffer, cl_mem patchIndexBuffer, cl_mem patchParamsBuffer, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1978
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic stencil function.
Definition: clEvaluator.h:561
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: clEvaluator.h:1379
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create
static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, DEVICE_CONTEXT deviceContext)
Generic creator template.
Definition: clEvaluator.h:106
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2061
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create
static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, DEVICE_CONTEXT deviceContext)
Generic creator template.
Definition: clEvaluator.h:131
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst, BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc, cl_mem dv, BufferDescriptor const &dvDesc, cl_mem duu, BufferDescriptor const &duuDesc, cl_mem duv, BufferDescriptor const &duvDesc, cl_mem dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, cl_mem patchCoordsBuffer, cl_mem patchArrayBuffer, cl_mem patchIndexBuffer, cl_mem patchParamsBuffer, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst, BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc, cl_mem dv, BufferDescriptor const &dvDesc, cl_mem duu, BufferDescriptor const &duuDesc, cl_mem duv, BufferDescriptor const &duvDesc, cl_mem dvv, BufferDescriptor const &dvvDesc, cl_mem sizes, cl_mem offsets, cl_mem indices, cl_mem weights, cl_mem duWeights, cl_mem dvWeights, cl_mem duuWeights, cl_mem duvWeights, cl_mem dvvWeights, int start, int end, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Dispatch the CL compute kernel asynchronously. returns false if the kernel hasn't been compiled yet.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2564
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1217
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic stencil function.
Definition: clEvaluator.h:646
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1653
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Synchronize
static void Synchronize(DEVICE_CONTEXT deviceContext)
Wait the OpenCL kernels finish.
Definition: clEvaluator.h:2613
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1125
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: clEvaluator.h:421
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst, BufferDescriptor const &dstDesc, cl_mem sizes, cl_mem offsets, cl_mem indices, cl_mem weights, int start, int end, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1489
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2329
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2437
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create
static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, cl_context clContext, cl_command_queue clCommandQueue)
Definition: clEvaluator.h:145
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1003
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1749
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:902
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1568
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1854
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic stencil function.
Definition: clEvaluator.h:500
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: clEvaluator.h:1288
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::CLEvaluator
CLEvaluator(cl_context context, cl_command_queue queue)
Constructor.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Compile
bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc=BufferDescriptor(), BufferDescriptor const &dvDesc=BufferDescriptor(), BufferDescriptor const &duuDesc=BufferDescriptor(), BufferDescriptor const &duvDesc=BufferDescriptor(), BufferDescriptor const &dvvDesc=BufferDescriptor())
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2143
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchParamBuffer
cl_mem GetPatchParamBuffer() const
Returns the CL memory of the array of Osd::PatchParam buffer.
Definition: clPatchTable.h:74
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchArrayBuffer
cl_mem GetFVarPatchArrayBuffer(int fvarChannel=0) const
Returns the CL memory of the array of Osd::PatchArray buffer.
Definition: clPatchTable.h:86
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchArrayBuffer
cl_mem GetPatchArrayBuffer() const
Returns the CL memory of the array of Osd::PatchArray buffer.
Definition: clPatchTable.h:68
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::Create
static CLPatchTable * Create(Far::PatchTable const *patchTable, cl_context clContext)
Creator. Returns NULL if error.
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchIndexBuffer
cl_mem GetFVarPatchIndexBuffer(int fvarChannel=0) const
Returns the CL memory of the face-varying control vertices.
Definition: clPatchTable.h:89
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetVaryingPatchIndexBuffer
cl_mem GetVaryingPatchIndexBuffer() const
Returns the CL memory of the varying control vertices.
Definition: clPatchTable.h:80
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetNumFVarChannels
int GetNumFVarChannels() const
Returns the number of face-varying channel buffers.
Definition: clPatchTable.h:83
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchIndexBuffer
cl_mem GetPatchIndexBuffer() const
Returns the CL memory of the patch control vertices.
Definition: clPatchTable.h:71
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::Create
static CLPatchTable * Create(Far::PatchTable const *patchTable, DEVICE_CONTEXT context)
Definition: clPatchTable.h:59
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchParamBuffer
cl_mem GetFVarPatchParamBuffer(int fvarChannel=0) const
Returns the CL memory of the array of Osd::PatchParam buffer.
Definition: clPatchTable.h:92
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::allocate
bool allocate(Far::PatchTable const *patchTable, cl_context clContext)
\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetVaryingPatchArrayBuffer
cl_mem GetVaryingPatchArrayBuffer() const
Returns the CL memory of the array of Osd::PatchArray buffer.
Definition: clPatchTable.h:77
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-clEvaluator.h\n+clPatchTable.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,1556 +30,174 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_CL_EVALUATOR_H\n- 26#define OPENSUBDIV3_OSD_CL_EVALUATOR_H\n+ 25#ifndef OPENSUBDIV3_OSD_CL_PATCH_TABLE_H\n+ 26#define OPENSUBDIV3_OSD_CL_PATCH_TABLE_H\n 27\n 28#include \"../version.h\"\n 29\n 30#include \"../osd/opencl.h\"\n- 31#include \"../osd/types.h\"\n- 32#include \"../osd/bufferDescriptor.h\"\n+ 31#include \"../osd/nonCopyable.h\"\n+ 32#include \"../osd/types.h\"\n 33\n- 34namespace OpenSubdiv {\n- 35namespace OPENSUBDIV_VERSION {\n- 36\n- 37namespace Far {\n- 38 class PatchTable;\n- 39 class StencilTable;\n- 40 class LimitStencilTable;\n- 41}\n+ 34#include \n+ 35\n+ 36namespace OpenSubdiv {\n+ 37namespace OPENSUBDIV_VERSION {\n+ 38\n+ 39namespace Far{\n+ 40 class PatchTable;\n+ 41};\n 42\n 43namespace Osd {\n 44\n-52class CLStencilTable {\n+52class CLPatchTable : private NonCopyable {\n 53public:\n- 54 template \n-55 static CLStencilTable *Create(Far::StencilTable const *stencilTable,\n- 56 DEVICE_CONTEXT context) {\n- 57 return new CLStencilTable(stencilTable, context->GetContext());\n- 58 }\n- 59\n- 60 template \n-61 static CLStencilTable *Create(\n- 62 Far::LimitStencilTable const *limitStencilTable,\n- 63 DEVICE_CONTEXT context) {\n- 64 return new CLStencilTable(limitStencilTable, context->GetContext());\n- 65 }\n+55 static CLPatchTable *Create(Far::PatchTable const *patchTable,\n+ 56 cl_context clContext);\n+ 57\n+ 58 template \n+59 static CLPatchTable * Create(Far::PatchTable const *patchTable,\n+ 60 DEVICE_CONTEXT context) {\n+ 61 return Create(patchTable, context->GetContext());\n+ 62 }\n+ 63\n+65 ~CLPatchTable();\n 66\n-67 CLStencilTable(Far::StencilTable const *stencilTable,\n- 68 cl_context clContext);\n-69 CLStencilTable(Far::LimitStencilTable const *limitStencilTable,\n- 70 cl_context clContext);\n-71 ~CLStencilTable();\n+68 cl_mem GetPatchArrayBuffer() const { return _patchArrays; }\n+ 69\n+71 cl_mem GetPatchIndexBuffer() const { return _indexBuffer; }\n 72\n- 73 // interfaces needed for CLComputeKernel\n-74 cl_mem GetSizesBuffer() const { return _sizes; }\n-75 cl_mem GetOffsetsBuffer() const { return _offsets; }\n-76 cl_mem GetIndicesBuffer() const { return _indices; }\n-77 cl_mem GetWeightsBuffer() const { return _weights; }\n-78 cl_mem GetDuWeightsBuffer() const { return _duWeights; }\n-79 cl_mem GetDvWeightsBuffer() const { return _dvWeights; }\n-80 cl_mem GetDuuWeightsBuffer() const { return _duuWeights; }\n-81 cl_mem GetDuvWeightsBuffer() const { return _duvWeights; }\n-82 cl_mem GetDvvWeightsBuffer() const { return _dvvWeights; }\n-83 int GetNumStencils() const { return _numStencils; }\n+74 cl_mem GetPatchParamBuffer() const { return _patchParamBuffer; }\n+ 75\n+77 cl_mem GetVaryingPatchArrayBuffer() const { return _varyingPatchArrays; }\n+ 78\n+80 cl_mem GetVaryingPatchIndexBuffer() const { return _varyingIndexBuffer; }\n+ 81\n+83 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }\n 84\n- 85private:\n- 86 cl_mem _sizes;\n- 87 cl_mem _offsets;\n- 88 cl_mem _indices;\n- 89 cl_mem _weights;\n- 90 cl_mem _duWeights;\n- 91 cl_mem _dvWeights;\n- 92 cl_mem _duuWeights;\n- 93 cl_mem _duvWeights;\n- 94 cl_mem _dvvWeights;\n- 95 int _numStencils;\n- 96};\n- 97\n- 98// -------------------------------------------------------------------------\n---\n- 99\n-100class CLEvaluator {\n- 101public:\n-102 typedef bool Instantiatable;\n- 103\n- 105 template \n-106 static CLEvaluator *Create(BufferDescriptor const &srcDesc,\n- 107 BufferDescriptor const &dstDesc,\n- 108 BufferDescriptor const &duDesc,\n- 109 BufferDescriptor const &dvDesc,\n- 110 DEVICE_CONTEXT deviceContext) {\n- 111 return Create(srcDesc, dstDesc, duDesc, dvDesc,\n- 112 deviceContext->GetContext(),\n- 113 deviceContext->GetCommandQueue());\n- 114 }\n- 115\n-116 static CLEvaluator * Create(BufferDescriptor const &srcDesc,\n- 117 BufferDescriptor const &dstDesc,\n- 118 BufferDescriptor const &duDesc,\n- 119 BufferDescriptor const &dvDesc,\n- 120 cl_context clContext,\n- 121 cl_command_queue clCommandQueue) {\n- 122 CLEvaluator *instance = new CLEvaluator(clContext, clCommandQueue);\n- 123 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc))\n- 124 return instance;\n- 125 delete instance;\n- 126 return NULL;\n- 127 }\n- 128\n- 130 template \n-131 static CLEvaluator *Create(BufferDescriptor const &srcDesc,\n- 132 BufferDescriptor const &dstDesc,\n- 133 BufferDescriptor const &duDesc,\n- 134 BufferDescriptor const &dvDesc,\n- 135 BufferDescriptor const &duuDesc,\n- 136 BufferDescriptor const &duvDesc,\n- 137 BufferDescriptor const &dvvDesc,\n- 138 DEVICE_CONTEXT deviceContext) {\n- 139 return Create(srcDesc, dstDesc, duDesc, dvDesc,\n- 140 duuDesc, duvDesc, dvvDesc,\n- 141 deviceContext->GetContext(),\n- 142 deviceContext->GetCommandQueue());\n- 143 }\n- 144\n-145 static CLEvaluator * Create(BufferDescriptor const &srcDesc,\n- 146 BufferDescriptor const &dstDesc,\n- 147 BufferDescriptor const &duDesc,\n- 148 BufferDescriptor const &dvDesc,\n- 149 BufferDescriptor const &duuDesc,\n- 150 BufferDescriptor const &duvDesc,\n- 151 BufferDescriptor const &dvvDesc,\n- 152 cl_context clContext,\n- 153 cl_command_queue clCommandQueue) {\n- 154 CLEvaluator *instance = new CLEvaluator(clContext, clCommandQueue);\n- 155 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc,\n- 156 duuDesc, duvDesc, dvvDesc))\n- 157 return instance;\n- 158 delete instance;\n- 159 return NULL;\n- 160 }\n- 161\n-163 CLEvaluator(cl_context context, cl_command_queue queue);\n- 164\n-166 ~CLEvaluator();\n- 167\n- 173\n- 217 template \n-219 static bool EvalStencils(\n- 220 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 221 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 222 STENCIL_TABLE const *stencilTable,\n- 223 CLEvaluator const *instance,\n- 224 DEVICE_CONTEXT deviceContext,\n- 225 unsigned int numStartEvents=0,\n- 226 const cl_event* startEvents=NULL,\n- 227 cl_event* endEvent=NULL) {\n- 228\n- 229 if (instance) {\n- 230 return instance->EvalStencils(srcBuffer, srcDesc,\n- 231 dstBuffer, dstDesc,\n- 232 stencilTable,\n- 233 numStartEvents, startEvents, endEvent);\n- 234 } else {\n- 235 // Create an instance on demand (slow)\n- 236 instance = Create(srcDesc, dstDesc,\n- 237 BufferDescriptor(),\n- 238 BufferDescriptor(),\n- 239 deviceContext);\n- 240 if (instance) {\n- 241 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 242 dstBuffer, dstDesc,\n- 243 stencilTable,\n- 244 numStartEvents, startEvents, endEvent);\n- 245 delete instance;\n- 246 return r;\n- 247 }\n- 248 return false;\n- 249 }\n- 250 }\n- 251\n- 307 template \n-309 static bool EvalStencils(\n- 310 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 311 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 312 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 313 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 314 STENCIL_TABLE const *stencilTable,\n- 315 CLEvaluator const *instance,\n- 316 DEVICE_CONTEXT deviceContext,\n- 317 unsigned int numStartEvents=0,\n- 318 const cl_event* startEvents=NULL,\n- 319 cl_event* endEvent=NULL) {\n- 320\n- 321 if (instance) {\n- 322 return instance->EvalStencils(srcBuffer, srcDesc,\n- 323 dstBuffer, dstDesc,\n- 324 duBuffer, duDesc,\n- 325 dvBuffer, dvDesc,\n- 326 stencilTable,\n- 327 numStartEvents, startEvents, endEvent);\n- 328 } else {\n- 329 // Create an instance on demand (slow)\n- 330 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,\n- 331 deviceContext);\n- 332 if (instance) {\n- 333 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 334 dstBuffer, dstDesc,\n- 335 duBuffer, duDesc,\n- 336 dvBuffer, dvDesc,\n- 337 stencilTable,\n- 338 numStartEvents, startEvents, endEvent);\n- 339 delete instance;\n- 340 return r;\n- 341 }\n- 342 return false;\n- 343 }\n- 344 }\n- 345\n- 419 template \n-421 static bool EvalStencils(\n- 422 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 423 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 424 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 425 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 426 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 427 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 428 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 429 STENCIL_TABLE const *stencilTable,\n- 430 CLEvaluator const *instance,\n- 431 DEVICE_CONTEXT deviceContext,\n- 432 unsigned int numStartEvents=0,\n- 433 const cl_event* startEvents=NULL,\n- 434 cl_event* endEvent=NULL) {\n- 435\n- 436 if (instance) {\n- 437 return instance->EvalStencils(srcBuffer, srcDesc,\n- 438 dstBuffer, dstDesc,\n- 439 duBuffer, duDesc,\n- 440 dvBuffer, dvDesc,\n- 441 duuBuffer, duuDesc,\n- 442 duvBuffer, duvDesc,\n- 443 dvvBuffer, dvvDesc,\n- 444 stencilTable,\n- 445 numStartEvents, startEvents, endEvent);\n- 446 } else {\n- 447 // Create an instance on demand (slow)\n- 448 instance = Create(srcDesc, dstDesc,\n- 449 duDesc, dvDesc,\n- 450 duuDesc, duvDesc, dvvDesc,\n- 451 deviceContext);\n- 452 if (instance) {\n- 453 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 454 dstBuffer, dstDesc,\n- 455 duBuffer, duDesc,\n- 456 dvBuffer, dvDesc,\n- 457 duuBuffer, duuDesc,\n- 458 duvBuffer, duvDesc,\n- 459 dvvBuffer, dvvDesc,\n- 460 stencilTable,\n- 461 numStartEvents, startEvents, endEvent);\n- 462 delete instance;\n- 463 return r;\n- 464 }\n- 465 return false;\n- 466 }\n- 467 }\n- 468\n- 499 template \n-500 bool EvalStencils(\n- 501 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 502 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 503 STENCIL_TABLE const *stencilTable,\n- 504 unsigned int numStartEvents=0,\n- 505 const cl_event* startEvents=NULL,\n- 506 cl_event* endEvent=NULL) const {\n- 507 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 508 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 509 stencilTable->GetSizesBuffer(),\n- 510 stencilTable->GetOffsetsBuffer(),\n- 511 stencilTable->GetIndicesBuffer(),\n- 512 stencilTable->GetWeightsBuffer(),\n- 513 0,\n- 514 stencilTable->GetNumStencils(),\n- 515 numStartEvents, startEvents, endEvent);\n- 516 }\n- 517\n- 560 template \n-561 bool EvalStencils(\n- 562 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 563 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 564 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 565 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 566 STENCIL_TABLE const *stencilTable,\n- 567 unsigned int numStartEvents=0,\n- 568 const cl_event* startEvents=NULL,\n- 569 cl_event* endEvent=NULL) const {\n- 570 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 571 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 572 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n- 573 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n- 574 stencilTable->GetSizesBuffer(),\n- 575 stencilTable->GetOffsetsBuffer(),\n- 576 stencilTable->GetIndicesBuffer(),\n- 577 stencilTable->GetWeightsBuffer(),\n- 578 stencilTable->GetDuWeightsBuffer(),\n- 579 stencilTable->GetDvWeightsBuffer(),\n- 580 0,\n- 581 stencilTable->GetNumStencils(),\n- 582 numStartEvents, startEvents, endEvent);\n- 583 }\n- 584\n- 645 template \n-646 bool EvalStencils(\n- 647 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 648 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 649 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 650 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 651 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 652 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 653 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 654 STENCIL_TABLE const *stencilTable,\n- 655 unsigned int numStartEvents=0,\n- 656 const cl_event* startEvents=NULL,\n- 657 cl_event* endEvent=NULL) const {\n- 658 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 659 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 660 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n- 661 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n- 662 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,\n- 663 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,\n- 664 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,\n- 665 stencilTable->GetSizesBuffer(),\n- 666 stencilTable->GetOffsetsBuffer(),\n- 667 stencilTable->GetIndicesBuffer(),\n- 668 stencilTable->GetWeightsBuffer(),\n- 669 stencilTable->GetDuWeightsBuffer(),\n- 670 stencilTable->GetDvWeightsBuffer(),\n- 671 stencilTable->GetDuuWeightsBuffer(),\n- 672 stencilTable->GetDuvWeightsBuffer(),\n- 673 stencilTable->GetDvvWeightsBuffer(),\n- 674 0,\n- 675 stencilTable->GetNumStencils(),\n- 676 numStartEvents, startEvents, endEvent);\n- 677 }\n- 678\n-681 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,\n- 682 cl_mem dst, BufferDescriptor const &dstDesc,\n- 683 cl_mem sizes,\n- 684 cl_mem offsets,\n- 685 cl_mem indices,\n- 686 cl_mem weights,\n- 687 int start,\n- 688 int end,\n- 689 unsigned int numStartEvents=0,\n- 690 const cl_event* startEvents=NULL,\n- 691 cl_event* endEvent=NULL) const;\n- 692\n-741 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,\n- 742 cl_mem dst, BufferDescriptor const &dstDesc,\n- 743 cl_mem du, BufferDescriptor const &duDesc,\n- 744 cl_mem dv, BufferDescriptor const &dvDesc,\n- 745 cl_mem sizes,\n- 746 cl_mem offsets,\n- 747 cl_mem indices,\n- 748 cl_mem weights,\n- 749 cl_mem duWeights,\n- 750 cl_mem dvWeights,\n- 751 int start,\n- 752 int end,\n- 753 unsigned int numStartEvents=0,\n- 754 const cl_event* startEvents=NULL,\n- 755 cl_event* endEvent=NULL) const;\n- 756\n-823 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,\n- 824 cl_mem dst, BufferDescriptor const &dstDesc,\n- 825 cl_mem du, BufferDescriptor const &duDesc,\n- 826 cl_mem dv, BufferDescriptor const &dvDesc,\n- 827 cl_mem duu, BufferDescriptor const &duuDesc,\n- 828 cl_mem duv, BufferDescriptor const &duvDesc,\n- 829 cl_mem dvv, BufferDescriptor const &dvvDesc,\n- 830 cl_mem sizes,\n- 831 cl_mem offsets,\n- 832 cl_mem indices,\n- 833 cl_mem weights,\n- 834 cl_mem duWeights,\n- 835 cl_mem dvWeights,\n- 836 cl_mem duuWeights,\n- 837 cl_mem duvWeights,\n- 838 cl_mem dvvWeights,\n- 839 int start,\n- 840 int end,\n- 841 unsigned int numStartEvents=0,\n- 842 const cl_event* startEvents=NULL,\n- 843 cl_event* endEvent=NULL) const;\n- 844\n- 850\n- 899 template \n-902 static bool EvalPatches(\n- 903 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 904 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 905 int numPatchCoords,\n- 906 PATCHCOORD_BUFFER *patchCoords,\n- 907 PATCH_TABLE *patchTable,\n- 908 CLEvaluator const *instance,\n- 909 DEVICE_CONTEXT deviceContext,\n- 910 unsigned int numStartEvents=0,\n- 911 const cl_event* startEvents=NULL,\n- 912 cl_event* endEvent=NULL) {\n- 913\n- 914 if (instance) {\n- 915 return instance->EvalPatches(srcBuffer, srcDesc,\n- 916 dstBuffer, dstDesc,\n- 917 numPatchCoords, patchCoords,\n- 918 patchTable,\n- 919 numStartEvents, startEvents, endEvent);\n- 920 } else {\n- 921 // Create an instance on demand (slow)\n- 922 (void)deviceContext; // unused\n- 923 instance = Create(srcDesc, dstDesc,\n- 924 BufferDescriptor(),\n- 925 BufferDescriptor(),\n- 926 deviceContext);\n- 927 if (instance) {\n- 928 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 929 dstBuffer, dstDesc,\n- 930 numPatchCoords, patchCoords,\n- 931 patchTable,\n- 932 numStartEvents, startEvents, endEvent);\n- 933 delete instance;\n- 934 return r;\n- 935 }\n- 936 return false;\n- 937 }\n- 938 }\n- 939\n- 1000 template \n-1003 static bool EvalPatches(\n- 1004 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1005 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1006 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1007 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1008 int numPatchCoords,\n- 1009 PATCHCOORD_BUFFER *patchCoords,\n- 1010 PATCH_TABLE *patchTable,\n- 1011 CLEvaluator const *instance,\n- 1012 DEVICE_CONTEXT deviceContext,\n- 1013 unsigned int numStartEvents=0,\n- 1014 const cl_event* startEvents=NULL,\n- 1015 cl_event* endEvent=NULL) {\n- 1016\n- 1017 if (instance) {\n- 1018 return instance->EvalPatches(srcBuffer, srcDesc,\n- 1019 dstBuffer, dstDesc,\n- 1020 duBuffer, duDesc,\n- 1021 dvBuffer, dvDesc,\n- 1022 numPatchCoords, patchCoords,\n- 1023 patchTable,\n- 1024 numStartEvents, startEvents, endEvent);\n- 1025 } else {\n- 1026 // Create an instance on demand (slow)\n- 1027 (void)deviceContext; // unused\n- 1028 instance = Create(srcDesc, dstDesc, duDesc, dvDesc, deviceContext);\n- 1029 if (instance) {\n- 1030 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 1031 dstBuffer, dstDesc,\n- 1032 duBuffer, duDesc,\n- 1033 dvBuffer, dvDesc,\n- 1034 numPatchCoords, patchCoords,\n- 1035 patchTable,\n- 1036 numStartEvents, startEvents, endEvent);\n- 1037 delete instance;\n- 1038 return r;\n- 1039 }\n- 1040 return false;\n- 1041 }\n- 1042 }\n- 1043\n- 1122 template \n-1125 static bool EvalPatches(\n- 1126 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1127 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1128 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1129 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1130 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1131 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1132 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1133 int numPatchCoords,\n- 1134 PATCHCOORD_BUFFER *patchCoords,\n- 1135 PATCH_TABLE *patchTable,\n- 1136 CLEvaluator const *instance,\n- 1137 DEVICE_CONTEXT deviceContext,\n- 1138 unsigned int numStartEvents=0,\n- 1139 const cl_event* startEvents=NULL,\n- 1140 cl_event* endEvent=NULL) {\n- 1141\n- 1142 if (instance) {\n- 1143 return instance->EvalPatches(srcBuffer, srcDesc,\n- 1144 dstBuffer, dstDesc,\n- 1145 duBuffer, duDesc,\n- 1146 dvBuffer, dvDesc,\n- 1147 duuBuffer, duuDesc,\n- 1148 duvBuffer, duvDesc,\n- 1149 dvvBuffer, dvvDesc,\n- 1150 numPatchCoords, patchCoords,\n- 1151 patchTable,\n- 1152 numStartEvents, startEvents, endEvent);\n- 1153 } else {\n- 1154 // Create an instance on demand (slow)\n- 1155 (void)deviceContext; // unused\n- 1156 instance = Create(srcDesc, dstDesc,\n- 1157 duDesc, dvDesc,\n- 1158 duuDesc, duvDesc, dvvDesc,\n- 1159 deviceContext);\n- 1160 if (instance) {\n- 1161 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 1162 dstBuffer, dstDesc,\n- 1163 duBuffer, duDesc,\n- 1164 dvBuffer, dvDesc,\n- 1165 duuBuffer, duuDesc,\n- 1166 duvBuffer, duvDesc,\n- 1167 dvvBuffer, dvvDesc,\n- 1168 numPatchCoords, patchCoords,\n- 1169 patchTable,\n- 1170 numStartEvents, startEvents, endEvent);\n- 1171 delete instance;\n- 1172 return r;\n- 1173 }\n- 1174 return false;\n- 1175 }\n- 1176 }\n- 1177\n- 1215 template \n-1217 bool EvalPatches(\n- 1218 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1219 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1220 int numPatchCoords,\n- 1221 PATCHCOORD_BUFFER *patchCoords,\n- 1222 PATCH_TABLE *patchTable,\n- 1223 unsigned int numStartEvents=0,\n- 1224 const cl_event* startEvents=NULL,\n- 1225 cl_event* endEvent=NULL) const {\n- 1226\n- 1227 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 1228 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 1229 0, BufferDescriptor(),\n- 1230 0, BufferDescriptor(),\n- 1231 numPatchCoords,\n- 1232 patchCoords->BindCLBuffer(_clCommandQueue),\n- 1233 patchTable->GetPatchArrayBuffer(),\n- 1234 patchTable->GetPatchIndexBuffer(),\n- 1235 patchTable->GetPatchParamBuffer(),\n- 1236 numStartEvents, startEvents, endEvent);\n- 1237 }\n- 1238\n- 1286 template \n-1288 bool EvalPatches(\n- 1289 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1290 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1291 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1292 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1293 int numPatchCoords,\n- 1294 PATCHCOORD_BUFFER *patchCoords,\n- 1295 PATCH_TABLE *patchTable,\n- 1296 unsigned int numStartEvents=0,\n- 1297 const cl_event* startEvents=NULL,\n- 1298 cl_event* endEvent=NULL) const {\n- 1299\n- 1300 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 1301 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 1302 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n- 1303 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n- 1304 numPatchCoords,\n- 1305 patchCoords->BindCLBuffer(_clCommandQueue),\n- 1306 patchTable->GetPatchArrayBuffer(),\n- 1307 patchTable->GetPatchIndexBuffer(),\n- 1308 patchTable->GetPatchParamBuffer(),\n- 1309 numStartEvents, startEvents, endEvent);\n- 1310 }\n- 1311\n- 1377 template \n-1379 bool EvalPatches(\n- 1380 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1381 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1382 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1383 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1384 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1385 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1386 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1387 int numPatchCoords,\n- 1388 PATCHCOORD_BUFFER *patchCoords,\n- 1389 PATCH_TABLE *patchTable,\n- 1390 unsigned int numStartEvents=0,\n- 1391 const cl_event* startEvents=NULL,\n- 1392 cl_event* endEvent=NULL) const {\n- 1393\n- 1394 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 1395 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 1396 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n- 1397 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n- 1398 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,\n- 1399 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,\n- 1400 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,\n- 1401 numPatchCoords,\n- 1402 patchCoords->BindCLBuffer(_clCommandQueue),\n- 1403 patchTable->GetPatchArrayBuffer(),\n- 1404 patchTable->GetPatchIndexBuffer(),\n- 1405 patchTable->GetPatchParamBuffer(),\n- 1406 numStartEvents, startEvents, endEvent);\n- 1407 }\n- 1408\n-1409 bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc,\n- 1410 cl_mem dst, BufferDescriptor const &dstDesc,\n- 1411 cl_mem du, BufferDescriptor const &duDesc,\n- 1412 cl_mem dv, BufferDescriptor const &dvDesc,\n- 1413 int numPatchCoords,\n- 1414 cl_mem patchCoordsBuffer,\n- 1415 cl_mem patchArrayBuffer,\n- 1416 cl_mem patchIndexBuffer,\n- 1417 cl_mem patchParamsBuffer,\n- 1418 unsigned int numStartEvents=0,\n- 1419 const cl_event* startEvents=NULL,\n- 1420 cl_event* endEvent=NULL) const;\n- 1421\n-1422 bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc,\n- 1423 cl_mem dst, BufferDescriptor const &dstDesc,\n- 1424 cl_mem du, BufferDescriptor const &duDesc,\n- 1425 cl_mem dv, BufferDescriptor const &dvDesc,\n- 1426 cl_mem duu, BufferDescriptor const &duuDesc,\n- 1427 cl_mem duv, BufferDescriptor const &duvDesc,\n- 1428 cl_mem dvv, BufferDescriptor const &dvvDesc,\n- 1429 int numPatchCoords,\n- 1430 cl_mem patchCoordsBuffer,\n- 1431 cl_mem patchArrayBuffer,\n- 1432 cl_mem patchIndexBuffer,\n- 1433 cl_mem patchParamsBuffer,\n- 1434 unsigned int numStartEvents=0,\n- 1435 const cl_event* startEvents=NULL,\n- 1436 cl_event* endEvent=NULL) const;\n- 1437\n- 1486 template \n-1489 static bool EvalPatchesVarying(\n- 1490 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1491 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1492 int numPatchCoords,\n- 1493 PATCHCOORD_BUFFER *patchCoords,\n- 1494 PATCH_TABLE *patchTable,\n- 1495 CLEvaluator const *instance,\n- 1496 DEVICE_CONTEXT deviceContext,\n- 1497 unsigned int numStartEvents=0,\n- 1498 const cl_event* startEvents=NULL,\n- 1499 cl_event* endEvent=NULL) {\n- 1500\n- 1501 if (instance) {\n- 1502 return instance->EvalPatchesVarying(\n- 1503 srcBuffer, srcDesc,\n- 1504 dstBuffer, dstDesc,\n- 1505 numPatchCoords, patchCoords,\n- 1506 patchTable,\n- 1507 numStartEvents, startEvents, endEvent);\n- 1508 } else {\n- 1509 // Create an instance on demand (slow)\n- 1510 (void)deviceContext; // unused\n- 1511 instance = Create(srcDesc, dstDesc,\n- 1512 BufferDescriptor(),\n- 1513 BufferDescriptor(),\n- 1514 deviceContext);\n- 1515 if (instance) {\n- 1516 bool r = instance->EvalPatchesVarying(\n- 1517 srcBuffer, srcDesc,\n- 1518 dstBuffer, dstDesc,\n- 1519 numPatchCoords, patchCoords,\n- 1520 patchTable,\n- 1521 numStartEvents, startEvents, endEvent);\n- 1522 delete instance;\n- 1523 return r;\n- 1524 }\n- 1525 return false;\n- 1526 }\n- 1527 }\n- 1528\n- 1566 template \n-1568 bool EvalPatchesVarying(\n- 1569 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1570 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1571 int numPatchCoords,\n- 1572 PATCHCOORD_BUFFER *patchCoords,\n- 1573 PATCH_TABLE *patchTable,\n- 1574 unsigned int numStartEvents=0,\n- 1575 const cl_event* startEvents=NULL,\n- 1576 cl_event* endEvent=NULL) const {\n- 1577\n- 1578 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 1579 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 1580 0, BufferDescriptor(),\n- 1581 0, BufferDescriptor(),\n- 1582 numPatchCoords,\n- 1583 patchCoords->BindCLBuffer(_clCommandQueue),\n- 1584 patchTable->GetVaryingPatchArrayBuffer(),\n- 1585 patchTable->GetVaryingPatchIndexBuffer(),\n- 1586 patchTable->GetPatchParamBuffer(),\n- 1587 numStartEvents, startEvents, endEvent);\n- 1588 }\n- 1589\n- 1650 template \n-1653 static bool EvalPatchesVarying(\n- 1654 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1655 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1656 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1657 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1658 int numPatchCoords,\n- 1659 PATCHCOORD_BUFFER *patchCoords,\n- 1660 PATCH_TABLE *patchTable,\n- 1661 CLEvaluator const *instance,\n- 1662 DEVICE_CONTEXT deviceContext,\n- 1663 unsigned int numStartEvents=0,\n- 1664 const cl_event* startEvents=NULL,\n- 1665 cl_event* endEvent=NULL) {\n- 1666\n- 1667 if (instance) {\n- 1668 return instance->EvalPatchesVarying(\n- 1669 srcBuffer, srcDesc,\n- 1670 dstBuffer, dstDesc,\n- 1671 duBuffer, duDesc,\n- 1672 dvBuffer, dvDesc,\n- 1673 numPatchCoords, patchCoords,\n- 1674 patchTable,\n- 1675 numStartEvents, startEvents, endEvent);\n- 1676 } else {\n- 1677 // Create an instance on demand (slow)\n- 1678 (void)deviceContext; // unused\n- 1679 instance = Create(srcDesc, dstDesc,\n- 1680 duDesc, dvDesc,\n- 1681 deviceContext);\n- 1682 if (instance) {\n- 1683 bool r = instance->EvalPatchesVarying(\n- 1684 srcBuffer, srcDesc,\n- 1685 dstBuffer, dstDesc,\n- 1686 duBuffer, duDesc,\n- 1687 dvBuffer, dvDesc,\n- 1688 numPatchCoords, patchCoords,\n- 1689 patchTable,\n- 1690 numStartEvents, startEvents, endEvent);\n- 1691 delete instance;\n- 1692 return r;\n- 1693 }\n- 1694 return false;\n- 1695 }\n- 1696 }\n- 1697\n- 1747 template \n-1749 bool EvalPatchesVarying(\n- 1750 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1751 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1752 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1753 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1754 int numPatchCoords,\n- 1755 PATCHCOORD_BUFFER *patchCoords,\n- 1756 PATCH_TABLE *patchTable,\n- 1757 unsigned int numStartEvents=0,\n- 1758 const cl_event* startEvents=NULL,\n- 1759 cl_event* endEvent=NULL) const {\n- 1760\n- 1761 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 1762 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 1763 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n- 1764 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n- 1765 numPatchCoords,\n- 1766 patchCoords->BindCLBuffer(_clCommandQueue),\n- 1767 patchTable->GetVaryingPatchArrayBuffer(),\n- 1768 patchTable->GetVaryingPatchIndexBuffer(),\n- 1769 patchTable->GetPatchParamBuffer(),\n- 1770 numStartEvents, startEvents, endEvent);\n- 1771 }\n- 1772\n- 1851 template \n-1854 static bool EvalPatchesVarying(\n- 1855 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1856 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1857 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1858 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1859 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1860 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1861 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1862 int numPatchCoords,\n- 1863 PATCHCOORD_BUFFER *patchCoords,\n- 1864 PATCH_TABLE *patchTable,\n- 1865 CLEvaluator const *instance,\n- 1866 DEVICE_CONTEXT deviceContext,\n- 1867 unsigned int numStartEvents=0,\n- 1868 const cl_event* startEvents=NULL,\n- 1869 cl_event* endEvent=NULL) {\n- 1870\n- 1871 if (instance) {\n- 1872 return instance->EvalPatchesVarying(\n- 1873 srcBuffer, srcDesc,\n- 1874 dstBuffer, dstDesc,\n- 1875 duBuffer, duDesc,\n- 1876 dvBuffer, dvDesc,\n- 1877 duuBuffer, duuDesc,\n- 1878 duvBuffer, duvDesc,\n- 1879 dvvBuffer, dvvDesc,\n- 1880 numPatchCoords, patchCoords,\n- 1881 patchTable,\n- 1882 numStartEvents, startEvents, endEvent);\n- 1883 } else {\n- 1884 // Create an instance on demand (slow)\n- 1885 (void)deviceContext; // unused\n- 1886 instance = Create(srcDesc, dstDesc,\n- 1887 duDesc, dvDesc,\n- 1888 duuDesc, duvDesc, dvvDesc,\n- 1889 deviceContext);\n- 1890 if (instance) {\n- 1891 bool r = instance->EvalPatchesVarying(\n- 1892 srcBuffer, srcDesc,\n- 1893 dstBuffer, dstDesc,\n- 1894 duBuffer, duDesc,\n- 1895 dvBuffer, dvDesc,\n- 1896 duuBuffer, duuDesc,\n- 1897 duvBuffer, duvDesc,\n- 1898 dvvBuffer, dvvDesc,\n- 1899 numPatchCoords, patchCoords,\n- 1900 patchTable,\n- 1901 numStartEvents, startEvents, endEvent);\n- 1902 delete instance;\n- 1903 return r;\n- 1904 }\n- 1905 return false;\n- 1906 }\n- 1907 }\n- 1908\n- 1976 template \n-1978 bool EvalPatchesVarying(\n- 1979 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1980 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1981 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1982 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1983 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1984 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1985 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1986 int numPatchCoords,\n- 1987 PATCHCOORD_BUFFER *patchCoords,\n- 1988 PATCH_TABLE *patchTable,\n- 1989 unsigned int numStartEvents=0,\n- 1990 const cl_event* startEvents=NULL,\n- 1991 cl_event* endEvent=NULL) const {\n- 1992\n- 1993 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 1994 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 1995 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n- 1996 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n- 1997 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,\n- 1998 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,\n- 1999 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,\n- 2000 numPatchCoords,\n- 2001 patchCoords->BindCLBuffer(_clCommandQueue),\n- 2002 patchTable->GetVaryingPatchArrayBuffer(),\n- 2003 patchTable->GetVaryingPatchIndexBuffer(),\n- 2004 patchTable->GetPatchParamBuffer(),\n- 2005 numStartEvents, startEvents, endEvent);\n- 2006 }\n- 2007\n- 2058 template \n-2061 static bool EvalPatchesFaceVarying(\n- 2062 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 2063 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 2064 int numPatchCoords,\n- 2065 PATCHCOORD_BUFFER *patchCoords,\n- 2066 PATCH_TABLE *patchTable,\n- 2067 int fvarChannel,\n- 2068 CLEvaluator const *instance,\n- 2069 DEVICE_CONTEXT deviceContext,\n- 2070 unsigned int numStartEvents=0,\n- 2071 const cl_event* startEvents=NULL,\n- 2072 cl_event* endEvent=NULL) {\n- 2073\n- 2074 if (instance) {\n- 2075 return instance->EvalPatchesFaceVarying(\n- 2076 srcBuffer, srcDesc,\n- 2077 dstBuffer, dstDesc,\n- 2078 numPatchCoords, patchCoords,\n- 2079 patchTable, fvarChannel,\n- 2080 numStartEvents, startEvents, endEvent);\n- 2081 } else {\n- 2082 // Create an instance on demand (slow)\n- 2083 (void)deviceContext; // unused\n- 2084 instance = Create(srcDesc, dstDesc,\n- 2085 BufferDescriptor(),\n- 2086 BufferDescriptor(),\n- 2087 deviceContext);\n- 2088 if (instance) {\n- 2089 bool r = instance->EvalPatchesFaceVarying(\n- 2090 srcBuffer, srcDesc,\n- 2091 dstBuffer, dstDesc,\n- 2092 numPatchCoords, patchCoords,\n- 2093 patchTable, fvarChannel,\n- 2094 numStartEvents, startEvents, endEvent);\n- 2095 delete instance;\n- 2096 return r;\n- 2097 }\n- 2098 return false;\n- 2099 }\n- 2100 }\n- 2101\n- 2141 template \n-2143 bool EvalPatchesFaceVarying(\n- 2144 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 2145 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 2146 int numPatchCoords,\n- 2147 PATCHCOORD_BUFFER *patchCoords,\n- 2148 PATCH_TABLE *patchTable,\n- 2149 int fvarChannel = 0,\n- 2150 unsigned int numStartEvents=0,\n- 2151 const cl_event* startEvents=NULL,\n- 2152 cl_event* endEvent=NULL) const {\n- 2153\n- 2154 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 2155 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 2156 0, BufferDescriptor(),\n- 2157 0, BufferDescriptor(),\n- 2158 numPatchCoords,\n- 2159 patchCoords->BindCLBuffer(_clCommandQueue),\n- 2160 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 2161 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 2162 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n- 2163 numStartEvents, startEvents, endEvent);\n- 2164 }\n- 2165\n- 2228 template \n-2231 static bool EvalPatchesFaceVarying(\n- 2232 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 2233 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 2234 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 2235 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 2236 int numPatchCoords,\n- 2237 PATCHCOORD_BUFFER *patchCoords,\n- 2238 PATCH_TABLE *patchTable,\n- 2239 int fvarChannel,\n- 2240 CLEvaluator const *instance,\n- 2241 DEVICE_CONTEXT deviceContext,\n- 2242 unsigned int numStartEvents=0,\n- 2243 const cl_event* startEvents=NULL,\n- 2244 cl_event* endEvent=NULL) {\n- 2245\n- 2246 if (instance) {\n- 2247 return instance->EvalPatchesFaceVarying(\n- 2248 srcBuffer, srcDesc,\n- 2249 dstBuffer, dstDesc,\n- 2250 duBuffer, duDesc,\n- 2251 dvBuffer, dvDesc,\n- 2252 numPatchCoords, patchCoords,\n- 2253 patchTable, fvarChannel,\n- 2254 numStartEvents, startEvents, endEvent);\n- 2255 } else {\n- 2256 // Create an instance on demand (slow)\n- 2257 (void)deviceContext; // unused\n- 2258 instance = Create(srcDesc, dstDesc,\n- 2259 duDesc, dvDesc, deviceContext);\n- 2260 if (instance) {\n- 2261 bool r = instance->EvalPatchesFaceVarying(\n- 2262 srcBuffer, srcDesc,\n- 2263 dstBuffer, dstDesc,\n- 2264 duBuffer, duDesc,\n- 2265 dvBuffer, dvDesc,\n- 2266 numPatchCoords, patchCoords,\n- 2267 patchTable, fvarChannel,\n- 2268 numStartEvents, startEvents, endEvent);\n- 2269 delete instance;\n- 2270 return r;\n- 2271 }\n- 2272 return false;\n- 2273 }\n- 2274 }\n- 2275\n- 2327 template \n-2329 bool EvalPatchesFaceVarying(\n- 2330 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 2331 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 2332 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 2333 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 2334 int numPatchCoords,\n- 2335 PATCHCOORD_BUFFER *patchCoords,\n- 2336 PATCH_TABLE *patchTable,\n- 2337 int fvarChannel = 0,\n- 2338 unsigned int numStartEvents=0,\n- 2339 const cl_event* startEvents=NULL,\n- 2340 cl_event* endEvent=NULL) const {\n- 2341\n- 2342 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 2343 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 2344 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n- 2345 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n- 2346 numPatchCoords,\n- 2347 patchCoords->BindCLBuffer(_clCommandQueue),\n- 2348 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 2349 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 2350 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n- 2351 numStartEvents, startEvents, endEvent);\n- 2352 }\n- 2353\n- 2434 template \n-2437 static bool EvalPatchesFaceVarying(\n- 2438 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 2439 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 2440 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 2441 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 2442 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 2443 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 2444 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 2445 int numPatchCoords,\n- 2446 PATCHCOORD_BUFFER *patchCoords,\n- 2447 PATCH_TABLE *patchTable,\n- 2448 int fvarChannel,\n- 2449 CLEvaluator const *instance,\n- 2450 DEVICE_CONTEXT deviceContext,\n- 2451 unsigned int numStartEvents=0,\n- 2452 const cl_event* startEvents=NULL,\n- 2453 cl_event* endEvent=NULL) {\n- 2454\n- 2455 if (instance) {\n- 2456 return instance->EvalPatchesFaceVarying(\n- 2457 srcBuffer, srcDesc,\n- 2458 dstBuffer, dstDesc,\n- 2459 duBuffer, duDesc,\n- 2460 dvBuffer, dvDesc,\n- 2461 duuBuffer, duuDesc,\n- 2462 duvBuffer, duvDesc,\n- 2463 dvvBuffer, dvvDesc,\n- 2464 numPatchCoords, patchCoords,\n- 2465 patchTable, fvarChannel,\n- 2466 numStartEvents, startEvents, endEvent);\n- 2467 } else {\n- 2468 // Create an instance on demand (slow)\n- 2469 (void)deviceContext; // unused\n- 2470 instance = Create(srcDesc, dstDesc,\n- 2471 duDesc, dvDesc,\n- 2472 duuDesc, duvDesc, dvvDesc,\n- 2473 deviceContext);\n- 2474 if (instance) {\n- 2475 bool r = instance->EvalPatchesFaceVarying(\n- 2476 srcBuffer, srcDesc,\n- 2477 dstBuffer, dstDesc,\n- 2478 duBuffer, duDesc,\n- 2479 dvBuffer, dvDesc,\n- 2480 duuBuffer, duuDesc,\n- 2481 duvBuffer, duvDesc,\n- 2482 dvvBuffer, dvvDesc,\n- 2483 numPatchCoords, patchCoords,\n- 2484 patchTable, fvarChannel,\n- 2485 numStartEvents, startEvents, endEvent);\n- 2486 delete instance;\n- 2487 return r;\n- 2488 }\n- 2489 return false;\n- 2490 }\n- 2491 }\n- 2492\n- 2562 template \n-2564 bool EvalPatchesFaceVarying(\n- 2565 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 2566 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 2567 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 2568 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 2569 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 2570 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 2571 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 2572 int numPatchCoords,\n- 2573 PATCHCOORD_BUFFER *patchCoords,\n- 2574 PATCH_TABLE *patchTable,\n- 2575 int fvarChannel = 0,\n- 2576 unsigned int numStartEvents=0,\n- 2577 const cl_event* startEvents=NULL,\n- 2578 cl_event* endEvent=NULL) const {\n- 2579\n- 2580 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n- 2581 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n- 2582 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n- 2583 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n- 2584 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,\n- 2585 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,\n- 2586 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,\n- 2587 numPatchCoords,\n- 2588 patchCoords->BindCLBuffer(_clCommandQueue),\n- 2589 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 2590 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 2591 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n- 2592 numStartEvents, startEvents, endEvent);\n- 2593 }\n- 2594\n- 2600\n-2603 bool Compile(BufferDescriptor const &srcDesc,\n- 2604 BufferDescriptor const &dstDesc,\n- 2605 BufferDescriptor const &duDesc = BufferDescriptor(),\n- 2606 BufferDescriptor const &dvDesc = BufferDescriptor(),\n- 2607 BufferDescriptor const &duuDesc = BufferDescriptor(),\n- 2608 BufferDescriptor const &duvDesc = BufferDescriptor(),\n- 2609 BufferDescriptor const &dvvDesc = BufferDescriptor());\n- 2610\n- 2612 template \n-2613 static void Synchronize(DEVICE_CONTEXT deviceContext) {\n- 2614 Synchronize(deviceContext->GetCommandQueue());\n- 2615 }\n- 2616\n-2617 static void Synchronize(cl_command_queue queue);\n- 2618\n- 2619private:\n- 2620 cl_context _clContext;\n- 2621 cl_command_queue _clCommandQueue;\n- 2622 cl_program _program;\n- 2623 cl_kernel _stencilKernel;\n- 2624 cl_kernel _stencilDerivKernel;\n- 2625 cl_kernel _patchKernel;\n- 2626};\n- 2627\n- 2628} // end namespace Osd\n- 2629\n- 2630} // end namespace OPENSUBDIV_VERSION\n- 2631using namespace OPENSUBDIV_VERSION;\n- 2632\n- 2633} // end namespace OpenSubdiv\n- 2634\n- 2635\n- 2636#endif // OPENSUBDIV3_OSD_CL_EVALUATOR_H\n+86 cl_mem GetFVarPatchArrayBuffer(int fvarChannel = 0) const { return\n+_fvarPatchArrays[fvarChannel]; }\n+ 87\n+89 cl_mem GetFVarPatchIndexBuffer(int fvarChannel = 0) const { return\n+_fvarIndexBuffers[fvarChannel]; }\n+ 90\n+92 cl_mem GetFVarPatchParamBuffer(int fvarChannel = 0) const { return\n+_fvarParamBuffers[fvarChannel]; }\n+ 93\n+ 94protected:\n+95 CLPatchTable();\n+ 96\n+97 bool allocate(Far::PatchTable const *patchTable, cl_context clContext);\n+ 98\n+99 cl_mem _patchArrays;\n+100 cl_mem _indexBuffer;\n+101 cl_mem _patchParamBuffer;\n+ 102\n+103 cl_mem _varyingPatchArrays;\n+104 cl_mem _varyingIndexBuffer;\n+ 105\n+106 std::vector _fvarPatchArrays;\n+107 std::vector _fvarIndexBuffers;\n+108 std::vector _fvarParamBuffers;\n+ 109\n+ 110};\n+ 111\n+ 112} // end namespace Osd\n+ 113\n+ 114} // end namespace OPENSUBDIV_VERSION\n+ 115using namespace OPENSUBDIV_VERSION;\n+ 116\n+ 117} // end namespace OpenSubdiv\n+ 118\n+ 119#endif // OPENSUBDIV3_OSD_CL_PATCH_TABLE_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n-Stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:273\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n-Limit stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:583\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n-BufferDescriptor is a struct which describes buffer elements in interleaved\n-data buffers....\n-Definition: bufferDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable\n-OpenCL stencil table.\n-Definition: clEvaluator.h:52\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetDvvWeightsBuffer\n-cl_mem GetDvvWeightsBuffer() const\n-Definition: clEvaluator.h:82\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::CLStencilTable\n-CLStencilTable(Far::StencilTable const *stencilTable, cl_context clContext)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetOffsetsBuffer\n-cl_mem GetOffsetsBuffer() const\n-Definition: clEvaluator.h:75\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::Create\n-static CLStencilTable * Create(Far::LimitStencilTable const *limitStencilTable,\n-DEVICE_CONTEXT context)\n-Definition: clEvaluator.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetIndicesBuffer\n-cl_mem GetIndicesBuffer() const\n-Definition: clEvaluator.h:76\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetNumStencils\n-int GetNumStencils() const\n-Definition: clEvaluator.h:83\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetSizesBuffer\n-cl_mem GetSizesBuffer() const\n-Definition: clEvaluator.h:74\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetDuuWeightsBuffer\n-cl_mem GetDuuWeightsBuffer() const\n-Definition: clEvaluator.h:80\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetDvWeightsBuffer\n-cl_mem GetDvWeightsBuffer() const\n-Definition: clEvaluator.h:79\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetDuvWeightsBuffer\n-cl_mem GetDuvWeightsBuffer() const\n-Definition: clEvaluator.h:81\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetDuWeightsBuffer\n-cl_mem GetDuWeightsBuffer() const\n-Definition: clEvaluator.h:78\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::~CLStencilTable\n-~CLStencilTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::Create\n-static CLStencilTable * Create(Far::StencilTable const *stencilTable,\n-DEVICE_CONTEXT context)\n-Definition: clEvaluator.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::CLStencilTable\n-CLStencilTable(Far::LimitStencilTable const *limitStencilTable, cl_context\n-clContext)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetWeightsBuffer\n-cl_mem GetWeightsBuffer() const\n-Definition: clEvaluator.h:77\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator\n-Definition: clEvaluator.h:100\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, CLEvaluator\n-const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0,\n-const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n-Generic static stencil function. This function has a same signature as other\n-device kernels have so t...\n-Definition: clEvaluator.h:309\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Synchronize\n-static void Synchronize(cl_command_queue queue)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::~CLEvaluator\n-~CLEvaluator()\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n+Container for arrays of parametric patches.\n+Definition: patchTable.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable\n+CL patch table.\n+Definition: clPatchTable.h:52\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_fvarPatchArrays\n+std::vector< cl_mem > _fvarPatchArrays\n+Definition: clPatchTable.h:106\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::~CLPatchTable\n+~CLPatchTable()\n Destructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n-bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst,\n-BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc,\n-cl_mem dv, BufferDescriptor const &dvDesc, cl_mem sizes, cl_mem offsets, cl_mem\n-indices, cl_mem weights, cl_mem duWeights, cl_mem dvWeights, int start, int\n-end, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n-*endEvent=NULL) const\n-Dispatch the CL compute kernel asynchronously. returns false if the kernel\n-hasn't been compiled yet.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n-const *stencilTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext,\n-unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n-*endEvent=NULL)\n-Generic static stencil function. This function has a same signature as other\n-device kernels have so t...\n-Definition: clEvaluator.h:219\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create\n-static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor\n-const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc,\n-cl_context clContext, cl_command_queue clCommandQueue)\n-Definition: clEvaluator.h:116\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CLEvaluator const\n-*instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const\n-cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:2231\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n-bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst,\n-BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc,\n-cl_mem dv, BufferDescriptor const &dvDesc, int numPatchCoords, cl_mem\n-patchCoordsBuffer, cl_mem patchArrayBuffer, cl_mem patchIndexBuffer, cl_mem\n-patchParamsBuffer, unsigned int numStartEvents=0, const cl_event\n-*startEvents=NULL, cl_event *endEvent=NULL) const\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n-bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const\n-cl_event *startEvents=NULL, cl_event *endEvent=NULL) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:1978\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n-bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, STENCIL_TABLE const *stencilTable, unsigned int numStartEvents=0,\n-const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const\n-Generic stencil function.\n-Definition: clEvaluator.h:561\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n-bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const\n-cl_event *startEvents=NULL, cl_event *endEvent=NULL) const\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: clEvaluator.h:1379\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Instantiatable\n-bool Instantiatable\n-Definition: clEvaluator.h:102\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create\n-static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor\n-const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc,\n-DEVICE_CONTEXT deviceContext)\n-Generic creator template.\n-Definition: clEvaluator.h:106\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n-fvarChannel, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext,\n-unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n-*endEvent=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:2061\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create\n-static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor\n-const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc,\n-BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc,\n-BufferDescriptor const &dvvDesc, DEVICE_CONTEXT deviceContext)\n-Generic creator template.\n-Definition: clEvaluator.h:131\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n-bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst,\n-BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc,\n-cl_mem dv, BufferDescriptor const &dvDesc, cl_mem duu, BufferDescriptor const\n-&duuDesc, cl_mem duv, BufferDescriptor const &duvDesc, cl_mem dvv,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, cl_mem patchCoordsBuffer,\n-cl_mem patchArrayBuffer, cl_mem patchIndexBuffer, cl_mem patchParamsBuffer,\n-unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n-*endEvent=NULL) const\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n-bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst,\n-BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc,\n-cl_mem dv, BufferDescriptor const &dvDesc, cl_mem duu, BufferDescriptor const\n-&duuDesc, cl_mem duv, BufferDescriptor const &duvDesc, cl_mem dvv,\n-BufferDescriptor const &dvvDesc, cl_mem sizes, cl_mem offsets, cl_mem indices,\n-cl_mem weights, cl_mem duWeights, cl_mem dvWeights, cl_mem duuWeights, cl_mem\n-duvWeights, cl_mem dvvWeights, int start, int end, unsigned int\n-numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n-const\n-Dispatch the CL compute kernel asynchronously. returns false if the kernel\n-hasn't been compiled yet.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n-bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0,\n-unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n-*endEvent=NULL) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:2564\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n-bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int\n-numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n-const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:1217\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n-bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, unsigned\n-int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n-*endEvent=NULL) const\n-Generic stencil function.\n-Definition: clEvaluator.h:646\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance,\n-DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event\n-*startEvents=NULL, cl_event *endEvent=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:1653\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Synchronize\n-static void Synchronize(DEVICE_CONTEXT deviceContext)\n-Wait the OpenCL kernels finish.\n-Definition: clEvaluator.h:2613\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance,\n-DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event\n-*startEvents=NULL, cl_event *endEvent=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:1125\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n-CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int\n-numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n-Generic static stencil function. This function has a same signature as other\n-device kernels have so t...\n-Definition: clEvaluator.h:421\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n-bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst,\n-BufferDescriptor const &dstDesc, cl_mem sizes, cl_mem offsets, cl_mem indices,\n-cl_mem weights, int start, int end, unsigned int numStartEvents=0, const\n-cl_event *startEvents=NULL, cl_event *endEvent=NULL) const\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n-CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int\n-numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:1489\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n-bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0, unsigned int\n-numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n-const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:2329\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n-CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int\n-numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:2437\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create\n-static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor\n-const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc,\n-BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc,\n-BufferDescriptor const &dvvDesc, cl_context clContext, cl_command_queue\n-clCommandQueue)\n-Definition: clEvaluator.h:145\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext,\n-unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n-*endEvent=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:1003\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n-bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL,\n-cl_event *endEvent=NULL) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:1749\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const\n-*instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const\n-cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:902\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n-bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int\n-numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n-const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:1568\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const\n-*instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const\n-cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:1854\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n-bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const\n-*stencilTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL,\n-cl_event *endEvent=NULL) const\n-Generic stencil function.\n-Definition: clEvaluator.h:500\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n-bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL,\n-cl_event *endEvent=NULL) const\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: clEvaluator.h:1288\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::CLEvaluator\n-CLEvaluator(cl_context context, cl_command_queue queue)\n-Constructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Compile\n-bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc,\n-BufferDescriptor const &duDesc=BufferDescriptor(), BufferDescriptor const\n-&dvDesc=BufferDescriptor(), BufferDescriptor const &duuDesc=BufferDescriptor(),\n-BufferDescriptor const &duvDesc=BufferDescriptor(), BufferDescriptor const\n-&dvvDesc=BufferDescriptor())\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n-bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n-fvarChannel=0, unsigned int numStartEvents=0, const cl_event *startEvents=NULL,\n-cl_event *endEvent=NULL) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: clEvaluator.h:2143\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_patchParamBuffer\n+cl_mem _patchParamBuffer\n+Definition: clPatchTable.h:101\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchParamBuffer\n+cl_mem GetPatchParamBuffer() const\n+Returns the CL memory of the array of Osd::PatchParam buffer.\n+Definition: clPatchTable.h:74\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_indexBuffer\n+cl_mem _indexBuffer\n+Definition: clPatchTable.h:100\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchArrayBuffer\n+cl_mem GetFVarPatchArrayBuffer(int fvarChannel=0) const\n+Returns the CL memory of the array of Osd::PatchArray buffer.\n+Definition: clPatchTable.h:86\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchArrayBuffer\n+cl_mem GetPatchArrayBuffer() const\n+Returns the CL memory of the array of Osd::PatchArray buffer.\n+Definition: clPatchTable.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::Create\n+static CLPatchTable * Create(Far::PatchTable const *patchTable, cl_context\n+clContext)\n+Creator. Returns NULL if error.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_patchArrays\n+cl_mem _patchArrays\n+Definition: clPatchTable.h:99\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchIndexBuffer\n+cl_mem GetFVarPatchIndexBuffer(int fvarChannel=0) const\n+Returns the CL memory of the face-varying control vertices.\n+Definition: clPatchTable.h:89\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_varyingIndexBuffer\n+cl_mem _varyingIndexBuffer\n+Definition: clPatchTable.h:104\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetVaryingPatchIndexBuffer\n+cl_mem GetVaryingPatchIndexBuffer() const\n+Returns the CL memory of the varying control vertices.\n+Definition: clPatchTable.h:80\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetNumFVarChannels\n+int GetNumFVarChannels() const\n+Returns the number of face-varying channel buffers.\n+Definition: clPatchTable.h:83\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchIndexBuffer\n+cl_mem GetPatchIndexBuffer() const\n+Returns the CL memory of the patch control vertices.\n+Definition: clPatchTable.h:71\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::Create\n+static CLPatchTable * Create(Far::PatchTable const *patchTable, DEVICE_CONTEXT\n+context)\n+Definition: clPatchTable.h:59\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchParamBuffer\n+cl_mem GetFVarPatchParamBuffer(int fvarChannel=0) const\n+Returns the CL memory of the array of Osd::PatchParam buffer.\n+Definition: clPatchTable.h:92\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::CLPatchTable\n+CLPatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::allocate\n+bool allocate(Far::PatchTable const *patchTable, cl_context clContext)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_fvarParamBuffers\n+std::vector< cl_mem > _fvarParamBuffers\n+Definition: clPatchTable.h:108\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_fvarIndexBuffers\n+std::vector< cl_mem > _fvarIndexBuffers\n+Definition: clPatchTable.h:107\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_varyingPatchArrays\n+cl_mem _varyingPatchArrays\n+Definition: clPatchTable.h:103\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetVaryingPatchArrayBuffer\n+cl_mem GetVaryingPatchArrayBuffer() const\n+Returns the CL memory of the array of Osd::PatchArray buffer.\n+Definition: clPatchTable.h:77\n * opensubdiv\n * osd\n- * clEvaluator.h\n+ * clPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00803.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00803.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/clPatchTable.h File Reference\n+OpenSubdiv: opensubdiv/osd/clEvaluator.h File Reference\n \n \n \n \n \n \n \n@@ -83,29 +83,30 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
clPatchTable.h File Reference
\n+
clEvaluator.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../osd/opencl.h"
\n-#include "../osd/nonCopyable.h"
\n+#include "../osd/opencl.h"
\n #include "../osd/types.h"
\n-#include <vector>
\n+#include "../osd/bufferDescriptor.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n+\n+\n+\n+\n \n

\n Classes

class  CLPatchTable
 CL patch table. More...
class  CLStencilTable
 OpenCL stencil table. More...
 
class  CLEvaluator
 
\n \n \n \n \n@@ -116,13 +117,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,31 +5,32 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-clPatchTable.h File Reference\n+clEvaluator.h File Reference\n #include \"../version.h\"\n #include \"../osd/opencl.h\"\n-#include \"../osd/nonCopyable.h\"\n #include \"../osd/types.h\"\n-#include \n+#include \"../osd/bufferDescriptor.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0CLPatchTable\n-\u00a0 CL patch table. More...\n+class \u00a0CLStencilTable\n+\u00a0 OpenCL stencil table. More...\n+\u00a0\n+class \u00a0CLEvaluator\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * clPatchTable.h\n+ * clEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00803.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00803.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,4 @@\n var a00803 = [\n- [\"CLPatchTable\", \"a01145.html\", \"a01145\"]\n+ [\"CLStencilTable\", \"a01137.html\", \"a01137\"],\n+ [\"CLEvaluator\", \"a01141.html\", \"a01141\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00803_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00803_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/clPatchTable.h Source File\n+OpenSubdiv: opensubdiv/osd/clEvaluator.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
clPatchTable.h
\n+
clEvaluator.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,120 +107,1217 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_CL_PATCH_TABLE_H
\n-
26#define OPENSUBDIV3_OSD_CL_PATCH_TABLE_H
\n+
25#ifndef OPENSUBDIV3_OSD_CL_EVALUATOR_H
\n+
26#define OPENSUBDIV3_OSD_CL_EVALUATOR_H
\n
27
\n
28#include "../version.h"
\n
29
\n
30#include "../osd/opencl.h"
\n-
31#include "../osd/nonCopyable.h"
\n-
32#include "../osd/types.h"
\n+
31#include "../osd/types.h"
\n+
32#include "../osd/bufferDescriptor.h"
\n
33
\n-
34#include <vector>
\n-
35
\n-
36namespace OpenSubdiv {
\n-
37namespace OPENSUBDIV_VERSION {
\n-
38
\n-
39namespace Far{
\n-
40 class PatchTable;
\n-
41};
\n+
34namespace OpenSubdiv {
\n+
35namespace OPENSUBDIV_VERSION {
\n+
36
\n+
37namespace Far {
\n+
38 class PatchTable;
\n+
39 class StencilTable;
\n+
40 class LimitStencilTable;
\n+
41}
\n
42
\n
43namespace Osd {
\n
44
\n-
52class CLPatchTable : private NonCopyable<CLPatchTable> {
\n+
52class CLStencilTable {
\n
53public:
\n-
55 static CLPatchTable *Create(Far::PatchTable const *patchTable,
\n-
56 cl_context clContext);
\n-
57
\n-
58 template <typename DEVICE_CONTEXT>
\n-
59 static CLPatchTable * Create(Far::PatchTable const *patchTable,
\n-
60 DEVICE_CONTEXT context) {
\n-
61 return Create(patchTable, context->GetContext());
\n-
62 }
\n-
63
\n-
65 ~CLPatchTable();
\n+
54 template <typename DEVICE_CONTEXT>
\n+
55 static CLStencilTable *Create(Far::StencilTable const *stencilTable,
\n+
56 DEVICE_CONTEXT context) {
\n+
57 return new CLStencilTable(stencilTable, context->GetContext());
\n+
58 }
\n+
59
\n+
60 template <typename DEVICE_CONTEXT>
\n+
61 static CLStencilTable *Create(
\n+
62 Far::LimitStencilTable const *limitStencilTable,
\n+
63 DEVICE_CONTEXT context) {
\n+
64 return new CLStencilTable(limitStencilTable, context->GetContext());
\n+
65 }
\n
66
\n-
68 cl_mem GetPatchArrayBuffer() const { return _patchArrays; }
\n-
69
\n-
71 cl_mem GetPatchIndexBuffer() const { return _indexBuffer; }
\n+
67 CLStencilTable(Far::StencilTable const *stencilTable,
\n+
68 cl_context clContext);
\n+
69 CLStencilTable(Far::LimitStencilTable const *limitStencilTable,
\n+
70 cl_context clContext);
\n+
71 ~CLStencilTable();
\n
72
\n-
74 cl_mem GetPatchParamBuffer() const { return _patchParamBuffer; }
\n-
75
\n-
77 cl_mem GetVaryingPatchArrayBuffer() const { return _varyingPatchArrays; }
\n-
78
\n-
80 cl_mem GetVaryingPatchIndexBuffer() const { return _varyingIndexBuffer; }
\n-
81
\n-
83 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }
\n+
73 // interfaces needed for CLComputeKernel
\n+
74 cl_mem GetSizesBuffer() const { return _sizes; }
\n+
75 cl_mem GetOffsetsBuffer() const { return _offsets; }
\n+
76 cl_mem GetIndicesBuffer() const { return _indices; }
\n+
77 cl_mem GetWeightsBuffer() const { return _weights; }
\n+
78 cl_mem GetDuWeightsBuffer() const { return _duWeights; }
\n+
79 cl_mem GetDvWeightsBuffer() const { return _dvWeights; }
\n+
80 cl_mem GetDuuWeightsBuffer() const { return _duuWeights; }
\n+
81 cl_mem GetDuvWeightsBuffer() const { return _duvWeights; }
\n+
82 cl_mem GetDvvWeightsBuffer() const { return _dvvWeights; }
\n+
83 int GetNumStencils() const { return _numStencils; }
\n
84
\n-
86 cl_mem GetFVarPatchArrayBuffer(int fvarChannel = 0) const { return _fvarPatchArrays[fvarChannel]; }
\n-
87
\n-
89 cl_mem GetFVarPatchIndexBuffer(int fvarChannel = 0) const { return _fvarIndexBuffers[fvarChannel]; }
\n-
90
\n-
92 cl_mem GetFVarPatchParamBuffer(int fvarChannel = 0) const { return _fvarParamBuffers[fvarChannel]; }
\n-
93
\n-
94protected:
\n-
95 CLPatchTable();
\n-
96
\n-
97 bool allocate(Far::PatchTable const *patchTable, cl_context clContext);
\n-
98
\n-
99 cl_mem _patchArrays;
\n-
100 cl_mem _indexBuffer;
\n-
101 cl_mem _patchParamBuffer;
\n-
102
\n-
103 cl_mem _varyingPatchArrays;
\n-
104 cl_mem _varyingIndexBuffer;
\n-
105
\n-
106 std::vector<cl_mem> _fvarPatchArrays;
\n-
107 std::vector<cl_mem> _fvarIndexBuffers;
\n-
108 std::vector<cl_mem> _fvarParamBuffers;
\n-
109
\n-
110};
\n-
111
\n-
112} // end namespace Osd
\n-
113
\n-
114} // end namespace OPENSUBDIV_VERSION
\n-
115using namespace OPENSUBDIV_VERSION;
\n-
116
\n-
117} // end namespace OpenSubdiv
\n-
118
\n-
119#endif // OPENSUBDIV3_OSD_CL_PATCH_TABLE_H
\n+
85private:
\n+
86 cl_mem _sizes;
\n+
87 cl_mem _offsets;
\n+
88 cl_mem _indices;
\n+
89 cl_mem _weights;
\n+
90 cl_mem _duWeights;
\n+
91 cl_mem _dvWeights;
\n+
92 cl_mem _duuWeights;
\n+
93 cl_mem _duvWeights;
\n+
94 cl_mem _dvvWeights;
\n+
95 int _numStencils;
\n+
96};
\n+
97
\n+
98// ---------------------------------------------------------------------------
\n+
99
\n+
100class CLEvaluator {
\n+
101public:
\n+
102 typedef bool Instantiatable;
\n+
103
\n+
105 template <typename DEVICE_CONTEXT>
\n+
106 static CLEvaluator *Create(BufferDescriptor const &srcDesc,
\n+
107 BufferDescriptor const &dstDesc,
\n+
108 BufferDescriptor const &duDesc,
\n+
109 BufferDescriptor const &dvDesc,
\n+
110 DEVICE_CONTEXT deviceContext) {
\n+
111 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n+
112 deviceContext->GetContext(),
\n+
113 deviceContext->GetCommandQueue());
\n+
114 }
\n+
115
\n+
116 static CLEvaluator * Create(BufferDescriptor const &srcDesc,
\n+
117 BufferDescriptor const &dstDesc,
\n+
118 BufferDescriptor const &duDesc,
\n+
119 BufferDescriptor const &dvDesc,
\n+
120 cl_context clContext,
\n+
121 cl_command_queue clCommandQueue) {
\n+
122 CLEvaluator *instance = new CLEvaluator(clContext, clCommandQueue);
\n+
123 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc))
\n+
124 return instance;
\n+
125 delete instance;
\n+
126 return NULL;
\n+
127 }
\n+
128
\n+
130 template <typename DEVICE_CONTEXT>
\n+
131 static CLEvaluator *Create(BufferDescriptor const &srcDesc,
\n+
132 BufferDescriptor const &dstDesc,
\n+
133 BufferDescriptor const &duDesc,
\n+
134 BufferDescriptor const &dvDesc,
\n+
135 BufferDescriptor const &duuDesc,
\n+
136 BufferDescriptor const &duvDesc,
\n+
137 BufferDescriptor const &dvvDesc,
\n+
138 DEVICE_CONTEXT deviceContext) {
\n+
139 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n+
140 duuDesc, duvDesc, dvvDesc,
\n+
141 deviceContext->GetContext(),
\n+
142 deviceContext->GetCommandQueue());
\n+
143 }
\n+
144
\n+
145 static CLEvaluator * Create(BufferDescriptor const &srcDesc,
\n+
146 BufferDescriptor const &dstDesc,
\n+
147 BufferDescriptor const &duDesc,
\n+
148 BufferDescriptor const &dvDesc,
\n+
149 BufferDescriptor const &duuDesc,
\n+
150 BufferDescriptor const &duvDesc,
\n+
151 BufferDescriptor const &dvvDesc,
\n+
152 cl_context clContext,
\n+
153 cl_command_queue clCommandQueue) {
\n+
154 CLEvaluator *instance = new CLEvaluator(clContext, clCommandQueue);
\n+
155 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc,
\n+
156 duuDesc, duvDesc, dvvDesc))
\n+
157 return instance;
\n+
158 delete instance;
\n+
159 return NULL;
\n+
160 }
\n+
161
\n+
163 CLEvaluator(cl_context context, cl_command_queue queue);
\n+
164
\n+
166 ~CLEvaluator();
\n+
167
\n+
173
\n+
217 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
218 typename STENCIL_TABLE, typename DEVICE_CONTEXT>
\n+
219 static bool EvalStencils(
\n+
220 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
221 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
222 STENCIL_TABLE const *stencilTable,
\n+
223 CLEvaluator const *instance,
\n+
224 DEVICE_CONTEXT deviceContext,
\n+
225 unsigned int numStartEvents=0,
\n+
226 const cl_event* startEvents=NULL,
\n+
227 cl_event* endEvent=NULL) {
\n+
228
\n+
229 if (instance) {
\n+
230 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
231 dstBuffer, dstDesc,
\n+
232 stencilTable,
\n+
233 numStartEvents, startEvents, endEvent);
\n+
234 } else {
\n+
235 // Create an instance on demand (slow)
\n+
236 instance = Create(srcDesc, dstDesc,
\n+
237 BufferDescriptor(),
\n+
238 BufferDescriptor(),
\n+
239 deviceContext);
\n+
240 if (instance) {
\n+
241 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
242 dstBuffer, dstDesc,
\n+
243 stencilTable,
\n+
244 numStartEvents, startEvents, endEvent);
\n+
245 delete instance;
\n+
246 return r;
\n+
247 }
\n+
248 return false;
\n+
249 }
\n+
250 }
\n+
251
\n+
307 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
308 typename STENCIL_TABLE, typename DEVICE_CONTEXT>
\n+
309 static bool EvalStencils(
\n+
310 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
311 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
312 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
313 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
314 STENCIL_TABLE const *stencilTable,
\n+
315 CLEvaluator const *instance,
\n+
316 DEVICE_CONTEXT deviceContext,
\n+
317 unsigned int numStartEvents=0,
\n+
318 const cl_event* startEvents=NULL,
\n+
319 cl_event* endEvent=NULL) {
\n+
320
\n+
321 if (instance) {
\n+
322 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
323 dstBuffer, dstDesc,
\n+
324 duBuffer, duDesc,
\n+
325 dvBuffer, dvDesc,
\n+
326 stencilTable,
\n+
327 numStartEvents, startEvents, endEvent);
\n+
328 } else {
\n+
329 // Create an instance on demand (slow)
\n+
330 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,
\n+
331 deviceContext);
\n+
332 if (instance) {
\n+
333 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
334 dstBuffer, dstDesc,
\n+
335 duBuffer, duDesc,
\n+
336 dvBuffer, dvDesc,
\n+
337 stencilTable,
\n+
338 numStartEvents, startEvents, endEvent);
\n+
339 delete instance;
\n+
340 return r;
\n+
341 }
\n+
342 return false;
\n+
343 }
\n+
344 }
\n+
345
\n+
419 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
420 typename STENCIL_TABLE, typename DEVICE_CONTEXT>
\n+
421 static bool EvalStencils(
\n+
422 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
423 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
424 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
425 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
426 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
427 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
428 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
429 STENCIL_TABLE const *stencilTable,
\n+
430 CLEvaluator const *instance,
\n+
431 DEVICE_CONTEXT deviceContext,
\n+
432 unsigned int numStartEvents=0,
\n+
433 const cl_event* startEvents=NULL,
\n+
434 cl_event* endEvent=NULL) {
\n+
435
\n+
436 if (instance) {
\n+
437 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
438 dstBuffer, dstDesc,
\n+
439 duBuffer, duDesc,
\n+
440 dvBuffer, dvDesc,
\n+
441 duuBuffer, duuDesc,
\n+
442 duvBuffer, duvDesc,
\n+
443 dvvBuffer, dvvDesc,
\n+
444 stencilTable,
\n+
445 numStartEvents, startEvents, endEvent);
\n+
446 } else {
\n+
447 // Create an instance on demand (slow)
\n+
448 instance = Create(srcDesc, dstDesc,
\n+
449 duDesc, dvDesc,
\n+
450 duuDesc, duvDesc, dvvDesc,
\n+
451 deviceContext);
\n+
452 if (instance) {
\n+
453 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
454 dstBuffer, dstDesc,
\n+
455 duBuffer, duDesc,
\n+
456 dvBuffer, dvDesc,
\n+
457 duuBuffer, duuDesc,
\n+
458 duvBuffer, duvDesc,
\n+
459 dvvBuffer, dvvDesc,
\n+
460 stencilTable,
\n+
461 numStartEvents, startEvents, endEvent);
\n+
462 delete instance;
\n+
463 return r;
\n+
464 }
\n+
465 return false;
\n+
466 }
\n+
467 }
\n+
468
\n+
499 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
500 bool EvalStencils(
\n+
501 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
502 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
503 STENCIL_TABLE const *stencilTable,
\n+
504 unsigned int numStartEvents=0,
\n+
505 const cl_event* startEvents=NULL,
\n+
506 cl_event* endEvent=NULL) const {
\n+
507 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
508 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
509 stencilTable->GetSizesBuffer(),
\n+
510 stencilTable->GetOffsetsBuffer(),
\n+
511 stencilTable->GetIndicesBuffer(),
\n+
512 stencilTable->GetWeightsBuffer(),
\n+
513 0,
\n+
514 stencilTable->GetNumStencils(),
\n+
515 numStartEvents, startEvents, endEvent);
\n+
516 }
\n+
517
\n+
560 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
561 bool EvalStencils(
\n+
562 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
563 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
564 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
565 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
566 STENCIL_TABLE const *stencilTable,
\n+
567 unsigned int numStartEvents=0,
\n+
568 const cl_event* startEvents=NULL,
\n+
569 cl_event* endEvent=NULL) const {
\n+
570 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
571 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
572 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n+
573 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n+
574 stencilTable->GetSizesBuffer(),
\n+
575 stencilTable->GetOffsetsBuffer(),
\n+
576 stencilTable->GetIndicesBuffer(),
\n+
577 stencilTable->GetWeightsBuffer(),
\n+
578 stencilTable->GetDuWeightsBuffer(),
\n+
579 stencilTable->GetDvWeightsBuffer(),
\n+
580 0,
\n+
581 stencilTable->GetNumStencils(),
\n+
582 numStartEvents, startEvents, endEvent);
\n+
583 }
\n+
584
\n+
645 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
646 bool EvalStencils(
\n+
647 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
648 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
649 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
650 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
651 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
652 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
653 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
654 STENCIL_TABLE const *stencilTable,
\n+
655 unsigned int numStartEvents=0,
\n+
656 const cl_event* startEvents=NULL,
\n+
657 cl_event* endEvent=NULL) const {
\n+
658 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
659 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
660 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n+
661 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n+
662 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,
\n+
663 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,
\n+
664 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,
\n+
665 stencilTable->GetSizesBuffer(),
\n+
666 stencilTable->GetOffsetsBuffer(),
\n+
667 stencilTable->GetIndicesBuffer(),
\n+
668 stencilTable->GetWeightsBuffer(),
\n+
669 stencilTable->GetDuWeightsBuffer(),
\n+
670 stencilTable->GetDvWeightsBuffer(),
\n+
671 stencilTable->GetDuuWeightsBuffer(),
\n+
672 stencilTable->GetDuvWeightsBuffer(),
\n+
673 stencilTable->GetDvvWeightsBuffer(),
\n+
674 0,
\n+
675 stencilTable->GetNumStencils(),
\n+
676 numStartEvents, startEvents, endEvent);
\n+
677 }
\n+
678
\n+
681 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,
\n+
682 cl_mem dst, BufferDescriptor const &dstDesc,
\n+
683 cl_mem sizes,
\n+
684 cl_mem offsets,
\n+
685 cl_mem indices,
\n+
686 cl_mem weights,
\n+
687 int start,
\n+
688 int end,
\n+
689 unsigned int numStartEvents=0,
\n+
690 const cl_event* startEvents=NULL,
\n+
691 cl_event* endEvent=NULL) const;
\n+
692
\n+
741 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,
\n+
742 cl_mem dst, BufferDescriptor const &dstDesc,
\n+
743 cl_mem du, BufferDescriptor const &duDesc,
\n+
744 cl_mem dv, BufferDescriptor const &dvDesc,
\n+
745 cl_mem sizes,
\n+
746 cl_mem offsets,
\n+
747 cl_mem indices,
\n+
748 cl_mem weights,
\n+
749 cl_mem duWeights,
\n+
750 cl_mem dvWeights,
\n+
751 int start,
\n+
752 int end,
\n+
753 unsigned int numStartEvents=0,
\n+
754 const cl_event* startEvents=NULL,
\n+
755 cl_event* endEvent=NULL) const;
\n+
756
\n+
823 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,
\n+
824 cl_mem dst, BufferDescriptor const &dstDesc,
\n+
825 cl_mem du, BufferDescriptor const &duDesc,
\n+
826 cl_mem dv, BufferDescriptor const &dvDesc,
\n+
827 cl_mem duu, BufferDescriptor const &duuDesc,
\n+
828 cl_mem duv, BufferDescriptor const &duvDesc,
\n+
829 cl_mem dvv, BufferDescriptor const &dvvDesc,
\n+
830 cl_mem sizes,
\n+
831 cl_mem offsets,
\n+
832 cl_mem indices,
\n+
833 cl_mem weights,
\n+
834 cl_mem duWeights,
\n+
835 cl_mem dvWeights,
\n+
836 cl_mem duuWeights,
\n+
837 cl_mem duvWeights,
\n+
838 cl_mem dvvWeights,
\n+
839 int start,
\n+
840 int end,
\n+
841 unsigned int numStartEvents=0,
\n+
842 const cl_event* startEvents=NULL,
\n+
843 cl_event* endEvent=NULL) const;
\n+
844
\n+
850
\n+
899 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
900 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n+
901 typename DEVICE_CONTEXT>
\n+
902 static bool EvalPatches(
\n+
903 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
904 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
905 int numPatchCoords,
\n+
906 PATCHCOORD_BUFFER *patchCoords,
\n+
907 PATCH_TABLE *patchTable,
\n+
908 CLEvaluator const *instance,
\n+
909 DEVICE_CONTEXT deviceContext,
\n+
910 unsigned int numStartEvents=0,
\n+
911 const cl_event* startEvents=NULL,
\n+
912 cl_event* endEvent=NULL) {
\n+
913
\n+
914 if (instance) {
\n+
915 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
916 dstBuffer, dstDesc,
\n+
917 numPatchCoords, patchCoords,
\n+
918 patchTable,
\n+
919 numStartEvents, startEvents, endEvent);
\n+
920 } else {
\n+
921 // Create an instance on demand (slow)
\n+
922 (void)deviceContext; // unused
\n+
923 instance = Create(srcDesc, dstDesc,
\n+
924 BufferDescriptor(),
\n+
925 BufferDescriptor(),
\n+
926 deviceContext);
\n+
927 if (instance) {
\n+
928 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
929 dstBuffer, dstDesc,
\n+
930 numPatchCoords, patchCoords,
\n+
931 patchTable,
\n+
932 numStartEvents, startEvents, endEvent);
\n+
933 delete instance;
\n+
934 return r;
\n+
935 }
\n+
936 return false;
\n+
937 }
\n+
938 }
\n+
939
\n+
1000 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1001 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n+
1002 typename DEVICE_CONTEXT>
\n+
1003 static bool EvalPatches(
\n+
1004 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1005 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1006 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1007 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1008 int numPatchCoords,
\n+
1009 PATCHCOORD_BUFFER *patchCoords,
\n+
1010 PATCH_TABLE *patchTable,
\n+
1011 CLEvaluator const *instance,
\n+
1012 DEVICE_CONTEXT deviceContext,
\n+
1013 unsigned int numStartEvents=0,
\n+
1014 const cl_event* startEvents=NULL,
\n+
1015 cl_event* endEvent=NULL) {
\n+
1016
\n+
1017 if (instance) {
\n+
1018 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
1019 dstBuffer, dstDesc,
\n+
1020 duBuffer, duDesc,
\n+
1021 dvBuffer, dvDesc,
\n+
1022 numPatchCoords, patchCoords,
\n+
1023 patchTable,
\n+
1024 numStartEvents, startEvents, endEvent);
\n+
1025 } else {
\n+
1026 // Create an instance on demand (slow)
\n+
1027 (void)deviceContext; // unused
\n+
1028 instance = Create(srcDesc, dstDesc, duDesc, dvDesc, deviceContext);
\n+
1029 if (instance) {
\n+
1030 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
1031 dstBuffer, dstDesc,
\n+
1032 duBuffer, duDesc,
\n+
1033 dvBuffer, dvDesc,
\n+
1034 numPatchCoords, patchCoords,
\n+
1035 patchTable,
\n+
1036 numStartEvents, startEvents, endEvent);
\n+
1037 delete instance;
\n+
1038 return r;
\n+
1039 }
\n+
1040 return false;
\n+
1041 }
\n+
1042 }
\n+
1043
\n+
1122 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1123 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n+
1124 typename DEVICE_CONTEXT>
\n+
1125 static bool EvalPatches(
\n+
1126 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1127 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1128 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1129 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1130 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1131 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1132 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1133 int numPatchCoords,
\n+
1134 PATCHCOORD_BUFFER *patchCoords,
\n+
1135 PATCH_TABLE *patchTable,
\n+
1136 CLEvaluator const *instance,
\n+
1137 DEVICE_CONTEXT deviceContext,
\n+
1138 unsigned int numStartEvents=0,
\n+
1139 const cl_event* startEvents=NULL,
\n+
1140 cl_event* endEvent=NULL) {
\n+
1141
\n+
1142 if (instance) {
\n+
1143 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
1144 dstBuffer, dstDesc,
\n+
1145 duBuffer, duDesc,
\n+
1146 dvBuffer, dvDesc,
\n+
1147 duuBuffer, duuDesc,
\n+
1148 duvBuffer, duvDesc,
\n+
1149 dvvBuffer, dvvDesc,
\n+
1150 numPatchCoords, patchCoords,
\n+
1151 patchTable,
\n+
1152 numStartEvents, startEvents, endEvent);
\n+
1153 } else {
\n+
1154 // Create an instance on demand (slow)
\n+
1155 (void)deviceContext; // unused
\n+
1156 instance = Create(srcDesc, dstDesc,
\n+
1157 duDesc, dvDesc,
\n+
1158 duuDesc, duvDesc, dvvDesc,
\n+
1159 deviceContext);
\n+
1160 if (instance) {
\n+
1161 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
1162 dstBuffer, dstDesc,
\n+
1163 duBuffer, duDesc,
\n+
1164 dvBuffer, dvDesc,
\n+
1165 duuBuffer, duuDesc,
\n+
1166 duvBuffer, duvDesc,
\n+
1167 dvvBuffer, dvvDesc,
\n+
1168 numPatchCoords, patchCoords,
\n+
1169 patchTable,
\n+
1170 numStartEvents, startEvents, endEvent);
\n+
1171 delete instance;
\n+
1172 return r;
\n+
1173 }
\n+
1174 return false;
\n+
1175 }
\n+
1176 }
\n+
1177
\n+
1215 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1216 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1217 bool EvalPatches(
\n+
1218 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1219 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1220 int numPatchCoords,
\n+
1221 PATCHCOORD_BUFFER *patchCoords,
\n+
1222 PATCH_TABLE *patchTable,
\n+
1223 unsigned int numStartEvents=0,
\n+
1224 const cl_event* startEvents=NULL,
\n+
1225 cl_event* endEvent=NULL) const {
\n+
1226
\n+
1227 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
1228 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
1229 0, BufferDescriptor(),
\n+
1230 0, BufferDescriptor(),
\n+
1231 numPatchCoords,
\n+
1232 patchCoords->BindCLBuffer(_clCommandQueue),
\n+
1233 patchTable->GetPatchArrayBuffer(),
\n+
1234 patchTable->GetPatchIndexBuffer(),
\n+
1235 patchTable->GetPatchParamBuffer(),
\n+
1236 numStartEvents, startEvents, endEvent);
\n+
1237 }
\n+
1238
\n+
1286 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1287 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1288 bool EvalPatches(
\n+
1289 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1290 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1291 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1292 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1293 int numPatchCoords,
\n+
1294 PATCHCOORD_BUFFER *patchCoords,
\n+
1295 PATCH_TABLE *patchTable,
\n+
1296 unsigned int numStartEvents=0,
\n+
1297 const cl_event* startEvents=NULL,
\n+
1298 cl_event* endEvent=NULL) const {
\n+
1299
\n+
1300 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
1301 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
1302 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n+
1303 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n+
1304 numPatchCoords,
\n+
1305 patchCoords->BindCLBuffer(_clCommandQueue),
\n+
1306 patchTable->GetPatchArrayBuffer(),
\n+
1307 patchTable->GetPatchIndexBuffer(),
\n+
1308 patchTable->GetPatchParamBuffer(),
\n+
1309 numStartEvents, startEvents, endEvent);
\n+
1310 }
\n+
1311
\n+
1377 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1378 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1379 bool EvalPatches(
\n+
1380 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1381 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1382 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1383 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1384 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1385 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1386 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1387 int numPatchCoords,
\n+
1388 PATCHCOORD_BUFFER *patchCoords,
\n+
1389 PATCH_TABLE *patchTable,
\n+
1390 unsigned int numStartEvents=0,
\n+
1391 const cl_event* startEvents=NULL,
\n+
1392 cl_event* endEvent=NULL) const {
\n+
1393
\n+
1394 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
1395 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
1396 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n+
1397 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n+
1398 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,
\n+
1399 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,
\n+
1400 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,
\n+
1401 numPatchCoords,
\n+
1402 patchCoords->BindCLBuffer(_clCommandQueue),
\n+
1403 patchTable->GetPatchArrayBuffer(),
\n+
1404 patchTable->GetPatchIndexBuffer(),
\n+
1405 patchTable->GetPatchParamBuffer(),
\n+
1406 numStartEvents, startEvents, endEvent);
\n+
1407 }
\n+
1408
\n+
1409 bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc,
\n+
1410 cl_mem dst, BufferDescriptor const &dstDesc,
\n+
1411 cl_mem du, BufferDescriptor const &duDesc,
\n+
1412 cl_mem dv, BufferDescriptor const &dvDesc,
\n+
1413 int numPatchCoords,
\n+
1414 cl_mem patchCoordsBuffer,
\n+
1415 cl_mem patchArrayBuffer,
\n+
1416 cl_mem patchIndexBuffer,
\n+
1417 cl_mem patchParamsBuffer,
\n+
1418 unsigned int numStartEvents=0,
\n+
1419 const cl_event* startEvents=NULL,
\n+
1420 cl_event* endEvent=NULL) const;
\n+
1421
\n+
1422 bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc,
\n+
1423 cl_mem dst, BufferDescriptor const &dstDesc,
\n+
1424 cl_mem du, BufferDescriptor const &duDesc,
\n+
1425 cl_mem dv, BufferDescriptor const &dvDesc,
\n+
1426 cl_mem duu, BufferDescriptor const &duuDesc,
\n+
1427 cl_mem duv, BufferDescriptor const &duvDesc,
\n+
1428 cl_mem dvv, BufferDescriptor const &dvvDesc,
\n+
1429 int numPatchCoords,
\n+
1430 cl_mem patchCoordsBuffer,
\n+
1431 cl_mem patchArrayBuffer,
\n+
1432 cl_mem patchIndexBuffer,
\n+
1433 cl_mem patchParamsBuffer,
\n+
1434 unsigned int numStartEvents=0,
\n+
1435 const cl_event* startEvents=NULL,
\n+
1436 cl_event* endEvent=NULL) const;
\n+
1437
\n+
1486 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1487 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n+
1488 typename DEVICE_CONTEXT>
\n+
1489 static bool EvalPatchesVarying(
\n+
1490 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1491 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1492 int numPatchCoords,
\n+
1493 PATCHCOORD_BUFFER *patchCoords,
\n+
1494 PATCH_TABLE *patchTable,
\n+
1495 CLEvaluator const *instance,
\n+
1496 DEVICE_CONTEXT deviceContext,
\n+
1497 unsigned int numStartEvents=0,
\n+
1498 const cl_event* startEvents=NULL,
\n+
1499 cl_event* endEvent=NULL) {
\n+
1500
\n+
1501 if (instance) {
\n+
1502 return instance->EvalPatchesVarying(
\n+
1503 srcBuffer, srcDesc,
\n+
1504 dstBuffer, dstDesc,
\n+
1505 numPatchCoords, patchCoords,
\n+
1506 patchTable,
\n+
1507 numStartEvents, startEvents, endEvent);
\n+
1508 } else {
\n+
1509 // Create an instance on demand (slow)
\n+
1510 (void)deviceContext; // unused
\n+
1511 instance = Create(srcDesc, dstDesc,
\n+
1512 BufferDescriptor(),
\n+
1513 BufferDescriptor(),
\n+
1514 deviceContext);
\n+
1515 if (instance) {
\n+
1516 bool r = instance->EvalPatchesVarying(
\n+
1517 srcBuffer, srcDesc,
\n+
1518 dstBuffer, dstDesc,
\n+
1519 numPatchCoords, patchCoords,
\n+
1520 patchTable,
\n+
1521 numStartEvents, startEvents, endEvent);
\n+
1522 delete instance;
\n+
1523 return r;
\n+
1524 }
\n+
1525 return false;
\n+
1526 }
\n+
1527 }
\n+
1528
\n+
1566 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1567 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1568 bool EvalPatchesVarying(
\n+
1569 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1570 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1571 int numPatchCoords,
\n+
1572 PATCHCOORD_BUFFER *patchCoords,
\n+
1573 PATCH_TABLE *patchTable,
\n+
1574 unsigned int numStartEvents=0,
\n+
1575 const cl_event* startEvents=NULL,
\n+
1576 cl_event* endEvent=NULL) const {
\n+
1577
\n+
1578 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
1579 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
1580 0, BufferDescriptor(),
\n+
1581 0, BufferDescriptor(),
\n+
1582 numPatchCoords,
\n+
1583 patchCoords->BindCLBuffer(_clCommandQueue),
\n+
1584 patchTable->GetVaryingPatchArrayBuffer(),
\n+
1585 patchTable->GetVaryingPatchIndexBuffer(),
\n+
1586 patchTable->GetPatchParamBuffer(),
\n+
1587 numStartEvents, startEvents, endEvent);
\n+
1588 }
\n+
1589
\n+
1650 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1651 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n+
1652 typename DEVICE_CONTEXT>
\n+
1653 static bool EvalPatchesVarying(
\n+
1654 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1655 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1656 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1657 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1658 int numPatchCoords,
\n+
1659 PATCHCOORD_BUFFER *patchCoords,
\n+
1660 PATCH_TABLE *patchTable,
\n+
1661 CLEvaluator const *instance,
\n+
1662 DEVICE_CONTEXT deviceContext,
\n+
1663 unsigned int numStartEvents=0,
\n+
1664 const cl_event* startEvents=NULL,
\n+
1665 cl_event* endEvent=NULL) {
\n+
1666
\n+
1667 if (instance) {
\n+
1668 return instance->EvalPatchesVarying(
\n+
1669 srcBuffer, srcDesc,
\n+
1670 dstBuffer, dstDesc,
\n+
1671 duBuffer, duDesc,
\n+
1672 dvBuffer, dvDesc,
\n+
1673 numPatchCoords, patchCoords,
\n+
1674 patchTable,
\n+
1675 numStartEvents, startEvents, endEvent);
\n+
1676 } else {
\n+
1677 // Create an instance on demand (slow)
\n+
1678 (void)deviceContext; // unused
\n+
1679 instance = Create(srcDesc, dstDesc,
\n+
1680 duDesc, dvDesc,
\n+
1681 deviceContext);
\n+
1682 if (instance) {
\n+
1683 bool r = instance->EvalPatchesVarying(
\n+
1684 srcBuffer, srcDesc,
\n+
1685 dstBuffer, dstDesc,
\n+
1686 duBuffer, duDesc,
\n+
1687 dvBuffer, dvDesc,
\n+
1688 numPatchCoords, patchCoords,
\n+
1689 patchTable,
\n+
1690 numStartEvents, startEvents, endEvent);
\n+
1691 delete instance;
\n+
1692 return r;
\n+
1693 }
\n+
1694 return false;
\n+
1695 }
\n+
1696 }
\n+
1697
\n+
1747 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1748 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1749 bool EvalPatchesVarying(
\n+
1750 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1751 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1752 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1753 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1754 int numPatchCoords,
\n+
1755 PATCHCOORD_BUFFER *patchCoords,
\n+
1756 PATCH_TABLE *patchTable,
\n+
1757 unsigned int numStartEvents=0,
\n+
1758 const cl_event* startEvents=NULL,
\n+
1759 cl_event* endEvent=NULL) const {
\n+
1760
\n+
1761 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
1762 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
1763 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n+
1764 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n+
1765 numPatchCoords,
\n+
1766 patchCoords->BindCLBuffer(_clCommandQueue),
\n+
1767 patchTable->GetVaryingPatchArrayBuffer(),
\n+
1768 patchTable->GetVaryingPatchIndexBuffer(),
\n+
1769 patchTable->GetPatchParamBuffer(),
\n+
1770 numStartEvents, startEvents, endEvent);
\n+
1771 }
\n+
1772
\n+
1851 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1852 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n+
1853 typename DEVICE_CONTEXT>
\n+
1854 static bool EvalPatchesVarying(
\n+
1855 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1856 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1857 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1858 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1859 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1860 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1861 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1862 int numPatchCoords,
\n+
1863 PATCHCOORD_BUFFER *patchCoords,
\n+
1864 PATCH_TABLE *patchTable,
\n+
1865 CLEvaluator const *instance,
\n+
1866 DEVICE_CONTEXT deviceContext,
\n+
1867 unsigned int numStartEvents=0,
\n+
1868 const cl_event* startEvents=NULL,
\n+
1869 cl_event* endEvent=NULL) {
\n+
1870
\n+
1871 if (instance) {
\n+
1872 return instance->EvalPatchesVarying(
\n+
1873 srcBuffer, srcDesc,
\n+
1874 dstBuffer, dstDesc,
\n+
1875 duBuffer, duDesc,
\n+
1876 dvBuffer, dvDesc,
\n+
1877 duuBuffer, duuDesc,
\n+
1878 duvBuffer, duvDesc,
\n+
1879 dvvBuffer, dvvDesc,
\n+
1880 numPatchCoords, patchCoords,
\n+
1881 patchTable,
\n+
1882 numStartEvents, startEvents, endEvent);
\n+
1883 } else {
\n+
1884 // Create an instance on demand (slow)
\n+
1885 (void)deviceContext; // unused
\n+
1886 instance = Create(srcDesc, dstDesc,
\n+
1887 duDesc, dvDesc,
\n+
1888 duuDesc, duvDesc, dvvDesc,
\n+
1889 deviceContext);
\n+
1890 if (instance) {
\n+
1891 bool r = instance->EvalPatchesVarying(
\n+
1892 srcBuffer, srcDesc,
\n+
1893 dstBuffer, dstDesc,
\n+
1894 duBuffer, duDesc,
\n+
1895 dvBuffer, dvDesc,
\n+
1896 duuBuffer, duuDesc,
\n+
1897 duvBuffer, duvDesc,
\n+
1898 dvvBuffer, dvvDesc,
\n+
1899 numPatchCoords, patchCoords,
\n+
1900 patchTable,
\n+
1901 numStartEvents, startEvents, endEvent);
\n+
1902 delete instance;
\n+
1903 return r;
\n+
1904 }
\n+
1905 return false;
\n+
1906 }
\n+
1907 }
\n+
1908
\n+
1976 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1977 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1978 bool EvalPatchesVarying(
\n+
1979 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1980 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1981 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1982 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1983 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1984 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1985 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1986 int numPatchCoords,
\n+
1987 PATCHCOORD_BUFFER *patchCoords,
\n+
1988 PATCH_TABLE *patchTable,
\n+
1989 unsigned int numStartEvents=0,
\n+
1990 const cl_event* startEvents=NULL,
\n+
1991 cl_event* endEvent=NULL) const {
\n+
1992
\n+
1993 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
1994 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
1995 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n+
1996 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n+
1997 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,
\n+
1998 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,
\n+
1999 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,
\n+
2000 numPatchCoords,
\n+
2001 patchCoords->BindCLBuffer(_clCommandQueue),
\n+
2002 patchTable->GetVaryingPatchArrayBuffer(),
\n+
2003 patchTable->GetVaryingPatchIndexBuffer(),
\n+
2004 patchTable->GetPatchParamBuffer(),
\n+
2005 numStartEvents, startEvents, endEvent);
\n+
2006 }
\n+
2007
\n+
2058 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
2059 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n+
2060 typename DEVICE_CONTEXT>
\n+
2061 static bool EvalPatchesFaceVarying(
\n+
2062 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
2063 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
2064 int numPatchCoords,
\n+
2065 PATCHCOORD_BUFFER *patchCoords,
\n+
2066 PATCH_TABLE *patchTable,
\n+
2067 int fvarChannel,
\n+
2068 CLEvaluator const *instance,
\n+
2069 DEVICE_CONTEXT deviceContext,
\n+
2070 unsigned int numStartEvents=0,
\n+
2071 const cl_event* startEvents=NULL,
\n+
2072 cl_event* endEvent=NULL) {
\n+
2073
\n+
2074 if (instance) {
\n+
2075 return instance->EvalPatchesFaceVarying(
\n+
2076 srcBuffer, srcDesc,
\n+
2077 dstBuffer, dstDesc,
\n+
2078 numPatchCoords, patchCoords,
\n+
2079 patchTable, fvarChannel,
\n+
2080 numStartEvents, startEvents, endEvent);
\n+
2081 } else {
\n+
2082 // Create an instance on demand (slow)
\n+
2083 (void)deviceContext; // unused
\n+
2084 instance = Create(srcDesc, dstDesc,
\n+
2085 BufferDescriptor(),
\n+
2086 BufferDescriptor(),
\n+
2087 deviceContext);
\n+
2088 if (instance) {
\n+
2089 bool r = instance->EvalPatchesFaceVarying(
\n+
2090 srcBuffer, srcDesc,
\n+
2091 dstBuffer, dstDesc,
\n+
2092 numPatchCoords, patchCoords,
\n+
2093 patchTable, fvarChannel,
\n+
2094 numStartEvents, startEvents, endEvent);
\n+
2095 delete instance;
\n+
2096 return r;
\n+
2097 }
\n+
2098 return false;
\n+
2099 }
\n+
2100 }
\n+
2101
\n+
2141 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
2142 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
2143 bool EvalPatchesFaceVarying(
\n+
2144 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
2145 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
2146 int numPatchCoords,
\n+
2147 PATCHCOORD_BUFFER *patchCoords,
\n+
2148 PATCH_TABLE *patchTable,
\n+
2149 int fvarChannel = 0,
\n+
2150 unsigned int numStartEvents=0,
\n+
2151 const cl_event* startEvents=NULL,
\n+
2152 cl_event* endEvent=NULL) const {
\n+
2153
\n+
2154 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
2155 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
2156 0, BufferDescriptor(),
\n+
2157 0, BufferDescriptor(),
\n+
2158 numPatchCoords,
\n+
2159 patchCoords->BindCLBuffer(_clCommandQueue),
\n+
2160 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
2161 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
2162 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n+
2163 numStartEvents, startEvents, endEvent);
\n+
2164 }
\n+
2165
\n+
2228 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
2229 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n+
2230 typename DEVICE_CONTEXT>
\n+
2231 static bool EvalPatchesFaceVarying(
\n+
2232 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
2233 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
2234 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
2235 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
2236 int numPatchCoords,
\n+
2237 PATCHCOORD_BUFFER *patchCoords,
\n+
2238 PATCH_TABLE *patchTable,
\n+
2239 int fvarChannel,
\n+
2240 CLEvaluator const *instance,
\n+
2241 DEVICE_CONTEXT deviceContext,
\n+
2242 unsigned int numStartEvents=0,
\n+
2243 const cl_event* startEvents=NULL,
\n+
2244 cl_event* endEvent=NULL) {
\n+
2245
\n+
2246 if (instance) {
\n+
2247 return instance->EvalPatchesFaceVarying(
\n+
2248 srcBuffer, srcDesc,
\n+
2249 dstBuffer, dstDesc,
\n+
2250 duBuffer, duDesc,
\n+
2251 dvBuffer, dvDesc,
\n+
2252 numPatchCoords, patchCoords,
\n+
2253 patchTable, fvarChannel,
\n+
2254 numStartEvents, startEvents, endEvent);
\n+
2255 } else {
\n+
2256 // Create an instance on demand (slow)
\n+
2257 (void)deviceContext; // unused
\n+
2258 instance = Create(srcDesc, dstDesc,
\n+
2259 duDesc, dvDesc, deviceContext);
\n+
2260 if (instance) {
\n+
2261 bool r = instance->EvalPatchesFaceVarying(
\n+
2262 srcBuffer, srcDesc,
\n+
2263 dstBuffer, dstDesc,
\n+
2264 duBuffer, duDesc,
\n+
2265 dvBuffer, dvDesc,
\n+
2266 numPatchCoords, patchCoords,
\n+
2267 patchTable, fvarChannel,
\n+
2268 numStartEvents, startEvents, endEvent);
\n+
2269 delete instance;
\n+
2270 return r;
\n+
2271 }
\n+
2272 return false;
\n+
2273 }
\n+
2274 }
\n+
2275
\n+
2327 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
2328 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
2329 bool EvalPatchesFaceVarying(
\n+
2330 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
2331 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
2332 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
2333 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
2334 int numPatchCoords,
\n+
2335 PATCHCOORD_BUFFER *patchCoords,
\n+
2336 PATCH_TABLE *patchTable,
\n+
2337 int fvarChannel = 0,
\n+
2338 unsigned int numStartEvents=0,
\n+
2339 const cl_event* startEvents=NULL,
\n+
2340 cl_event* endEvent=NULL) const {
\n+
2341
\n+
2342 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
2343 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
2344 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n+
2345 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n+
2346 numPatchCoords,
\n+
2347 patchCoords->BindCLBuffer(_clCommandQueue),
\n+
2348 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
2349 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
2350 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n+
2351 numStartEvents, startEvents, endEvent);
\n+
2352 }
\n+
2353
\n+
2434 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
2435 typename PATCHCOORD_BUFFER, typename PATCH_TABLE,
\n+
2436 typename DEVICE_CONTEXT>
\n+
2437 static bool EvalPatchesFaceVarying(
\n+
2438 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
2439 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
2440 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
2441 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
2442 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
2443 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
2444 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
2445 int numPatchCoords,
\n+
2446 PATCHCOORD_BUFFER *patchCoords,
\n+
2447 PATCH_TABLE *patchTable,
\n+
2448 int fvarChannel,
\n+
2449 CLEvaluator const *instance,
\n+
2450 DEVICE_CONTEXT deviceContext,
\n+
2451 unsigned int numStartEvents=0,
\n+
2452 const cl_event* startEvents=NULL,
\n+
2453 cl_event* endEvent=NULL) {
\n+
2454
\n+
2455 if (instance) {
\n+
2456 return instance->EvalPatchesFaceVarying(
\n+
2457 srcBuffer, srcDesc,
\n+
2458 dstBuffer, dstDesc,
\n+
2459 duBuffer, duDesc,
\n+
2460 dvBuffer, dvDesc,
\n+
2461 duuBuffer, duuDesc,
\n+
2462 duvBuffer, duvDesc,
\n+
2463 dvvBuffer, dvvDesc,
\n+
2464 numPatchCoords, patchCoords,
\n+
2465 patchTable, fvarChannel,
\n+
2466 numStartEvents, startEvents, endEvent);
\n+
2467 } else {
\n+
2468 // Create an instance on demand (slow)
\n+
2469 (void)deviceContext; // unused
\n+
2470 instance = Create(srcDesc, dstDesc,
\n+
2471 duDesc, dvDesc,
\n+
2472 duuDesc, duvDesc, dvvDesc,
\n+
2473 deviceContext);
\n+
2474 if (instance) {
\n+
2475 bool r = instance->EvalPatchesFaceVarying(
\n+
2476 srcBuffer, srcDesc,
\n+
2477 dstBuffer, dstDesc,
\n+
2478 duBuffer, duDesc,
\n+
2479 dvBuffer, dvDesc,
\n+
2480 duuBuffer, duuDesc,
\n+
2481 duvBuffer, duvDesc,
\n+
2482 dvvBuffer, dvvDesc,
\n+
2483 numPatchCoords, patchCoords,
\n+
2484 patchTable, fvarChannel,
\n+
2485 numStartEvents, startEvents, endEvent);
\n+
2486 delete instance;
\n+
2487 return r;
\n+
2488 }
\n+
2489 return false;
\n+
2490 }
\n+
2491 }
\n+
2492
\n+
2562 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
2563 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
2564 bool EvalPatchesFaceVarying(
\n+
2565 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
2566 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
2567 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
2568 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
2569 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
2570 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
2571 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
2572 int numPatchCoords,
\n+
2573 PATCHCOORD_BUFFER *patchCoords,
\n+
2574 PATCH_TABLE *patchTable,
\n+
2575 int fvarChannel = 0,
\n+
2576 unsigned int numStartEvents=0,
\n+
2577 const cl_event* startEvents=NULL,
\n+
2578 cl_event* endEvent=NULL) const {
\n+
2579
\n+
2580 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,
\n+
2581 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,
\n+
2582 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,
\n+
2583 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,
\n+
2584 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,
\n+
2585 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,
\n+
2586 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,
\n+
2587 numPatchCoords,
\n+
2588 patchCoords->BindCLBuffer(_clCommandQueue),
\n+
2589 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
2590 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
2591 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n+
2592 numStartEvents, startEvents, endEvent);
\n+
2593 }
\n+
2594
\n+
2600
\n+
2603 bool Compile(BufferDescriptor const &srcDesc,
\n+
2604 BufferDescriptor const &dstDesc,
\n+
2605 BufferDescriptor const &duDesc = BufferDescriptor(),
\n+
2606 BufferDescriptor const &dvDesc = BufferDescriptor(),
\n+
2607 BufferDescriptor const &duuDesc = BufferDescriptor(),
\n+
2608 BufferDescriptor const &duvDesc = BufferDescriptor(),
\n+
2609 BufferDescriptor const &dvvDesc = BufferDescriptor());
\n+
2610
\n+
2612 template <typename DEVICE_CONTEXT>
\n+
2613 static void Synchronize(DEVICE_CONTEXT deviceContext) {
\n+
2614 Synchronize(deviceContext->GetCommandQueue());
\n+
2615 }
\n+
2616
\n+
2617 static void Synchronize(cl_command_queue queue);
\n+
2618
\n+
2619private:
\n+
2620 cl_context _clContext;
\n+
2621 cl_command_queue _clCommandQueue;
\n+
2622 cl_program _program;
\n+
2623 cl_kernel _stencilKernel;
\n+
2624 cl_kernel _stencilDerivKernel;
\n+
2625 cl_kernel _patchKernel;
\n+
2626};
\n+
2627
\n+
2628} // end namespace Osd
\n+
2629
\n+
2630} // end namespace OPENSUBDIV_VERSION
\n+
2631using namespace OPENSUBDIV_VERSION;
\n+
2632
\n+
2633} // end namespace OpenSubdiv
\n+
2634
\n+
2635
\n+
2636#endif // OPENSUBDIV3_OSD_CL_EVALUATOR_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchParamBuffer
cl_mem GetPatchParamBuffer() const
Returns the CL memory of the array of Osd::PatchParam buffer.
Definition: clPatchTable.h:74
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchArrayBuffer
cl_mem GetFVarPatchArrayBuffer(int fvarChannel=0) const
Returns the CL memory of the array of Osd::PatchArray buffer.
Definition: clPatchTable.h:86
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchArrayBuffer
cl_mem GetPatchArrayBuffer() const
Returns the CL memory of the array of Osd::PatchArray buffer.
Definition: clPatchTable.h:68
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::Create
static CLPatchTable * Create(Far::PatchTable const *patchTable, cl_context clContext)
Creator. Returns NULL if error.
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchIndexBuffer
cl_mem GetFVarPatchIndexBuffer(int fvarChannel=0) const
Returns the CL memory of the face-varying control vertices.
Definition: clPatchTable.h:89
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetVaryingPatchIndexBuffer
cl_mem GetVaryingPatchIndexBuffer() const
Returns the CL memory of the varying control vertices.
Definition: clPatchTable.h:80
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetNumFVarChannels
int GetNumFVarChannels() const
Returns the number of face-varying channel buffers.
Definition: clPatchTable.h:83
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchIndexBuffer
cl_mem GetPatchIndexBuffer() const
Returns the CL memory of the patch control vertices.
Definition: clPatchTable.h:71
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::Create
static CLPatchTable * Create(Far::PatchTable const *patchTable, DEVICE_CONTEXT context)
Definition: clPatchTable.h:59
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchParamBuffer
cl_mem GetFVarPatchParamBuffer(int fvarChannel=0) const
Returns the CL memory of the array of Osd::PatchParam buffer.
Definition: clPatchTable.h:92
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::allocate
bool allocate(Far::PatchTable const *patchTable, cl_context clContext)
\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetVaryingPatchArrayBuffer
cl_mem GetVaryingPatchArrayBuffer() const
Returns the CL memory of the array of Osd::PatchArray buffer.
Definition: clPatchTable.h:77
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::CLStencilTable
CLStencilTable(Far::StencilTable const *stencilTable, cl_context clContext)
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::Create
static CLStencilTable * Create(Far::LimitStencilTable const *limitStencilTable, DEVICE_CONTEXT context)
Definition: clEvaluator.h:61
\n+\n+\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::Create
static CLStencilTable * Create(Far::StencilTable const *stencilTable, DEVICE_CONTEXT context)
Definition: clEvaluator.h:55
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::CLStencilTable
CLStencilTable(Far::LimitStencilTable const *limitStencilTable, cl_context clContext)
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: clEvaluator.h:309
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Synchronize
static void Synchronize(cl_command_queue queue)
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst, BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc, cl_mem dv, BufferDescriptor const &dvDesc, cl_mem sizes, cl_mem offsets, cl_mem indices, cl_mem weights, cl_mem duWeights, cl_mem dvWeights, int start, int end, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Dispatch the CL compute kernel asynchronously. returns false if the kernel hasn't been compiled yet.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: clEvaluator.h:219
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create
static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, cl_context clContext, cl_command_queue clCommandQueue)
Definition: clEvaluator.h:116
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2231
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst, BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc, cl_mem dv, BufferDescriptor const &dvDesc, int numPatchCoords, cl_mem patchCoordsBuffer, cl_mem patchArrayBuffer, cl_mem patchIndexBuffer, cl_mem patchParamsBuffer, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1978
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic stencil function.
Definition: clEvaluator.h:561
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: clEvaluator.h:1379
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create
static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, DEVICE_CONTEXT deviceContext)
Generic creator template.
Definition: clEvaluator.h:106
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2061
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create
static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, DEVICE_CONTEXT deviceContext)
Generic creator template.
Definition: clEvaluator.h:131
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst, BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc, cl_mem dv, BufferDescriptor const &dvDesc, cl_mem duu, BufferDescriptor const &duuDesc, cl_mem duv, BufferDescriptor const &duvDesc, cl_mem dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, cl_mem patchCoordsBuffer, cl_mem patchArrayBuffer, cl_mem patchIndexBuffer, cl_mem patchParamsBuffer, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst, BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc, cl_mem dv, BufferDescriptor const &dvDesc, cl_mem duu, BufferDescriptor const &duuDesc, cl_mem duv, BufferDescriptor const &duvDesc, cl_mem dvv, BufferDescriptor const &dvvDesc, cl_mem sizes, cl_mem offsets, cl_mem indices, cl_mem weights, cl_mem duWeights, cl_mem dvWeights, cl_mem duuWeights, cl_mem duvWeights, cl_mem dvvWeights, int start, int end, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Dispatch the CL compute kernel asynchronously. returns false if the kernel hasn't been compiled yet.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2564
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1217
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic stencil function.
Definition: clEvaluator.h:646
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1653
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Synchronize
static void Synchronize(DEVICE_CONTEXT deviceContext)
Wait the OpenCL kernels finish.
Definition: clEvaluator.h:2613
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1125
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: clEvaluator.h:421
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst, BufferDescriptor const &dstDesc, cl_mem sizes, cl_mem offsets, cl_mem indices, cl_mem weights, int start, int end, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1489
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2329
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2437
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create
static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, cl_context clContext, cl_command_queue clCommandQueue)
Definition: clEvaluator.h:145
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1003
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1749
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:902
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1568
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:1854
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic stencil function.
Definition: clEvaluator.h:500
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: clEvaluator.h:1288
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::CLEvaluator
CLEvaluator(cl_context context, cl_command_queue queue)
Constructor.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Compile
bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc=BufferDescriptor(), BufferDescriptor const &dvDesc=BufferDescriptor(), BufferDescriptor const &duuDesc=BufferDescriptor(), BufferDescriptor const &duvDesc=BufferDescriptor(), BufferDescriptor const &dvvDesc=BufferDescriptor())
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: clEvaluator.h:2143
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-clPatchTable.h\n+clEvaluator.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,174 +30,1556 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_CL_PATCH_TABLE_H\n- 26#define OPENSUBDIV3_OSD_CL_PATCH_TABLE_H\n+ 25#ifndef OPENSUBDIV3_OSD_CL_EVALUATOR_H\n+ 26#define OPENSUBDIV3_OSD_CL_EVALUATOR_H\n 27\n 28#include \"../version.h\"\n 29\n 30#include \"../osd/opencl.h\"\n- 31#include \"../osd/nonCopyable.h\"\n- 32#include \"../osd/types.h\"\n+ 31#include \"../osd/types.h\"\n+ 32#include \"../osd/bufferDescriptor.h\"\n 33\n- 34#include \n- 35\n- 36namespace OpenSubdiv {\n- 37namespace OPENSUBDIV_VERSION {\n- 38\n- 39namespace Far{\n- 40 class PatchTable;\n- 41};\n+ 34namespace OpenSubdiv {\n+ 35namespace OPENSUBDIV_VERSION {\n+ 36\n+ 37namespace Far {\n+ 38 class PatchTable;\n+ 39 class StencilTable;\n+ 40 class LimitStencilTable;\n+ 41}\n 42\n 43namespace Osd {\n 44\n-52class CLPatchTable : private NonCopyable {\n+52class CLStencilTable {\n 53public:\n-55 static CLPatchTable *Create(Far::PatchTable const *patchTable,\n- 56 cl_context clContext);\n- 57\n- 58 template \n-59 static CLPatchTable * Create(Far::PatchTable const *patchTable,\n- 60 DEVICE_CONTEXT context) {\n- 61 return Create(patchTable, context->GetContext());\n- 62 }\n- 63\n-65 ~CLPatchTable();\n+ 54 template \n+55 static CLStencilTable *Create(Far::StencilTable const *stencilTable,\n+ 56 DEVICE_CONTEXT context) {\n+ 57 return new CLStencilTable(stencilTable, context->GetContext());\n+ 58 }\n+ 59\n+ 60 template \n+61 static CLStencilTable *Create(\n+ 62 Far::LimitStencilTable const *limitStencilTable,\n+ 63 DEVICE_CONTEXT context) {\n+ 64 return new CLStencilTable(limitStencilTable, context->GetContext());\n+ 65 }\n 66\n-68 cl_mem GetPatchArrayBuffer() const { return _patchArrays; }\n- 69\n-71 cl_mem GetPatchIndexBuffer() const { return _indexBuffer; }\n+67 CLStencilTable(Far::StencilTable const *stencilTable,\n+ 68 cl_context clContext);\n+69 CLStencilTable(Far::LimitStencilTable const *limitStencilTable,\n+ 70 cl_context clContext);\n+71 ~CLStencilTable();\n 72\n-74 cl_mem GetPatchParamBuffer() const { return _patchParamBuffer; }\n- 75\n-77 cl_mem GetVaryingPatchArrayBuffer() const { return _varyingPatchArrays; }\n- 78\n-80 cl_mem GetVaryingPatchIndexBuffer() const { return _varyingIndexBuffer; }\n- 81\n-83 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }\n+ 73 // interfaces needed for CLComputeKernel\n+74 cl_mem GetSizesBuffer() const { return _sizes; }\n+75 cl_mem GetOffsetsBuffer() const { return _offsets; }\n+76 cl_mem GetIndicesBuffer() const { return _indices; }\n+77 cl_mem GetWeightsBuffer() const { return _weights; }\n+78 cl_mem GetDuWeightsBuffer() const { return _duWeights; }\n+79 cl_mem GetDvWeightsBuffer() const { return _dvWeights; }\n+80 cl_mem GetDuuWeightsBuffer() const { return _duuWeights; }\n+81 cl_mem GetDuvWeightsBuffer() const { return _duvWeights; }\n+82 cl_mem GetDvvWeightsBuffer() const { return _dvvWeights; }\n+83 int GetNumStencils() const { return _numStencils; }\n 84\n-86 cl_mem GetFVarPatchArrayBuffer(int fvarChannel = 0) const { return\n-_fvarPatchArrays[fvarChannel]; }\n- 87\n-89 cl_mem GetFVarPatchIndexBuffer(int fvarChannel = 0) const { return\n-_fvarIndexBuffers[fvarChannel]; }\n- 90\n-92 cl_mem GetFVarPatchParamBuffer(int fvarChannel = 0) const { return\n-_fvarParamBuffers[fvarChannel]; }\n- 93\n- 94protected:\n-95 CLPatchTable();\n- 96\n-97 bool allocate(Far::PatchTable const *patchTable, cl_context clContext);\n- 98\n-99 cl_mem _patchArrays;\n-100 cl_mem _indexBuffer;\n-101 cl_mem _patchParamBuffer;\n- 102\n-103 cl_mem _varyingPatchArrays;\n-104 cl_mem _varyingIndexBuffer;\n- 105\n-106 std::vector _fvarPatchArrays;\n-107 std::vector _fvarIndexBuffers;\n-108 std::vector _fvarParamBuffers;\n- 109\n- 110};\n- 111\n- 112} // end namespace Osd\n- 113\n- 114} // end namespace OPENSUBDIV_VERSION\n- 115using namespace OPENSUBDIV_VERSION;\n- 116\n- 117} // end namespace OpenSubdiv\n- 118\n- 119#endif // OPENSUBDIV3_OSD_CL_PATCH_TABLE_H\n+ 85private:\n+ 86 cl_mem _sizes;\n+ 87 cl_mem _offsets;\n+ 88 cl_mem _indices;\n+ 89 cl_mem _weights;\n+ 90 cl_mem _duWeights;\n+ 91 cl_mem _dvWeights;\n+ 92 cl_mem _duuWeights;\n+ 93 cl_mem _duvWeights;\n+ 94 cl_mem _dvvWeights;\n+ 95 int _numStencils;\n+ 96};\n+ 97\n+ 98// -------------------------------------------------------------------------\n+--\n+ 99\n+100class CLEvaluator {\n+ 101public:\n+102 typedef bool Instantiatable;\n+ 103\n+ 105 template \n+106 static CLEvaluator *Create(BufferDescriptor const &srcDesc,\n+ 107 BufferDescriptor const &dstDesc,\n+ 108 BufferDescriptor const &duDesc,\n+ 109 BufferDescriptor const &dvDesc,\n+ 110 DEVICE_CONTEXT deviceContext) {\n+ 111 return Create(srcDesc, dstDesc, duDesc, dvDesc,\n+ 112 deviceContext->GetContext(),\n+ 113 deviceContext->GetCommandQueue());\n+ 114 }\n+ 115\n+116 static CLEvaluator * Create(BufferDescriptor const &srcDesc,\n+ 117 BufferDescriptor const &dstDesc,\n+ 118 BufferDescriptor const &duDesc,\n+ 119 BufferDescriptor const &dvDesc,\n+ 120 cl_context clContext,\n+ 121 cl_command_queue clCommandQueue) {\n+ 122 CLEvaluator *instance = new CLEvaluator(clContext, clCommandQueue);\n+ 123 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc))\n+ 124 return instance;\n+ 125 delete instance;\n+ 126 return NULL;\n+ 127 }\n+ 128\n+ 130 template \n+131 static CLEvaluator *Create(BufferDescriptor const &srcDesc,\n+ 132 BufferDescriptor const &dstDesc,\n+ 133 BufferDescriptor const &duDesc,\n+ 134 BufferDescriptor const &dvDesc,\n+ 135 BufferDescriptor const &duuDesc,\n+ 136 BufferDescriptor const &duvDesc,\n+ 137 BufferDescriptor const &dvvDesc,\n+ 138 DEVICE_CONTEXT deviceContext) {\n+ 139 return Create(srcDesc, dstDesc, duDesc, dvDesc,\n+ 140 duuDesc, duvDesc, dvvDesc,\n+ 141 deviceContext->GetContext(),\n+ 142 deviceContext->GetCommandQueue());\n+ 143 }\n+ 144\n+145 static CLEvaluator * Create(BufferDescriptor const &srcDesc,\n+ 146 BufferDescriptor const &dstDesc,\n+ 147 BufferDescriptor const &duDesc,\n+ 148 BufferDescriptor const &dvDesc,\n+ 149 BufferDescriptor const &duuDesc,\n+ 150 BufferDescriptor const &duvDesc,\n+ 151 BufferDescriptor const &dvvDesc,\n+ 152 cl_context clContext,\n+ 153 cl_command_queue clCommandQueue) {\n+ 154 CLEvaluator *instance = new CLEvaluator(clContext, clCommandQueue);\n+ 155 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc,\n+ 156 duuDesc, duvDesc, dvvDesc))\n+ 157 return instance;\n+ 158 delete instance;\n+ 159 return NULL;\n+ 160 }\n+ 161\n+163 CLEvaluator(cl_context context, cl_command_queue queue);\n+ 164\n+166 ~CLEvaluator();\n+ 167\n+ 173\n+ 217 template \n+219 static bool EvalStencils(\n+ 220 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 221 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 222 STENCIL_TABLE const *stencilTable,\n+ 223 CLEvaluator const *instance,\n+ 224 DEVICE_CONTEXT deviceContext,\n+ 225 unsigned int numStartEvents=0,\n+ 226 const cl_event* startEvents=NULL,\n+ 227 cl_event* endEvent=NULL) {\n+ 228\n+ 229 if (instance) {\n+ 230 return instance->EvalStencils(srcBuffer, srcDesc,\n+ 231 dstBuffer, dstDesc,\n+ 232 stencilTable,\n+ 233 numStartEvents, startEvents, endEvent);\n+ 234 } else {\n+ 235 // Create an instance on demand (slow)\n+ 236 instance = Create(srcDesc, dstDesc,\n+ 237 BufferDescriptor(),\n+ 238 BufferDescriptor(),\n+ 239 deviceContext);\n+ 240 if (instance) {\n+ 241 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 242 dstBuffer, dstDesc,\n+ 243 stencilTable,\n+ 244 numStartEvents, startEvents, endEvent);\n+ 245 delete instance;\n+ 246 return r;\n+ 247 }\n+ 248 return false;\n+ 249 }\n+ 250 }\n+ 251\n+ 307 template \n+309 static bool EvalStencils(\n+ 310 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 311 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 312 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 313 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 314 STENCIL_TABLE const *stencilTable,\n+ 315 CLEvaluator const *instance,\n+ 316 DEVICE_CONTEXT deviceContext,\n+ 317 unsigned int numStartEvents=0,\n+ 318 const cl_event* startEvents=NULL,\n+ 319 cl_event* endEvent=NULL) {\n+ 320\n+ 321 if (instance) {\n+ 322 return instance->EvalStencils(srcBuffer, srcDesc,\n+ 323 dstBuffer, dstDesc,\n+ 324 duBuffer, duDesc,\n+ 325 dvBuffer, dvDesc,\n+ 326 stencilTable,\n+ 327 numStartEvents, startEvents, endEvent);\n+ 328 } else {\n+ 329 // Create an instance on demand (slow)\n+ 330 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,\n+ 331 deviceContext);\n+ 332 if (instance) {\n+ 333 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 334 dstBuffer, dstDesc,\n+ 335 duBuffer, duDesc,\n+ 336 dvBuffer, dvDesc,\n+ 337 stencilTable,\n+ 338 numStartEvents, startEvents, endEvent);\n+ 339 delete instance;\n+ 340 return r;\n+ 341 }\n+ 342 return false;\n+ 343 }\n+ 344 }\n+ 345\n+ 419 template \n+421 static bool EvalStencils(\n+ 422 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 423 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 424 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 425 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 426 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 427 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 428 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 429 STENCIL_TABLE const *stencilTable,\n+ 430 CLEvaluator const *instance,\n+ 431 DEVICE_CONTEXT deviceContext,\n+ 432 unsigned int numStartEvents=0,\n+ 433 const cl_event* startEvents=NULL,\n+ 434 cl_event* endEvent=NULL) {\n+ 435\n+ 436 if (instance) {\n+ 437 return instance->EvalStencils(srcBuffer, srcDesc,\n+ 438 dstBuffer, dstDesc,\n+ 439 duBuffer, duDesc,\n+ 440 dvBuffer, dvDesc,\n+ 441 duuBuffer, duuDesc,\n+ 442 duvBuffer, duvDesc,\n+ 443 dvvBuffer, dvvDesc,\n+ 444 stencilTable,\n+ 445 numStartEvents, startEvents, endEvent);\n+ 446 } else {\n+ 447 // Create an instance on demand (slow)\n+ 448 instance = Create(srcDesc, dstDesc,\n+ 449 duDesc, dvDesc,\n+ 450 duuDesc, duvDesc, dvvDesc,\n+ 451 deviceContext);\n+ 452 if (instance) {\n+ 453 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 454 dstBuffer, dstDesc,\n+ 455 duBuffer, duDesc,\n+ 456 dvBuffer, dvDesc,\n+ 457 duuBuffer, duuDesc,\n+ 458 duvBuffer, duvDesc,\n+ 459 dvvBuffer, dvvDesc,\n+ 460 stencilTable,\n+ 461 numStartEvents, startEvents, endEvent);\n+ 462 delete instance;\n+ 463 return r;\n+ 464 }\n+ 465 return false;\n+ 466 }\n+ 467 }\n+ 468\n+ 499 template \n+500 bool EvalStencils(\n+ 501 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 502 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 503 STENCIL_TABLE const *stencilTable,\n+ 504 unsigned int numStartEvents=0,\n+ 505 const cl_event* startEvents=NULL,\n+ 506 cl_event* endEvent=NULL) const {\n+ 507 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 508 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 509 stencilTable->GetSizesBuffer(),\n+ 510 stencilTable->GetOffsetsBuffer(),\n+ 511 stencilTable->GetIndicesBuffer(),\n+ 512 stencilTable->GetWeightsBuffer(),\n+ 513 0,\n+ 514 stencilTable->GetNumStencils(),\n+ 515 numStartEvents, startEvents, endEvent);\n+ 516 }\n+ 517\n+ 560 template \n+561 bool EvalStencils(\n+ 562 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 563 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 564 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 565 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 566 STENCIL_TABLE const *stencilTable,\n+ 567 unsigned int numStartEvents=0,\n+ 568 const cl_event* startEvents=NULL,\n+ 569 cl_event* endEvent=NULL) const {\n+ 570 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 571 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 572 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n+ 573 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n+ 574 stencilTable->GetSizesBuffer(),\n+ 575 stencilTable->GetOffsetsBuffer(),\n+ 576 stencilTable->GetIndicesBuffer(),\n+ 577 stencilTable->GetWeightsBuffer(),\n+ 578 stencilTable->GetDuWeightsBuffer(),\n+ 579 stencilTable->GetDvWeightsBuffer(),\n+ 580 0,\n+ 581 stencilTable->GetNumStencils(),\n+ 582 numStartEvents, startEvents, endEvent);\n+ 583 }\n+ 584\n+ 645 template \n+646 bool EvalStencils(\n+ 647 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 648 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 649 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 650 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 651 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 652 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 653 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 654 STENCIL_TABLE const *stencilTable,\n+ 655 unsigned int numStartEvents=0,\n+ 656 const cl_event* startEvents=NULL,\n+ 657 cl_event* endEvent=NULL) const {\n+ 658 return EvalStencils(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 659 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 660 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n+ 661 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n+ 662 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,\n+ 663 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,\n+ 664 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,\n+ 665 stencilTable->GetSizesBuffer(),\n+ 666 stencilTable->GetOffsetsBuffer(),\n+ 667 stencilTable->GetIndicesBuffer(),\n+ 668 stencilTable->GetWeightsBuffer(),\n+ 669 stencilTable->GetDuWeightsBuffer(),\n+ 670 stencilTable->GetDvWeightsBuffer(),\n+ 671 stencilTable->GetDuuWeightsBuffer(),\n+ 672 stencilTable->GetDuvWeightsBuffer(),\n+ 673 stencilTable->GetDvvWeightsBuffer(),\n+ 674 0,\n+ 675 stencilTable->GetNumStencils(),\n+ 676 numStartEvents, startEvents, endEvent);\n+ 677 }\n+ 678\n+681 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,\n+ 682 cl_mem dst, BufferDescriptor const &dstDesc,\n+ 683 cl_mem sizes,\n+ 684 cl_mem offsets,\n+ 685 cl_mem indices,\n+ 686 cl_mem weights,\n+ 687 int start,\n+ 688 int end,\n+ 689 unsigned int numStartEvents=0,\n+ 690 const cl_event* startEvents=NULL,\n+ 691 cl_event* endEvent=NULL) const;\n+ 692\n+741 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,\n+ 742 cl_mem dst, BufferDescriptor const &dstDesc,\n+ 743 cl_mem du, BufferDescriptor const &duDesc,\n+ 744 cl_mem dv, BufferDescriptor const &dvDesc,\n+ 745 cl_mem sizes,\n+ 746 cl_mem offsets,\n+ 747 cl_mem indices,\n+ 748 cl_mem weights,\n+ 749 cl_mem duWeights,\n+ 750 cl_mem dvWeights,\n+ 751 int start,\n+ 752 int end,\n+ 753 unsigned int numStartEvents=0,\n+ 754 const cl_event* startEvents=NULL,\n+ 755 cl_event* endEvent=NULL) const;\n+ 756\n+823 bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc,\n+ 824 cl_mem dst, BufferDescriptor const &dstDesc,\n+ 825 cl_mem du, BufferDescriptor const &duDesc,\n+ 826 cl_mem dv, BufferDescriptor const &dvDesc,\n+ 827 cl_mem duu, BufferDescriptor const &duuDesc,\n+ 828 cl_mem duv, BufferDescriptor const &duvDesc,\n+ 829 cl_mem dvv, BufferDescriptor const &dvvDesc,\n+ 830 cl_mem sizes,\n+ 831 cl_mem offsets,\n+ 832 cl_mem indices,\n+ 833 cl_mem weights,\n+ 834 cl_mem duWeights,\n+ 835 cl_mem dvWeights,\n+ 836 cl_mem duuWeights,\n+ 837 cl_mem duvWeights,\n+ 838 cl_mem dvvWeights,\n+ 839 int start,\n+ 840 int end,\n+ 841 unsigned int numStartEvents=0,\n+ 842 const cl_event* startEvents=NULL,\n+ 843 cl_event* endEvent=NULL) const;\n+ 844\n+ 850\n+ 899 template \n+902 static bool EvalPatches(\n+ 903 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 904 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 905 int numPatchCoords,\n+ 906 PATCHCOORD_BUFFER *patchCoords,\n+ 907 PATCH_TABLE *patchTable,\n+ 908 CLEvaluator const *instance,\n+ 909 DEVICE_CONTEXT deviceContext,\n+ 910 unsigned int numStartEvents=0,\n+ 911 const cl_event* startEvents=NULL,\n+ 912 cl_event* endEvent=NULL) {\n+ 913\n+ 914 if (instance) {\n+ 915 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 916 dstBuffer, dstDesc,\n+ 917 numPatchCoords, patchCoords,\n+ 918 patchTable,\n+ 919 numStartEvents, startEvents, endEvent);\n+ 920 } else {\n+ 921 // Create an instance on demand (slow)\n+ 922 (void)deviceContext; // unused\n+ 923 instance = Create(srcDesc, dstDesc,\n+ 924 BufferDescriptor(),\n+ 925 BufferDescriptor(),\n+ 926 deviceContext);\n+ 927 if (instance) {\n+ 928 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 929 dstBuffer, dstDesc,\n+ 930 numPatchCoords, patchCoords,\n+ 931 patchTable,\n+ 932 numStartEvents, startEvents, endEvent);\n+ 933 delete instance;\n+ 934 return r;\n+ 935 }\n+ 936 return false;\n+ 937 }\n+ 938 }\n+ 939\n+ 1000 template \n+1003 static bool EvalPatches(\n+ 1004 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1005 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1006 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1007 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1008 int numPatchCoords,\n+ 1009 PATCHCOORD_BUFFER *patchCoords,\n+ 1010 PATCH_TABLE *patchTable,\n+ 1011 CLEvaluator const *instance,\n+ 1012 DEVICE_CONTEXT deviceContext,\n+ 1013 unsigned int numStartEvents=0,\n+ 1014 const cl_event* startEvents=NULL,\n+ 1015 cl_event* endEvent=NULL) {\n+ 1016\n+ 1017 if (instance) {\n+ 1018 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 1019 dstBuffer, dstDesc,\n+ 1020 duBuffer, duDesc,\n+ 1021 dvBuffer, dvDesc,\n+ 1022 numPatchCoords, patchCoords,\n+ 1023 patchTable,\n+ 1024 numStartEvents, startEvents, endEvent);\n+ 1025 } else {\n+ 1026 // Create an instance on demand (slow)\n+ 1027 (void)deviceContext; // unused\n+ 1028 instance = Create(srcDesc, dstDesc, duDesc, dvDesc, deviceContext);\n+ 1029 if (instance) {\n+ 1030 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 1031 dstBuffer, dstDesc,\n+ 1032 duBuffer, duDesc,\n+ 1033 dvBuffer, dvDesc,\n+ 1034 numPatchCoords, patchCoords,\n+ 1035 patchTable,\n+ 1036 numStartEvents, startEvents, endEvent);\n+ 1037 delete instance;\n+ 1038 return r;\n+ 1039 }\n+ 1040 return false;\n+ 1041 }\n+ 1042 }\n+ 1043\n+ 1122 template \n+1125 static bool EvalPatches(\n+ 1126 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1127 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1128 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1129 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1130 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1131 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1132 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1133 int numPatchCoords,\n+ 1134 PATCHCOORD_BUFFER *patchCoords,\n+ 1135 PATCH_TABLE *patchTable,\n+ 1136 CLEvaluator const *instance,\n+ 1137 DEVICE_CONTEXT deviceContext,\n+ 1138 unsigned int numStartEvents=0,\n+ 1139 const cl_event* startEvents=NULL,\n+ 1140 cl_event* endEvent=NULL) {\n+ 1141\n+ 1142 if (instance) {\n+ 1143 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 1144 dstBuffer, dstDesc,\n+ 1145 duBuffer, duDesc,\n+ 1146 dvBuffer, dvDesc,\n+ 1147 duuBuffer, duuDesc,\n+ 1148 duvBuffer, duvDesc,\n+ 1149 dvvBuffer, dvvDesc,\n+ 1150 numPatchCoords, patchCoords,\n+ 1151 patchTable,\n+ 1152 numStartEvents, startEvents, endEvent);\n+ 1153 } else {\n+ 1154 // Create an instance on demand (slow)\n+ 1155 (void)deviceContext; // unused\n+ 1156 instance = Create(srcDesc, dstDesc,\n+ 1157 duDesc, dvDesc,\n+ 1158 duuDesc, duvDesc, dvvDesc,\n+ 1159 deviceContext);\n+ 1160 if (instance) {\n+ 1161 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 1162 dstBuffer, dstDesc,\n+ 1163 duBuffer, duDesc,\n+ 1164 dvBuffer, dvDesc,\n+ 1165 duuBuffer, duuDesc,\n+ 1166 duvBuffer, duvDesc,\n+ 1167 dvvBuffer, dvvDesc,\n+ 1168 numPatchCoords, patchCoords,\n+ 1169 patchTable,\n+ 1170 numStartEvents, startEvents, endEvent);\n+ 1171 delete instance;\n+ 1172 return r;\n+ 1173 }\n+ 1174 return false;\n+ 1175 }\n+ 1176 }\n+ 1177\n+ 1215 template \n+1217 bool EvalPatches(\n+ 1218 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1219 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1220 int numPatchCoords,\n+ 1221 PATCHCOORD_BUFFER *patchCoords,\n+ 1222 PATCH_TABLE *patchTable,\n+ 1223 unsigned int numStartEvents=0,\n+ 1224 const cl_event* startEvents=NULL,\n+ 1225 cl_event* endEvent=NULL) const {\n+ 1226\n+ 1227 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 1228 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 1229 0, BufferDescriptor(),\n+ 1230 0, BufferDescriptor(),\n+ 1231 numPatchCoords,\n+ 1232 patchCoords->BindCLBuffer(_clCommandQueue),\n+ 1233 patchTable->GetPatchArrayBuffer(),\n+ 1234 patchTable->GetPatchIndexBuffer(),\n+ 1235 patchTable->GetPatchParamBuffer(),\n+ 1236 numStartEvents, startEvents, endEvent);\n+ 1237 }\n+ 1238\n+ 1286 template \n+1288 bool EvalPatches(\n+ 1289 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1290 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1291 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1292 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1293 int numPatchCoords,\n+ 1294 PATCHCOORD_BUFFER *patchCoords,\n+ 1295 PATCH_TABLE *patchTable,\n+ 1296 unsigned int numStartEvents=0,\n+ 1297 const cl_event* startEvents=NULL,\n+ 1298 cl_event* endEvent=NULL) const {\n+ 1299\n+ 1300 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 1301 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 1302 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n+ 1303 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n+ 1304 numPatchCoords,\n+ 1305 patchCoords->BindCLBuffer(_clCommandQueue),\n+ 1306 patchTable->GetPatchArrayBuffer(),\n+ 1307 patchTable->GetPatchIndexBuffer(),\n+ 1308 patchTable->GetPatchParamBuffer(),\n+ 1309 numStartEvents, startEvents, endEvent);\n+ 1310 }\n+ 1311\n+ 1377 template \n+1379 bool EvalPatches(\n+ 1380 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1381 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1382 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1383 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1384 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1385 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1386 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1387 int numPatchCoords,\n+ 1388 PATCHCOORD_BUFFER *patchCoords,\n+ 1389 PATCH_TABLE *patchTable,\n+ 1390 unsigned int numStartEvents=0,\n+ 1391 const cl_event* startEvents=NULL,\n+ 1392 cl_event* endEvent=NULL) const {\n+ 1393\n+ 1394 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 1395 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 1396 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n+ 1397 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n+ 1398 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,\n+ 1399 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,\n+ 1400 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,\n+ 1401 numPatchCoords,\n+ 1402 patchCoords->BindCLBuffer(_clCommandQueue),\n+ 1403 patchTable->GetPatchArrayBuffer(),\n+ 1404 patchTable->GetPatchIndexBuffer(),\n+ 1405 patchTable->GetPatchParamBuffer(),\n+ 1406 numStartEvents, startEvents, endEvent);\n+ 1407 }\n+ 1408\n+1409 bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc,\n+ 1410 cl_mem dst, BufferDescriptor const &dstDesc,\n+ 1411 cl_mem du, BufferDescriptor const &duDesc,\n+ 1412 cl_mem dv, BufferDescriptor const &dvDesc,\n+ 1413 int numPatchCoords,\n+ 1414 cl_mem patchCoordsBuffer,\n+ 1415 cl_mem patchArrayBuffer,\n+ 1416 cl_mem patchIndexBuffer,\n+ 1417 cl_mem patchParamsBuffer,\n+ 1418 unsigned int numStartEvents=0,\n+ 1419 const cl_event* startEvents=NULL,\n+ 1420 cl_event* endEvent=NULL) const;\n+ 1421\n+1422 bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc,\n+ 1423 cl_mem dst, BufferDescriptor const &dstDesc,\n+ 1424 cl_mem du, BufferDescriptor const &duDesc,\n+ 1425 cl_mem dv, BufferDescriptor const &dvDesc,\n+ 1426 cl_mem duu, BufferDescriptor const &duuDesc,\n+ 1427 cl_mem duv, BufferDescriptor const &duvDesc,\n+ 1428 cl_mem dvv, BufferDescriptor const &dvvDesc,\n+ 1429 int numPatchCoords,\n+ 1430 cl_mem patchCoordsBuffer,\n+ 1431 cl_mem patchArrayBuffer,\n+ 1432 cl_mem patchIndexBuffer,\n+ 1433 cl_mem patchParamsBuffer,\n+ 1434 unsigned int numStartEvents=0,\n+ 1435 const cl_event* startEvents=NULL,\n+ 1436 cl_event* endEvent=NULL) const;\n+ 1437\n+ 1486 template \n+1489 static bool EvalPatchesVarying(\n+ 1490 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1491 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1492 int numPatchCoords,\n+ 1493 PATCHCOORD_BUFFER *patchCoords,\n+ 1494 PATCH_TABLE *patchTable,\n+ 1495 CLEvaluator const *instance,\n+ 1496 DEVICE_CONTEXT deviceContext,\n+ 1497 unsigned int numStartEvents=0,\n+ 1498 const cl_event* startEvents=NULL,\n+ 1499 cl_event* endEvent=NULL) {\n+ 1500\n+ 1501 if (instance) {\n+ 1502 return instance->EvalPatchesVarying(\n+ 1503 srcBuffer, srcDesc,\n+ 1504 dstBuffer, dstDesc,\n+ 1505 numPatchCoords, patchCoords,\n+ 1506 patchTable,\n+ 1507 numStartEvents, startEvents, endEvent);\n+ 1508 } else {\n+ 1509 // Create an instance on demand (slow)\n+ 1510 (void)deviceContext; // unused\n+ 1511 instance = Create(srcDesc, dstDesc,\n+ 1512 BufferDescriptor(),\n+ 1513 BufferDescriptor(),\n+ 1514 deviceContext);\n+ 1515 if (instance) {\n+ 1516 bool r = instance->EvalPatchesVarying(\n+ 1517 srcBuffer, srcDesc,\n+ 1518 dstBuffer, dstDesc,\n+ 1519 numPatchCoords, patchCoords,\n+ 1520 patchTable,\n+ 1521 numStartEvents, startEvents, endEvent);\n+ 1522 delete instance;\n+ 1523 return r;\n+ 1524 }\n+ 1525 return false;\n+ 1526 }\n+ 1527 }\n+ 1528\n+ 1566 template \n+1568 bool EvalPatchesVarying(\n+ 1569 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1570 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1571 int numPatchCoords,\n+ 1572 PATCHCOORD_BUFFER *patchCoords,\n+ 1573 PATCH_TABLE *patchTable,\n+ 1574 unsigned int numStartEvents=0,\n+ 1575 const cl_event* startEvents=NULL,\n+ 1576 cl_event* endEvent=NULL) const {\n+ 1577\n+ 1578 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 1579 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 1580 0, BufferDescriptor(),\n+ 1581 0, BufferDescriptor(),\n+ 1582 numPatchCoords,\n+ 1583 patchCoords->BindCLBuffer(_clCommandQueue),\n+ 1584 patchTable->GetVaryingPatchArrayBuffer(),\n+ 1585 patchTable->GetVaryingPatchIndexBuffer(),\n+ 1586 patchTable->GetPatchParamBuffer(),\n+ 1587 numStartEvents, startEvents, endEvent);\n+ 1588 }\n+ 1589\n+ 1650 template \n+1653 static bool EvalPatchesVarying(\n+ 1654 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1655 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1656 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1657 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1658 int numPatchCoords,\n+ 1659 PATCHCOORD_BUFFER *patchCoords,\n+ 1660 PATCH_TABLE *patchTable,\n+ 1661 CLEvaluator const *instance,\n+ 1662 DEVICE_CONTEXT deviceContext,\n+ 1663 unsigned int numStartEvents=0,\n+ 1664 const cl_event* startEvents=NULL,\n+ 1665 cl_event* endEvent=NULL) {\n+ 1666\n+ 1667 if (instance) {\n+ 1668 return instance->EvalPatchesVarying(\n+ 1669 srcBuffer, srcDesc,\n+ 1670 dstBuffer, dstDesc,\n+ 1671 duBuffer, duDesc,\n+ 1672 dvBuffer, dvDesc,\n+ 1673 numPatchCoords, patchCoords,\n+ 1674 patchTable,\n+ 1675 numStartEvents, startEvents, endEvent);\n+ 1676 } else {\n+ 1677 // Create an instance on demand (slow)\n+ 1678 (void)deviceContext; // unused\n+ 1679 instance = Create(srcDesc, dstDesc,\n+ 1680 duDesc, dvDesc,\n+ 1681 deviceContext);\n+ 1682 if (instance) {\n+ 1683 bool r = instance->EvalPatchesVarying(\n+ 1684 srcBuffer, srcDesc,\n+ 1685 dstBuffer, dstDesc,\n+ 1686 duBuffer, duDesc,\n+ 1687 dvBuffer, dvDesc,\n+ 1688 numPatchCoords, patchCoords,\n+ 1689 patchTable,\n+ 1690 numStartEvents, startEvents, endEvent);\n+ 1691 delete instance;\n+ 1692 return r;\n+ 1693 }\n+ 1694 return false;\n+ 1695 }\n+ 1696 }\n+ 1697\n+ 1747 template \n+1749 bool EvalPatchesVarying(\n+ 1750 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1751 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1752 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1753 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1754 int numPatchCoords,\n+ 1755 PATCHCOORD_BUFFER *patchCoords,\n+ 1756 PATCH_TABLE *patchTable,\n+ 1757 unsigned int numStartEvents=0,\n+ 1758 const cl_event* startEvents=NULL,\n+ 1759 cl_event* endEvent=NULL) const {\n+ 1760\n+ 1761 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 1762 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 1763 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n+ 1764 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n+ 1765 numPatchCoords,\n+ 1766 patchCoords->BindCLBuffer(_clCommandQueue),\n+ 1767 patchTable->GetVaryingPatchArrayBuffer(),\n+ 1768 patchTable->GetVaryingPatchIndexBuffer(),\n+ 1769 patchTable->GetPatchParamBuffer(),\n+ 1770 numStartEvents, startEvents, endEvent);\n+ 1771 }\n+ 1772\n+ 1851 template \n+1854 static bool EvalPatchesVarying(\n+ 1855 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1856 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1857 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1858 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1859 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1860 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1861 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1862 int numPatchCoords,\n+ 1863 PATCHCOORD_BUFFER *patchCoords,\n+ 1864 PATCH_TABLE *patchTable,\n+ 1865 CLEvaluator const *instance,\n+ 1866 DEVICE_CONTEXT deviceContext,\n+ 1867 unsigned int numStartEvents=0,\n+ 1868 const cl_event* startEvents=NULL,\n+ 1869 cl_event* endEvent=NULL) {\n+ 1870\n+ 1871 if (instance) {\n+ 1872 return instance->EvalPatchesVarying(\n+ 1873 srcBuffer, srcDesc,\n+ 1874 dstBuffer, dstDesc,\n+ 1875 duBuffer, duDesc,\n+ 1876 dvBuffer, dvDesc,\n+ 1877 duuBuffer, duuDesc,\n+ 1878 duvBuffer, duvDesc,\n+ 1879 dvvBuffer, dvvDesc,\n+ 1880 numPatchCoords, patchCoords,\n+ 1881 patchTable,\n+ 1882 numStartEvents, startEvents, endEvent);\n+ 1883 } else {\n+ 1884 // Create an instance on demand (slow)\n+ 1885 (void)deviceContext; // unused\n+ 1886 instance = Create(srcDesc, dstDesc,\n+ 1887 duDesc, dvDesc,\n+ 1888 duuDesc, duvDesc, dvvDesc,\n+ 1889 deviceContext);\n+ 1890 if (instance) {\n+ 1891 bool r = instance->EvalPatchesVarying(\n+ 1892 srcBuffer, srcDesc,\n+ 1893 dstBuffer, dstDesc,\n+ 1894 duBuffer, duDesc,\n+ 1895 dvBuffer, dvDesc,\n+ 1896 duuBuffer, duuDesc,\n+ 1897 duvBuffer, duvDesc,\n+ 1898 dvvBuffer, dvvDesc,\n+ 1899 numPatchCoords, patchCoords,\n+ 1900 patchTable,\n+ 1901 numStartEvents, startEvents, endEvent);\n+ 1902 delete instance;\n+ 1903 return r;\n+ 1904 }\n+ 1905 return false;\n+ 1906 }\n+ 1907 }\n+ 1908\n+ 1976 template \n+1978 bool EvalPatchesVarying(\n+ 1979 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1980 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1981 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1982 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1983 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1984 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1985 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1986 int numPatchCoords,\n+ 1987 PATCHCOORD_BUFFER *patchCoords,\n+ 1988 PATCH_TABLE *patchTable,\n+ 1989 unsigned int numStartEvents=0,\n+ 1990 const cl_event* startEvents=NULL,\n+ 1991 cl_event* endEvent=NULL) const {\n+ 1992\n+ 1993 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 1994 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 1995 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n+ 1996 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n+ 1997 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,\n+ 1998 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,\n+ 1999 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,\n+ 2000 numPatchCoords,\n+ 2001 patchCoords->BindCLBuffer(_clCommandQueue),\n+ 2002 patchTable->GetVaryingPatchArrayBuffer(),\n+ 2003 patchTable->GetVaryingPatchIndexBuffer(),\n+ 2004 patchTable->GetPatchParamBuffer(),\n+ 2005 numStartEvents, startEvents, endEvent);\n+ 2006 }\n+ 2007\n+ 2058 template \n+2061 static bool EvalPatchesFaceVarying(\n+ 2062 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 2063 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 2064 int numPatchCoords,\n+ 2065 PATCHCOORD_BUFFER *patchCoords,\n+ 2066 PATCH_TABLE *patchTable,\n+ 2067 int fvarChannel,\n+ 2068 CLEvaluator const *instance,\n+ 2069 DEVICE_CONTEXT deviceContext,\n+ 2070 unsigned int numStartEvents=0,\n+ 2071 const cl_event* startEvents=NULL,\n+ 2072 cl_event* endEvent=NULL) {\n+ 2073\n+ 2074 if (instance) {\n+ 2075 return instance->EvalPatchesFaceVarying(\n+ 2076 srcBuffer, srcDesc,\n+ 2077 dstBuffer, dstDesc,\n+ 2078 numPatchCoords, patchCoords,\n+ 2079 patchTable, fvarChannel,\n+ 2080 numStartEvents, startEvents, endEvent);\n+ 2081 } else {\n+ 2082 // Create an instance on demand (slow)\n+ 2083 (void)deviceContext; // unused\n+ 2084 instance = Create(srcDesc, dstDesc,\n+ 2085 BufferDescriptor(),\n+ 2086 BufferDescriptor(),\n+ 2087 deviceContext);\n+ 2088 if (instance) {\n+ 2089 bool r = instance->EvalPatchesFaceVarying(\n+ 2090 srcBuffer, srcDesc,\n+ 2091 dstBuffer, dstDesc,\n+ 2092 numPatchCoords, patchCoords,\n+ 2093 patchTable, fvarChannel,\n+ 2094 numStartEvents, startEvents, endEvent);\n+ 2095 delete instance;\n+ 2096 return r;\n+ 2097 }\n+ 2098 return false;\n+ 2099 }\n+ 2100 }\n+ 2101\n+ 2141 template \n+2143 bool EvalPatchesFaceVarying(\n+ 2144 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 2145 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 2146 int numPatchCoords,\n+ 2147 PATCHCOORD_BUFFER *patchCoords,\n+ 2148 PATCH_TABLE *patchTable,\n+ 2149 int fvarChannel = 0,\n+ 2150 unsigned int numStartEvents=0,\n+ 2151 const cl_event* startEvents=NULL,\n+ 2152 cl_event* endEvent=NULL) const {\n+ 2153\n+ 2154 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 2155 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 2156 0, BufferDescriptor(),\n+ 2157 0, BufferDescriptor(),\n+ 2158 numPatchCoords,\n+ 2159 patchCoords->BindCLBuffer(_clCommandQueue),\n+ 2160 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 2161 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 2162 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n+ 2163 numStartEvents, startEvents, endEvent);\n+ 2164 }\n+ 2165\n+ 2228 template \n+2231 static bool EvalPatchesFaceVarying(\n+ 2232 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 2233 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 2234 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 2235 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 2236 int numPatchCoords,\n+ 2237 PATCHCOORD_BUFFER *patchCoords,\n+ 2238 PATCH_TABLE *patchTable,\n+ 2239 int fvarChannel,\n+ 2240 CLEvaluator const *instance,\n+ 2241 DEVICE_CONTEXT deviceContext,\n+ 2242 unsigned int numStartEvents=0,\n+ 2243 const cl_event* startEvents=NULL,\n+ 2244 cl_event* endEvent=NULL) {\n+ 2245\n+ 2246 if (instance) {\n+ 2247 return instance->EvalPatchesFaceVarying(\n+ 2248 srcBuffer, srcDesc,\n+ 2249 dstBuffer, dstDesc,\n+ 2250 duBuffer, duDesc,\n+ 2251 dvBuffer, dvDesc,\n+ 2252 numPatchCoords, patchCoords,\n+ 2253 patchTable, fvarChannel,\n+ 2254 numStartEvents, startEvents, endEvent);\n+ 2255 } else {\n+ 2256 // Create an instance on demand (slow)\n+ 2257 (void)deviceContext; // unused\n+ 2258 instance = Create(srcDesc, dstDesc,\n+ 2259 duDesc, dvDesc, deviceContext);\n+ 2260 if (instance) {\n+ 2261 bool r = instance->EvalPatchesFaceVarying(\n+ 2262 srcBuffer, srcDesc,\n+ 2263 dstBuffer, dstDesc,\n+ 2264 duBuffer, duDesc,\n+ 2265 dvBuffer, dvDesc,\n+ 2266 numPatchCoords, patchCoords,\n+ 2267 patchTable, fvarChannel,\n+ 2268 numStartEvents, startEvents, endEvent);\n+ 2269 delete instance;\n+ 2270 return r;\n+ 2271 }\n+ 2272 return false;\n+ 2273 }\n+ 2274 }\n+ 2275\n+ 2327 template \n+2329 bool EvalPatchesFaceVarying(\n+ 2330 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 2331 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 2332 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 2333 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 2334 int numPatchCoords,\n+ 2335 PATCHCOORD_BUFFER *patchCoords,\n+ 2336 PATCH_TABLE *patchTable,\n+ 2337 int fvarChannel = 0,\n+ 2338 unsigned int numStartEvents=0,\n+ 2339 const cl_event* startEvents=NULL,\n+ 2340 cl_event* endEvent=NULL) const {\n+ 2341\n+ 2342 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 2343 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 2344 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n+ 2345 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n+ 2346 numPatchCoords,\n+ 2347 patchCoords->BindCLBuffer(_clCommandQueue),\n+ 2348 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 2349 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 2350 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n+ 2351 numStartEvents, startEvents, endEvent);\n+ 2352 }\n+ 2353\n+ 2434 template \n+2437 static bool EvalPatchesFaceVarying(\n+ 2438 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 2439 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 2440 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 2441 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 2442 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 2443 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 2444 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 2445 int numPatchCoords,\n+ 2446 PATCHCOORD_BUFFER *patchCoords,\n+ 2447 PATCH_TABLE *patchTable,\n+ 2448 int fvarChannel,\n+ 2449 CLEvaluator const *instance,\n+ 2450 DEVICE_CONTEXT deviceContext,\n+ 2451 unsigned int numStartEvents=0,\n+ 2452 const cl_event* startEvents=NULL,\n+ 2453 cl_event* endEvent=NULL) {\n+ 2454\n+ 2455 if (instance) {\n+ 2456 return instance->EvalPatchesFaceVarying(\n+ 2457 srcBuffer, srcDesc,\n+ 2458 dstBuffer, dstDesc,\n+ 2459 duBuffer, duDesc,\n+ 2460 dvBuffer, dvDesc,\n+ 2461 duuBuffer, duuDesc,\n+ 2462 duvBuffer, duvDesc,\n+ 2463 dvvBuffer, dvvDesc,\n+ 2464 numPatchCoords, patchCoords,\n+ 2465 patchTable, fvarChannel,\n+ 2466 numStartEvents, startEvents, endEvent);\n+ 2467 } else {\n+ 2468 // Create an instance on demand (slow)\n+ 2469 (void)deviceContext; // unused\n+ 2470 instance = Create(srcDesc, dstDesc,\n+ 2471 duDesc, dvDesc,\n+ 2472 duuDesc, duvDesc, dvvDesc,\n+ 2473 deviceContext);\n+ 2474 if (instance) {\n+ 2475 bool r = instance->EvalPatchesFaceVarying(\n+ 2476 srcBuffer, srcDesc,\n+ 2477 dstBuffer, dstDesc,\n+ 2478 duBuffer, duDesc,\n+ 2479 dvBuffer, dvDesc,\n+ 2480 duuBuffer, duuDesc,\n+ 2481 duvBuffer, duvDesc,\n+ 2482 dvvBuffer, dvvDesc,\n+ 2483 numPatchCoords, patchCoords,\n+ 2484 patchTable, fvarChannel,\n+ 2485 numStartEvents, startEvents, endEvent);\n+ 2486 delete instance;\n+ 2487 return r;\n+ 2488 }\n+ 2489 return false;\n+ 2490 }\n+ 2491 }\n+ 2492\n+ 2562 template \n+2564 bool EvalPatchesFaceVarying(\n+ 2565 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 2566 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 2567 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 2568 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 2569 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 2570 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 2571 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 2572 int numPatchCoords,\n+ 2573 PATCHCOORD_BUFFER *patchCoords,\n+ 2574 PATCH_TABLE *patchTable,\n+ 2575 int fvarChannel = 0,\n+ 2576 unsigned int numStartEvents=0,\n+ 2577 const cl_event* startEvents=NULL,\n+ 2578 cl_event* endEvent=NULL) const {\n+ 2579\n+ 2580 return EvalPatches(srcBuffer->BindCLBuffer(_clCommandQueue), srcDesc,\n+ 2581 dstBuffer->BindCLBuffer(_clCommandQueue), dstDesc,\n+ 2582 duBuffer->BindCLBuffer(_clCommandQueue), duDesc,\n+ 2583 dvBuffer->BindCLBuffer(_clCommandQueue), dvDesc,\n+ 2584 duuBuffer->BindCLBuffer(_clCommandQueue), duuDesc,\n+ 2585 duvBuffer->BindCLBuffer(_clCommandQueue), duvDesc,\n+ 2586 dvvBuffer->BindCLBuffer(_clCommandQueue), dvvDesc,\n+ 2587 numPatchCoords,\n+ 2588 patchCoords->BindCLBuffer(_clCommandQueue),\n+ 2589 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 2590 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 2591 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n+ 2592 numStartEvents, startEvents, endEvent);\n+ 2593 }\n+ 2594\n+ 2600\n+2603 bool Compile(BufferDescriptor const &srcDesc,\n+ 2604 BufferDescriptor const &dstDesc,\n+ 2605 BufferDescriptor const &duDesc = BufferDescriptor(),\n+ 2606 BufferDescriptor const &dvDesc = BufferDescriptor(),\n+ 2607 BufferDescriptor const &duuDesc = BufferDescriptor(),\n+ 2608 BufferDescriptor const &duvDesc = BufferDescriptor(),\n+ 2609 BufferDescriptor const &dvvDesc = BufferDescriptor());\n+ 2610\n+ 2612 template \n+2613 static void Synchronize(DEVICE_CONTEXT deviceContext) {\n+ 2614 Synchronize(deviceContext->GetCommandQueue());\n+ 2615 }\n+ 2616\n+2617 static void Synchronize(cl_command_queue queue);\n+ 2618\n+ 2619private:\n+ 2620 cl_context _clContext;\n+ 2621 cl_command_queue _clCommandQueue;\n+ 2622 cl_program _program;\n+ 2623 cl_kernel _stencilKernel;\n+ 2624 cl_kernel _stencilDerivKernel;\n+ 2625 cl_kernel _patchKernel;\n+ 2626};\n+ 2627\n+ 2628} // end namespace Osd\n+ 2629\n+ 2630} // end namespace OPENSUBDIV_VERSION\n+ 2631using namespace OPENSUBDIV_VERSION;\n+ 2632\n+ 2633} // end namespace OpenSubdiv\n+ 2634\n+ 2635\n+ 2636#endif // OPENSUBDIV3_OSD_CL_EVALUATOR_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n-Container for arrays of parametric patches.\n-Definition: patchTable.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable\n-CL patch table.\n-Definition: clPatchTable.h:52\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_fvarPatchArrays\n-std::vector< cl_mem > _fvarPatchArrays\n-Definition: clPatchTable.h:106\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::~CLPatchTable\n-~CLPatchTable()\n-Destructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_patchParamBuffer\n-cl_mem _patchParamBuffer\n-Definition: clPatchTable.h:101\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchParamBuffer\n-cl_mem GetPatchParamBuffer() const\n-Returns the CL memory of the array of Osd::PatchParam buffer.\n-Definition: clPatchTable.h:74\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_indexBuffer\n-cl_mem _indexBuffer\n-Definition: clPatchTable.h:100\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchArrayBuffer\n-cl_mem GetFVarPatchArrayBuffer(int fvarChannel=0) const\n-Returns the CL memory of the array of Osd::PatchArray buffer.\n-Definition: clPatchTable.h:86\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchArrayBuffer\n-cl_mem GetPatchArrayBuffer() const\n-Returns the CL memory of the array of Osd::PatchArray buffer.\n-Definition: clPatchTable.h:68\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::Create\n-static CLPatchTable * Create(Far::PatchTable const *patchTable, cl_context\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n+Stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:273\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n+Limit stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:583\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n+BufferDescriptor is a struct which describes buffer elements in interleaved\n+data buffers....\n+Definition: bufferDescriptor.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable\n+OpenCL stencil table.\n+Definition: clEvaluator.h:52\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetDvvWeightsBuffer\n+cl_mem GetDvvWeightsBuffer() const\n+Definition: clEvaluator.h:82\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::CLStencilTable\n+CLStencilTable(Far::StencilTable const *stencilTable, cl_context clContext)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetOffsetsBuffer\n+cl_mem GetOffsetsBuffer() const\n+Definition: clEvaluator.h:75\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::Create\n+static CLStencilTable * Create(Far::LimitStencilTable const *limitStencilTable,\n+DEVICE_CONTEXT context)\n+Definition: clEvaluator.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetIndicesBuffer\n+cl_mem GetIndicesBuffer() const\n+Definition: clEvaluator.h:76\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetNumStencils\n+int GetNumStencils() const\n+Definition: clEvaluator.h:83\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetSizesBuffer\n+cl_mem GetSizesBuffer() const\n+Definition: clEvaluator.h:74\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetDuuWeightsBuffer\n+cl_mem GetDuuWeightsBuffer() const\n+Definition: clEvaluator.h:80\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetDvWeightsBuffer\n+cl_mem GetDvWeightsBuffer() const\n+Definition: clEvaluator.h:79\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetDuvWeightsBuffer\n+cl_mem GetDuvWeightsBuffer() const\n+Definition: clEvaluator.h:81\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetDuWeightsBuffer\n+cl_mem GetDuWeightsBuffer() const\n+Definition: clEvaluator.h:78\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::~CLStencilTable\n+~CLStencilTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::Create\n+static CLStencilTable * Create(Far::StencilTable const *stencilTable,\n+DEVICE_CONTEXT context)\n+Definition: clEvaluator.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::CLStencilTable\n+CLStencilTable(Far::LimitStencilTable const *limitStencilTable, cl_context\n clContext)\n-Creator. Returns NULL if error.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_patchArrays\n-cl_mem _patchArrays\n-Definition: clPatchTable.h:99\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchIndexBuffer\n-cl_mem GetFVarPatchIndexBuffer(int fvarChannel=0) const\n-Returns the CL memory of the face-varying control vertices.\n-Definition: clPatchTable.h:89\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_varyingIndexBuffer\n-cl_mem _varyingIndexBuffer\n-Definition: clPatchTable.h:104\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetVaryingPatchIndexBuffer\n-cl_mem GetVaryingPatchIndexBuffer() const\n-Returns the CL memory of the varying control vertices.\n-Definition: clPatchTable.h:80\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetNumFVarChannels\n-int GetNumFVarChannels() const\n-Returns the number of face-varying channel buffers.\n-Definition: clPatchTable.h:83\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetPatchIndexBuffer\n-cl_mem GetPatchIndexBuffer() const\n-Returns the CL memory of the patch control vertices.\n-Definition: clPatchTable.h:71\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::Create\n-static CLPatchTable * Create(Far::PatchTable const *patchTable, DEVICE_CONTEXT\n-context)\n-Definition: clPatchTable.h:59\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetFVarPatchParamBuffer\n-cl_mem GetFVarPatchParamBuffer(int fvarChannel=0) const\n-Returns the CL memory of the array of Osd::PatchParam buffer.\n-Definition: clPatchTable.h:92\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::CLPatchTable\n-CLPatchTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::allocate\n-bool allocate(Far::PatchTable const *patchTable, cl_context clContext)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_fvarParamBuffers\n-std::vector< cl_mem > _fvarParamBuffers\n-Definition: clPatchTable.h:108\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_fvarIndexBuffers\n-std::vector< cl_mem > _fvarIndexBuffers\n-Definition: clPatchTable.h:107\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::_varyingPatchArrays\n-cl_mem _varyingPatchArrays\n-Definition: clPatchTable.h:103\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::GetVaryingPatchArrayBuffer\n-cl_mem GetVaryingPatchArrayBuffer() const\n-Returns the CL memory of the array of Osd::PatchArray buffer.\n-Definition: clPatchTable.h:77\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::GetWeightsBuffer\n+cl_mem GetWeightsBuffer() const\n+Definition: clEvaluator.h:77\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator\n+Definition: clEvaluator.h:100\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, CLEvaluator\n+const *instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0,\n+const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n+Generic static stencil function. This function has a same signature as other\n+device kernels have so t...\n+Definition: clEvaluator.h:309\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Synchronize\n+static void Synchronize(cl_command_queue queue)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::~CLEvaluator\n+~CLEvaluator()\n+Destructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n+bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst,\n+BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc,\n+cl_mem dv, BufferDescriptor const &dvDesc, cl_mem sizes, cl_mem offsets, cl_mem\n+indices, cl_mem weights, cl_mem duWeights, cl_mem dvWeights, int start, int\n+end, unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n+*endEvent=NULL) const\n+Dispatch the CL compute kernel asynchronously. returns false if the kernel\n+hasn't been compiled yet.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n+const *stencilTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext,\n+unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n+*endEvent=NULL)\n+Generic static stencil function. This function has a same signature as other\n+device kernels have so t...\n+Definition: clEvaluator.h:219\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create\n+static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor\n+const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc,\n+cl_context clContext, cl_command_queue clCommandQueue)\n+Definition: clEvaluator.h:116\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CLEvaluator const\n+*instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const\n+cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:2231\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n+bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst,\n+BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc,\n+cl_mem dv, BufferDescriptor const &dvDesc, int numPatchCoords, cl_mem\n+patchCoordsBuffer, cl_mem patchArrayBuffer, cl_mem patchIndexBuffer, cl_mem\n+patchParamsBuffer, unsigned int numStartEvents=0, const cl_event\n+*startEvents=NULL, cl_event *endEvent=NULL) const\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n+bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const\n+cl_event *startEvents=NULL, cl_event *endEvent=NULL) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:1978\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n+bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, STENCIL_TABLE const *stencilTable, unsigned int numStartEvents=0,\n+const cl_event *startEvents=NULL, cl_event *endEvent=NULL) const\n+Generic stencil function.\n+Definition: clEvaluator.h:561\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n+bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, unsigned int numStartEvents=0, const\n+cl_event *startEvents=NULL, cl_event *endEvent=NULL) const\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: clEvaluator.h:1379\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Instantiatable\n+bool Instantiatable\n+Definition: clEvaluator.h:102\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create\n+static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor\n+const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc,\n+DEVICE_CONTEXT deviceContext)\n+Generic creator template.\n+Definition: clEvaluator.h:106\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n+fvarChannel, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext,\n+unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n+*endEvent=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:2061\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create\n+static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor\n+const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc,\n+BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc,\n+BufferDescriptor const &dvvDesc, DEVICE_CONTEXT deviceContext)\n+Generic creator template.\n+Definition: clEvaluator.h:131\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n+bool EvalPatches(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst,\n+BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc,\n+cl_mem dv, BufferDescriptor const &dvDesc, cl_mem duu, BufferDescriptor const\n+&duuDesc, cl_mem duv, BufferDescriptor const &duvDesc, cl_mem dvv,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, cl_mem patchCoordsBuffer,\n+cl_mem patchArrayBuffer, cl_mem patchIndexBuffer, cl_mem patchParamsBuffer,\n+unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n+*endEvent=NULL) const\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n+bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst,\n+BufferDescriptor const &dstDesc, cl_mem du, BufferDescriptor const &duDesc,\n+cl_mem dv, BufferDescriptor const &dvDesc, cl_mem duu, BufferDescriptor const\n+&duuDesc, cl_mem duv, BufferDescriptor const &duvDesc, cl_mem dvv,\n+BufferDescriptor const &dvvDesc, cl_mem sizes, cl_mem offsets, cl_mem indices,\n+cl_mem weights, cl_mem duWeights, cl_mem dvWeights, cl_mem duuWeights, cl_mem\n+duvWeights, cl_mem dvvWeights, int start, int end, unsigned int\n+numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n+const\n+Dispatch the CL compute kernel asynchronously. returns false if the kernel\n+hasn't been compiled yet.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n+bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0,\n+unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n+*endEvent=NULL) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:2564\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n+bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int\n+numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n+const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:1217\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n+bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, unsigned\n+int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n+*endEvent=NULL) const\n+Generic stencil function.\n+Definition: clEvaluator.h:646\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance,\n+DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event\n+*startEvents=NULL, cl_event *endEvent=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:1653\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Synchronize\n+static void Synchronize(DEVICE_CONTEXT deviceContext)\n+Wait the OpenCL kernels finish.\n+Definition: clEvaluator.h:2613\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, CLEvaluator const *instance,\n+DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const cl_event\n+*startEvents=NULL, cl_event *endEvent=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:1125\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n+CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int\n+numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n+Generic static stencil function. This function has a same signature as other\n+device kernels have so t...\n+Definition: clEvaluator.h:421\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n+bool EvalStencils(cl_mem src, BufferDescriptor const &srcDesc, cl_mem dst,\n+BufferDescriptor const &dstDesc, cl_mem sizes, cl_mem offsets, cl_mem indices,\n+cl_mem weights, int start, int end, unsigned int numStartEvents=0, const\n+cl_event *startEvents=NULL, cl_event *endEvent=NULL) const\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n+CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int\n+numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:1489\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n+bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0, unsigned int\n+numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n+const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:2329\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n+CLEvaluator const *instance, DEVICE_CONTEXT deviceContext, unsigned int\n+numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:2437\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Create\n+static CLEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor\n+const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc,\n+BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc,\n+BufferDescriptor const &dvvDesc, cl_context clContext, cl_command_queue\n+clCommandQueue)\n+Definition: clEvaluator.h:145\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable, CLEvaluator const *instance, DEVICE_CONTEXT deviceContext,\n+unsigned int numStartEvents=0, const cl_event *startEvents=NULL, cl_event\n+*endEvent=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:1003\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n+bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL,\n+cl_event *endEvent=NULL) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:1749\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const\n+*instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const\n+cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:902\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n+bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, unsigned int\n+numStartEvents=0, const cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n+const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:1568\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CLEvaluator const\n+*instance, DEVICE_CONTEXT deviceContext, unsigned int numStartEvents=0, const\n+cl_event *startEvents=NULL, cl_event *endEvent=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:1854\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalStencils\n+bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const\n+*stencilTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL,\n+cl_event *endEvent=NULL) const\n+Generic stencil function.\n+Definition: clEvaluator.h:500\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatches\n+bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable, unsigned int numStartEvents=0, const cl_event *startEvents=NULL,\n+cl_event *endEvent=NULL) const\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: clEvaluator.h:1288\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::CLEvaluator\n+CLEvaluator(cl_context context, cl_command_queue queue)\n+Constructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::Compile\n+bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc,\n+BufferDescriptor const &duDesc=BufferDescriptor(), BufferDescriptor const\n+&dvDesc=BufferDescriptor(), BufferDescriptor const &duuDesc=BufferDescriptor(),\n+BufferDescriptor const &duvDesc=BufferDescriptor(), BufferDescriptor const\n+&dvvDesc=BufferDescriptor())\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::EvalPatchesFaceVarying\n+bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n+fvarChannel=0, unsigned int numStartEvents=0, const cl_event *startEvents=NULL,\n+cl_event *endEvent=NULL) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: clEvaluator.h:2143\n * opensubdiv\n * osd\n- * clPatchTable.h\n+ * clEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00809.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00809.html", "unified_diff": "@@ -87,15 +87,15 @@\n
\n Classes |\n Namespaces
\n
mtlPatchShaderSource.h File Reference
\n \n
\n
#import "../version.h"
\n-#import "../far/patchDescriptor.h"
\n+#import "../far/patchDescriptor.h"
\n #import <string>
\n
\n

Go to the source code of this file.

\n \n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00809_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00809_source.html", "unified_diff": "@@ -151,15 +151,15 @@\n
66} // end namespace OPENSUBDIV_VERSION
\n
67using namespace OPENSUBDIV_VERSION;
\n
68
\n
69} // end namespace OpenSubdiv
\n
70
\n
71#endif // OPENSUBDIV3_OSD_MTL_PATCH_SHADER_SOURCE
\n \n-\n+\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchShaderSource::GetVertexShaderSource
static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type)
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchShaderSource::GetHullShaderSource
static std::string GetHullShaderSource(Far::PatchDescriptor::Type type, Far::PatchDescriptor::Type fvarType)
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchShaderSource::GetDomainShaderSource
static std::string GetDomainShaderSource(Far::PatchDescriptor::Type type)
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchShaderSource::GetVertexShaderSource
static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type, Far::PatchDescriptor::Type fvarType)
Deprecated.
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchShaderSource::GetDomainShaderSource
static std::string GetDomainShaderSource(Far::PatchDescriptor::Type type, Far::PatchDescriptor::Type fvarType)
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchShaderSource::GetHullShaderSource
static std::string GetHullShaderSource(Far::PatchDescriptor::Type type)
\n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00812.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00812.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/mtlMesh.h File Reference\n+OpenSubdiv: opensubdiv/osd/mtlPatchTable.h File Reference\n \n \n \n \n \n \n \n@@ -81,43 +81,47 @@\n \n \n \n
\n
\n-Namespaces |\n-Typedefs
\n-
mtlMesh.h File Reference
\n+Classes |\n+Namespaces
\n+
mtlPatchTable.h File Reference
\n \n
\n
#include "../version.h"
\n-#include "../osd/mesh.h"
\n-#include "../osd/mtlPatchTable.h"
\n+#include "../far/patchDescriptor.h"
\n+#include "../osd/nonCopyable.h"
\n+#include "../osd/types.h"
\n+#include "../osd/mtlCommon.h"
\n
\n

Go to the source code of this file.

\n

\n Classes

class  MTLPatchShaderSource
\n+\n+\n+\n+

\n+Classes

class  MTLPatchTable
 
\n \n \n \n \n \n+\n+\n \n \n-

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n-\n-\n-\n

\n-Typedefs

typedef MeshInterface< MTLPatchTable > MTLMeshInterface
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,27 +4,31 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-Namespaces | Typedefs\n-mtlMesh.h File Reference\n+Classes | Namespaces\n+mtlPatchTable.h File Reference\n #include \"../version.h\"\n-#include \"../osd/mesh.h\"\n-#include \"../osd/mtlPatchTable.h\"\n+#include \"../far/patchDescriptor.h\"\n+#include \"../osd/nonCopyable.h\"\n+#include \"../osd/types.h\"\n+#include \"../osd/mtlCommon.h\"\n Go_to_the_source_code_of_this_file.\n+ Classes\n+class \u00a0MTLPatchTable\n+\u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n- Typedefs\n-typedef MeshInterface< MTLPatchTable >\u00a0MTLMeshInterface\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * mtlMesh.h\n+ * mtlPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00812.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00812.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00812 = [\n- [\"MTLMeshInterface\", \"a00812.html#aa41e5b43fe46eeb0f97ff6ebdb1935aa\", null]\n+ [\"MTLPatchTable\", \"a01297.html\", \"a01297\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00812_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00812_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/mtlMesh.h Source File\n+OpenSubdiv: opensubdiv/osd/mtlPatchTable.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
mtlMesh.h
\n+
mtlPatchTable.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,44 +107,119 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_MTL_MESH_H
\n-
26#define OPENSUBDIV3_OSD_MTL_MESH_H
\n+
25#ifndef OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H
\n+
26#define OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H
\n
27
\n
28#include "../version.h"
\n-
29#include "../osd/mesh.h"
\n-
30#include "../osd/mtlPatchTable.h"
\n-
31
\n-
32
\n-
33namespace OpenSubdiv {
\n-
34namespace OPENSUBDIV_VERSION {
\n-
35
\n-
36namespace Osd {
\n-
37
\n-
38typedef MeshInterface<MTLPatchTable> MTLMeshInterface;
\n+
29#include "../far/patchDescriptor.h"
\n+
30#include "../osd/nonCopyable.h"
\n+
31#include "../osd/types.h"
\n+
32#include "../osd/mtlCommon.h"
\n+
33
\n+
34@protocol MTLDevice;
\n+
35@protocol MTLBuffer;
\n+
36
\n+
37namespace OpenSubdiv {
\n+
38namespace OPENSUBDIV_VERSION {
\n
39
\n-
40} // end namespace Osd
\n-
41
\n-
42} // end namespace OPENSUBDIV_VERSION
\n-
43using namespace OPENSUBDIV_VERSION;
\n-
44
\n-
45} // end namespace OpenSubdiv
\n-
46
\n-
47#endif // OPENSUBDIV3_OSD_MTL_MESH_H
\n+
40namespace Far {
\n+
41 class PatchTable;
\n+
42};
\n+
43
\n+
44namespace Osd {
\n+
45
\n+
46class MTLPatchTable : private NonCopyable<MTLPatchTable> {
\n+
47public:
\n+
48 typedef id<MTLBuffer> VertexBufferBinding;
\n+
49
\n+
50 MTLPatchTable();
\n+
51 ~MTLPatchTable();
\n+
52
\n+
53 template<typename DEVICE_CONTEXT>
\n+
54 static MTLPatchTable *Create(Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context)
\n+
55 {
\n+
56 return Create(farPatchTable, context);
\n+
57 }
\n+
58
\n+
59 static MTLPatchTable *Create(Far::PatchTable const *farPatchTable, MTLContext* context);
\n+
60
\n+
61 PatchArrayVector const &GetPatchArrays() const { return _patchArrays; }
\n+
62 id<MTLBuffer> GetPatchIndexBuffer() const { return _indexBuffer; }
\n+
63 id<MTLBuffer> GetPatchParamBuffer() const { return _patchParamBuffer; }
\n+
64
\n+
65 PatchArrayVector const &GetVaryingPatchArrays() const { return _varyingPatchArrays; }
\n+
66 id<MTLBuffer> GetVaryingPatchIndexBuffer() const { return _varyingPatchIndexBuffer; }
\n+
67
\n+
68 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }
\n+
69 PatchArrayVector const &GetFVarPatchArrays(int fvarChannel = 0) const { return _fvarPatchArrays[fvarChannel]; }
\n+
70 id<MTLBuffer> GetFVarPatchIndexBuffer(int fvarChannel = 0) const { return _fvarIndexBuffers[fvarChannel]; }
\n+
71 id<MTLBuffer> GetFVarPatchParamBuffer(int fvarChannel = 0) const { return _fvarParamBuffers[fvarChannel]; }
\n+
72
\n+
73protected:
\n+
74 bool allocate(Far::PatchTable const *farPatchTable, MTLContext* context);
\n+
75
\n+
76 PatchArrayVector _patchArrays;
\n+
77
\n+
78 id<MTLBuffer> _indexBuffer;
\n+
79 id<MTLBuffer> _patchParamBuffer;
\n+
80
\n+
81 PatchArrayVector _varyingPatchArrays;
\n+
82
\n+
83 id<MTLBuffer> _varyingPatchIndexBuffer;
\n+
84
\n+
85 std::vector<PatchArrayVector> _fvarPatchArrays;
\n+
86 std::vector<id<MTLBuffer>> _fvarIndexBuffers;
\n+
87 std::vector<id<MTLBuffer>> _fvarParamBuffers;
\n+
88};
\n+
89
\n+
90} // end namespace Osd
\n+
91
\n+
92} //end namespace OPENSUBDIV_VERSION
\n+
93using namespace OPENSUBDIV_VERSION;
\n+
94
\n+
95} //end namespace OpenSubdiv
\n+
96
\n+
97#endif //end OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLMeshInterface
MeshInterface< MTLPatchTable > MTLMeshInterface
Definition: mtlMesh.h:38
\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetPatchArrays
PatchArrayVector const & GetPatchArrays() const
Definition: mtlPatchTable.h:61
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::allocate
bool allocate(Far::PatchTable const *farPatchTable, MTLContext *context)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::Create
static MTLPatchTable * Create(Far::PatchTable const *farPatchTable, MTLContext *context)
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarIndexBuffers
std::vector< id< MTLBuffer > > _fvarIndexBuffers
Definition: mtlPatchTable.h:86
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchParamBuffer
id< MTLBuffer > GetFVarPatchParamBuffer(int fvarChannel=0) const
Definition: mtlPatchTable.h:71
\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchArrays
PatchArrayVector const & GetFVarPatchArrays(int fvarChannel=0) const
Definition: mtlPatchTable.h:69
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarParamBuffers
std::vector< id< MTLBuffer > > _fvarParamBuffers
Definition: mtlPatchTable.h:87
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarPatchArrays
std::vector< PatchArrayVector > _fvarPatchArrays
Definition: mtlPatchTable.h:85
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::Create
static MTLPatchTable * Create(Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context)
Definition: mtlPatchTable.h:54
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchIndexBuffer
id< MTLBuffer > GetFVarPatchIndexBuffer(int fvarChannel=0) const
Definition: mtlPatchTable.h:70
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetVaryingPatchArrays
PatchArrayVector const & GetVaryingPatchArrays() const
Definition: mtlPatchTable.h:65
\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-mtlMesh.h\n+mtlPatchTable.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,41 +30,170 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_MTL_MESH_H\n- 26#define OPENSUBDIV3_OSD_MTL_MESH_H\n+ 25#ifndef OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H\n+ 26#define OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H\n 27\n 28#include \"../version.h\"\n- 29#include \"../osd/mesh.h\"\n- 30#include \"../osd/mtlPatchTable.h\"\n- 31\n- 32\n- 33namespace OpenSubdiv {\n- 34namespace OPENSUBDIV_VERSION {\n- 35\n- 36namespace Osd {\n- 37\n-38typedef MeshInterface MTLMeshInterface;\n+ 29#include \"../far/patchDescriptor.h\"\n+ 30#include \"../osd/nonCopyable.h\"\n+ 31#include \"../osd/types.h\"\n+ 32#include \"../osd/mtlCommon.h\"\n+ 33\n+ 34@protocol MTLDevice;\n+ 35@protocol MTLBuffer;\n+ 36\n+ 37namespace OpenSubdiv {\n+ 38namespace OPENSUBDIV_VERSION {\n 39\n- 40} // end namespace Osd\n- 41\n- 42} // end namespace OPENSUBDIV_VERSION\n- 43using namespace OPENSUBDIV_VERSION;\n- 44\n- 45} // end namespace OpenSubdiv\n- 46\n- 47#endif // OPENSUBDIV3_OSD_MTL_MESH_H\n+ 40namespace Far {\n+ 41 class PatchTable;\n+ 42};\n+ 43\n+ 44namespace Osd {\n+ 45\n+46class MTLPatchTable : private NonCopyable {\n+ 47public:\n+48 typedef id VertexBufferBinding;\n+ 49\n+50 MTLPatchTable();\n+51 ~MTLPatchTable();\n+ 52\n+ 53 template\n+54 static MTLPatchTable *Create(Far::PatchTable const *farPatchTable,\n+DEVICE_CONTEXT context)\n+ 55 {\n+ 56 return Create(farPatchTable, context);\n+ 57 }\n+ 58\n+59 static MTLPatchTable *Create(Far::PatchTable const *farPatchTable,\n+MTLContext* context);\n+ 60\n+61 PatchArrayVector const &GetPatchArrays() const { return _patchArrays; }\n+62 id GetPatchIndexBuffer() const { return _indexBuffer; }\n+63 id GetPatchParamBuffer() const { return _patchParamBuffer; }\n+ 64\n+65 PatchArrayVector const &GetVaryingPatchArrays() const { return\n+_varyingPatchArrays; }\n+66 id GetVaryingPatchIndexBuffer() const { return\n+_varyingPatchIndexBuffer; }\n+ 67\n+68 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }\n+69 PatchArrayVector const &GetFVarPatchArrays(int fvarChannel = 0) const\n+{ return _fvarPatchArrays[fvarChannel]; }\n+70 id GetFVarPatchIndexBuffer(int fvarChannel = 0) const { return\n+_fvarIndexBuffers[fvarChannel]; }\n+71 id GetFVarPatchParamBuffer(int fvarChannel = 0) const { return\n+_fvarParamBuffers[fvarChannel]; }\n+ 72\n+ 73protected:\n+74 bool allocate(Far::PatchTable const *farPatchTable, MTLContext* context);\n+ 75\n+76 PatchArrayVector _patchArrays;\n+ 77\n+78 id _indexBuffer;\n+79 id _patchParamBuffer;\n+ 80\n+81 PatchArrayVector _varyingPatchArrays;\n+ 82\n+83 id _varyingPatchIndexBuffer;\n+ 84\n+85 std::vector _fvarPatchArrays;\n+86 std::vector> _fvarIndexBuffers;\n+87 std::vector> _fvarParamBuffers;\n+ 88};\n+ 89\n+ 90} // end namespace Osd\n+ 91\n+ 92} //end namespace OPENSUBDIV_VERSION\n+ 93using namespace OPENSUBDIV_VERSION;\n+ 94\n+ 95} //end namespace OpenSubdiv\n+ 96\n+ 97#endif //end OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLMeshInterface\n-MeshInterface< MTLPatchTable > MTLMeshInterface\n-Definition: mtlMesh.h:38\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface\n-Definition: mesh.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n+std::vector< PatchArray > PatchArrayVector\n+Definition: types.h:132\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n+Container for arrays of parametric patches.\n+Definition: patchTable.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLContext\n+Definition: mtlCommon.h:40\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable\n+Definition: mtlPatchTable.h:46\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetPatchArrays\n+PatchArrayVector const & GetPatchArrays() const\n+Definition: mtlPatchTable.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::allocate\n+bool allocate(Far::PatchTable const *farPatchTable, MTLContext *context)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::Create\n+static MTLPatchTable * Create(Far::PatchTable const *farPatchTable, MTLContext\n+*context)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetVaryingPatchIndexBuffer\n+id< MTLBuffer > GetVaryingPatchIndexBuffer() const\n+Definition: mtlPatchTable.h:66\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarIndexBuffers\n+std::vector< id< MTLBuffer > > _fvarIndexBuffers\n+Definition: mtlPatchTable.h:86\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchParamBuffer\n+id< MTLBuffer > GetFVarPatchParamBuffer(int fvarChannel=0) const\n+Definition: mtlPatchTable.h:71\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::MTLPatchTable\n+MTLPatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetPatchParamBuffer\n+id< MTLBuffer > GetPatchParamBuffer() const\n+Definition: mtlPatchTable.h:63\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetPatchIndexBuffer\n+id< MTLBuffer > GetPatchIndexBuffer() const\n+Definition: mtlPatchTable.h:62\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_patchArrays\n+PatchArrayVector _patchArrays\n+Definition: mtlPatchTable.h:76\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::VertexBufferBinding\n+id< MTLBuffer > VertexBufferBinding\n+Definition: mtlPatchTable.h:48\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetNumFVarChannels\n+int GetNumFVarChannels() const\n+Definition: mtlPatchTable.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchArrays\n+PatchArrayVector const & GetFVarPatchArrays(int fvarChannel=0) const\n+Definition: mtlPatchTable.h:69\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_varyingPatchIndexBuffer\n+id< MTLBuffer > _varyingPatchIndexBuffer\n+Definition: mtlPatchTable.h:83\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarParamBuffers\n+std::vector< id< MTLBuffer > > _fvarParamBuffers\n+Definition: mtlPatchTable.h:87\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarPatchArrays\n+std::vector< PatchArrayVector > _fvarPatchArrays\n+Definition: mtlPatchTable.h:85\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_patchParamBuffer\n+id< MTLBuffer > _patchParamBuffer\n+Definition: mtlPatchTable.h:79\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_varyingPatchArrays\n+PatchArrayVector _varyingPatchArrays\n+Definition: mtlPatchTable.h:81\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::Create\n+static MTLPatchTable * Create(Far::PatchTable const *farPatchTable,\n+DEVICE_CONTEXT context)\n+Definition: mtlPatchTable.h:54\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::~MTLPatchTable\n+~MTLPatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchIndexBuffer\n+id< MTLBuffer > GetFVarPatchIndexBuffer(int fvarChannel=0) const\n+Definition: mtlPatchTable.h:70\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetVaryingPatchArrays\n+PatchArrayVector const & GetVaryingPatchArrays() const\n+Definition: mtlPatchTable.h:65\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_indexBuffer\n+id< MTLBuffer > _indexBuffer\n+Definition: mtlPatchTable.h:78\n * opensubdiv\n * osd\n- * mtlMesh.h\n+ * mtlPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00815.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00815.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/mtlPatchTable.h File Reference\n+OpenSubdiv: opensubdiv/osd/mtlLegacyGregoryPatchTable.h File Reference\n \n \n \n \n \n \n \n@@ -83,45 +83,42 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
mtlPatchTable.h File Reference
\n+
mtlLegacyGregoryPatchTable.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../far/patchDescriptor.h"
\n+#include "../far/patchTable.h"
\n #include "../osd/nonCopyable.h"
\n-#include "../osd/types.h"
\n #include "../osd/mtlCommon.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n+\n \n

\n Classes

class  MTLPatchTable
class  MTLLegacyGregoryPatchTable
 
\n \n \n \n \n \n-\n-\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,30 +5,27 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-mtlPatchTable.h File Reference\n+mtlLegacyGregoryPatchTable.h File Reference\n #include \"../version.h\"\n-#include \"../far/patchDescriptor.h\"\n+#include \"../far/patchTable.h\"\n #include \"../osd/nonCopyable.h\"\n-#include \"../osd/types.h\"\n #include \"../osd/mtlCommon.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0MTLPatchTable\n+class \u00a0MTLLegacyGregoryPatchTable\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n-\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * mtlPatchTable.h\n+ * mtlLegacyGregoryPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00815.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00815.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00815 = [\n- [\"MTLPatchTable\", \"a01297.html\", \"a01297\"]\n+ [\"MTLLegacyGregoryPatchTable\", \"a01289.html\", \"a01289\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00815_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00815_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/mtlPatchTable.h Source File\n+OpenSubdiv: opensubdiv/osd/mtlLegacyGregoryPatchTable.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
mtlPatchTable.h
\n+
mtlLegacyGregoryPatchTable.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,119 +107,100 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H
\n-
26#define OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H
\n+
25#ifndef OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H
\n+
26#define OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H
\n
27
\n
28#include "../version.h"
\n-
29#include "../far/patchDescriptor.h"
\n+
29#include "../far/patchTable.h"
\n
30#include "../osd/nonCopyable.h"
\n-
31#include "../osd/types.h"
\n-
32#include "../osd/mtlCommon.h"
\n-
33
\n-
34@protocol MTLDevice;
\n-
35@protocol MTLBuffer;
\n-
36
\n-
37namespace OpenSubdiv {
\n-
38namespace OPENSUBDIV_VERSION {
\n-
39
\n-
40namespace Far {
\n-
41 class PatchTable;
\n-
42};
\n-
43
\n-
44namespace Osd {
\n-
45
\n-
46class MTLPatchTable : private NonCopyable<MTLPatchTable> {
\n-
47public:
\n-
48 typedef id<MTLBuffer> VertexBufferBinding;
\n-
49
\n-
50 MTLPatchTable();
\n-
51 ~MTLPatchTable();
\n-
52
\n-
53 template<typename DEVICE_CONTEXT>
\n-
54 static MTLPatchTable *Create(Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context)
\n-
55 {
\n-
56 return Create(farPatchTable, context);
\n-
57 }
\n-
58
\n-
59 static MTLPatchTable *Create(Far::PatchTable const *farPatchTable, MTLContext* context);
\n+
31#include "../osd/mtlCommon.h"
\n+
32
\n+
33@protocol MTLDevice;
\n+
34@protocol MTLBuffer;
\n+
35
\n+
36namespace OpenSubdiv {
\n+
37namespace OPENSUBDIV_VERSION {
\n+
38
\n+
39namespace Osd {
\n+
40
\n+
41class MTLLegacyGregoryPatchTable
\n+
42: private NonCopyable<MTLLegacyGregoryPatchTable>
\n+
43{
\n+
44public:
\n+
45 ~MTLLegacyGregoryPatchTable();
\n+
46
\n+
47 template<typename DEVICE_CONTEXT>
\n+
48 static MTLLegacyGregoryPatchTable* Create(Far::PatchTable const* farPatchTable, DEVICE_CONTEXT context) {
\n+
49 return Create(farPatchTable, context);
\n+
50 }
\n+
51
\n+
52 static MTLLegacyGregoryPatchTable* Create(Far::PatchTable const* farPatchTable, MTLContext* context);
\n+
53
\n+
54 void UpdateVertexBuffer(id<MTLBuffer> vbo, int numVertices, int numVertexElements, MTLContext* context);
\n+
55
\n+
56 id<MTLBuffer> GetVertexBuffer() const
\n+
57 {
\n+
58 return _vertexBuffer;
\n+
59 }
\n
60
\n-
61 PatchArrayVector const &GetPatchArrays() const { return _patchArrays; }
\n-
62 id<MTLBuffer> GetPatchIndexBuffer() const { return _indexBuffer; }
\n-
63 id<MTLBuffer> GetPatchParamBuffer() const { return _patchParamBuffer; }
\n-
64
\n-
65 PatchArrayVector const &GetVaryingPatchArrays() const { return _varyingPatchArrays; }
\n-
66 id<MTLBuffer> GetVaryingPatchIndexBuffer() const { return _varyingPatchIndexBuffer; }
\n-
67
\n-
68 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }
\n-
69 PatchArrayVector const &GetFVarPatchArrays(int fvarChannel = 0) const { return _fvarPatchArrays[fvarChannel]; }
\n-
70 id<MTLBuffer> GetFVarPatchIndexBuffer(int fvarChannel = 0) const { return _fvarIndexBuffers[fvarChannel]; }
\n-
71 id<MTLBuffer> GetFVarPatchParamBuffer(int fvarChannel = 0) const { return _fvarParamBuffers[fvarChannel]; }
\n-
72
\n-
73protected:
\n-
74 bool allocate(Far::PatchTable const *farPatchTable, MTLContext* context);
\n-
75
\n-
76 PatchArrayVector _patchArrays;
\n+
61 id<MTLBuffer> GetVertexValenceBuffer() const
\n+
62 {
\n+
63 return _vertexValenceBuffer;
\n+
64 }
\n+
65
\n+
66 id<MTLBuffer> GetQuadOffsetsBuffer() const
\n+
67 {
\n+
68 return _quadOffsetsBuffer;
\n+
69 }
\n+
70
\n+
71 int GetQuadOffsetsBase(Far::PatchDescriptor::Type type)
\n+
72 {
\n+
73 if(type == Far::PatchDescriptor::GREGORY_BOUNDARY)
\n+
74 return _quadOffsetsBase[1];
\n+
75 return _quadOffsetsBase[0];
\n+
76 }
\n
77
\n-
78 id<MTLBuffer> _indexBuffer;
\n-
79 id<MTLBuffer> _patchParamBuffer;
\n-
80
\n-
81 PatchArrayVector _varyingPatchArrays;
\n-
82
\n-
83 id<MTLBuffer> _varyingPatchIndexBuffer;
\n+
78private:
\n+
79 id<MTLBuffer> _vertexBuffer;
\n+
80 id<MTLBuffer> _vertexValenceBuffer;
\n+
81 id<MTLBuffer> _quadOffsetsBuffer;
\n+
82 int _quadOffsetsBase[2];
\n+
83};
\n
84
\n-
85 std::vector<PatchArrayVector> _fvarPatchArrays;
\n-
86 std::vector<id<MTLBuffer>> _fvarIndexBuffers;
\n-
87 std::vector<id<MTLBuffer>> _fvarParamBuffers;
\n-
88};
\n+
85} //end namespace Osd
\n+
86
\n+
87} //end namespace OPENSUBDIV_VERSION
\n+
88using namespace OPENSUBDIV_VERSION;
\n
89
\n-
90} // end namespace Osd
\n+
90} //end namespace OpenSuddiv
\n
91
\n-
92} //end namespace OPENSUBDIV_VERSION
\n-
93using namespace OPENSUBDIV_VERSION;
\n-
94
\n-
95} //end namespace OpenSubdiv
\n-
96
\n-
97#endif //end OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H
\n+
92#endif // OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n+\n+\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n \n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetPatchArrays
PatchArrayVector const & GetPatchArrays() const
Definition: mtlPatchTable.h:61
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::allocate
bool allocate(Far::PatchTable const *farPatchTable, MTLContext *context)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::Create
static MTLPatchTable * Create(Far::PatchTable const *farPatchTable, MTLContext *context)
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarIndexBuffers
std::vector< id< MTLBuffer > > _fvarIndexBuffers
Definition: mtlPatchTable.h:86
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchParamBuffer
id< MTLBuffer > GetFVarPatchParamBuffer(int fvarChannel=0) const
Definition: mtlPatchTable.h:71
\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchArrays
PatchArrayVector const & GetFVarPatchArrays(int fvarChannel=0) const
Definition: mtlPatchTable.h:69
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarParamBuffers
std::vector< id< MTLBuffer > > _fvarParamBuffers
Definition: mtlPatchTable.h:87
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarPatchArrays
std::vector< PatchArrayVector > _fvarPatchArrays
Definition: mtlPatchTable.h:85
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::Create
static MTLPatchTable * Create(Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context)
Definition: mtlPatchTable.h:54
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchIndexBuffer
id< MTLBuffer > GetFVarPatchIndexBuffer(int fvarChannel=0) const
Definition: mtlPatchTable.h:70
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetVaryingPatchArrays
PatchArrayVector const & GetVaryingPatchArrays() const
Definition: mtlPatchTable.h:65
\n-\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::Create
static MTLLegacyGregoryPatchTable * Create(Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context)
Definition: mtlLegacyGregoryPatchTable.h:48
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::Create
static MTLLegacyGregoryPatchTable * Create(Far::PatchTable const *farPatchTable, MTLContext *context)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::UpdateVertexBuffer
void UpdateVertexBuffer(id< MTLBuffer > vbo, int numVertices, int numVertexElements, MTLContext *context)
\n+\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-mtlPatchTable.h\n+mtlLegacyGregoryPatchTable.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,170 +30,127 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H\n- 26#define OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H\n+ 25#ifndef OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H\n+ 26#define OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H\n 27\n 28#include \"../version.h\"\n- 29#include \"../far/patchDescriptor.h\"\n+ 29#include \"../far/patchTable.h\"\n 30#include \"../osd/nonCopyable.h\"\n- 31#include \"../osd/types.h\"\n- 32#include \"../osd/mtlCommon.h\"\n- 33\n- 34@protocol MTLDevice;\n- 35@protocol MTLBuffer;\n- 36\n- 37namespace OpenSubdiv {\n- 38namespace OPENSUBDIV_VERSION {\n- 39\n- 40namespace Far {\n- 41 class PatchTable;\n- 42};\n- 43\n- 44namespace Osd {\n- 45\n-46class MTLPatchTable : private NonCopyable {\n- 47public:\n-48 typedef id VertexBufferBinding;\n- 49\n-50 MTLPatchTable();\n-51 ~MTLPatchTable();\n- 52\n- 53 template\n-54 static MTLPatchTable *Create(Far::PatchTable const *farPatchTable,\n-DEVICE_CONTEXT context)\n- 55 {\n- 56 return Create(farPatchTable, context);\n- 57 }\n- 58\n-59 static MTLPatchTable *Create(Far::PatchTable const *farPatchTable,\n-MTLContext* context);\n+ 31#include \"../osd/mtlCommon.h\"\n+ 32\n+ 33@protocol MTLDevice;\n+ 34@protocol MTLBuffer;\n+ 35\n+ 36namespace OpenSubdiv {\n+ 37namespace OPENSUBDIV_VERSION {\n+ 38\n+ 39namespace Osd {\n+ 40\n+41class MTLLegacyGregoryPatchTable\n+ 42: private NonCopyable\n+ 43{\n+ 44public:\n+45 ~MTLLegacyGregoryPatchTable();\n+ 46\n+ 47 template\n+48 static MTLLegacyGregoryPatchTable* Create(Far::PatchTable const*\n+farPatchTable, DEVICE_CONTEXT context) {\n+ 49 return Create(farPatchTable, context);\n+ 50 }\n+ 51\n+52 static MTLLegacyGregoryPatchTable* Create(Far::PatchTable const*\n+farPatchTable, MTLContext* context);\n+ 53\n+54 void UpdateVertexBuffer(id vbo, int numVertices, int\n+numVertexElements, MTLContext* context);\n+ 55\n+56 id GetVertexBuffer() const\n+ 57 {\n+ 58 return _vertexBuffer;\n+ 59 }\n 60\n-61 PatchArrayVector const &GetPatchArrays() const { return _patchArrays; }\n-62 id GetPatchIndexBuffer() const { return _indexBuffer; }\n-63 id GetPatchParamBuffer() const { return _patchParamBuffer; }\n- 64\n-65 PatchArrayVector const &GetVaryingPatchArrays() const { return\n-_varyingPatchArrays; }\n-66 id GetVaryingPatchIndexBuffer() const { return\n-_varyingPatchIndexBuffer; }\n- 67\n-68 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }\n-69 PatchArrayVector const &GetFVarPatchArrays(int fvarChannel = 0) const\n-{ return _fvarPatchArrays[fvarChannel]; }\n-70 id GetFVarPatchIndexBuffer(int fvarChannel = 0) const { return\n-_fvarIndexBuffers[fvarChannel]; }\n-71 id GetFVarPatchParamBuffer(int fvarChannel = 0) const { return\n-_fvarParamBuffers[fvarChannel]; }\n- 72\n- 73protected:\n-74 bool allocate(Far::PatchTable const *farPatchTable, MTLContext* context);\n- 75\n-76 PatchArrayVector _patchArrays;\n+61 id GetVertexValenceBuffer() const\n+ 62 {\n+ 63 return _vertexValenceBuffer;\n+ 64 }\n+ 65\n+66 id GetQuadOffsetsBuffer() const\n+ 67 {\n+ 68 return _quadOffsetsBuffer;\n+ 69 }\n+ 70\n+71 int GetQuadOffsetsBase(Far::PatchDescriptor::Type type)\n+ 72 {\n+ 73 if(type == Far::PatchDescriptor::GREGORY_BOUNDARY)\n+ 74 return _quadOffsetsBase[1];\n+ 75 return _quadOffsetsBase[0];\n+ 76 }\n 77\n-78 id _indexBuffer;\n-79 id _patchParamBuffer;\n- 80\n-81 PatchArrayVector _varyingPatchArrays;\n- 82\n-83 id _varyingPatchIndexBuffer;\n+ 78private:\n+ 79 id _vertexBuffer;\n+ 80 id _vertexValenceBuffer;\n+ 81 id _quadOffsetsBuffer;\n+ 82 int _quadOffsetsBase[2];\n+ 83};\n 84\n-85 std::vector _fvarPatchArrays;\n-86 std::vector> _fvarIndexBuffers;\n-87 std::vector> _fvarParamBuffers;\n- 88};\n+ 85} //end namespace Osd\n+ 86\n+ 87} //end namespace OPENSUBDIV_VERSION\n+ 88using namespace OPENSUBDIV_VERSION;\n 89\n- 90} // end namespace Osd\n+ 90} //end namespace OpenSuddiv\n 91\n- 92} //end namespace OPENSUBDIV_VERSION\n- 93using namespace OPENSUBDIV_VERSION;\n- 94\n- 95} //end namespace OpenSubdiv\n- 96\n- 97#endif //end OPENSUBDIV3_OSD_MTL_PATCH_TABLE_H\n+ 92#endif // OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n-std::vector< PatchArray > PatchArrayVector\n-Definition: types.h:132\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n+Type\n+Definition: patchDescriptor.h:48\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_BOUNDARY\n+@ GREGORY_BOUNDARY\n+Definition: patchDescriptor.h:61\n OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n Container for arrays of parametric patches.\n Definition: patchTable.h:55\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLContext\n Definition: mtlCommon.h:40\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable\n-Definition: mtlPatchTable.h:46\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetPatchArrays\n-PatchArrayVector const & GetPatchArrays() const\n-Definition: mtlPatchTable.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::allocate\n-bool allocate(Far::PatchTable const *farPatchTable, MTLContext *context)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::Create\n-static MTLPatchTable * Create(Far::PatchTable const *farPatchTable, MTLContext\n-*context)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetVaryingPatchIndexBuffer\n-id< MTLBuffer > GetVaryingPatchIndexBuffer() const\n-Definition: mtlPatchTable.h:66\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarIndexBuffers\n-std::vector< id< MTLBuffer > > _fvarIndexBuffers\n-Definition: mtlPatchTable.h:86\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchParamBuffer\n-id< MTLBuffer > GetFVarPatchParamBuffer(int fvarChannel=0) const\n-Definition: mtlPatchTable.h:71\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::MTLPatchTable\n-MTLPatchTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetPatchParamBuffer\n-id< MTLBuffer > GetPatchParamBuffer() const\n-Definition: mtlPatchTable.h:63\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetPatchIndexBuffer\n-id< MTLBuffer > GetPatchIndexBuffer() const\n-Definition: mtlPatchTable.h:62\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_patchArrays\n-PatchArrayVector _patchArrays\n-Definition: mtlPatchTable.h:76\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::VertexBufferBinding\n-id< MTLBuffer > VertexBufferBinding\n-Definition: mtlPatchTable.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetNumFVarChannels\n-int GetNumFVarChannels() const\n-Definition: mtlPatchTable.h:68\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchArrays\n-PatchArrayVector const & GetFVarPatchArrays(int fvarChannel=0) const\n-Definition: mtlPatchTable.h:69\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_varyingPatchIndexBuffer\n-id< MTLBuffer > _varyingPatchIndexBuffer\n-Definition: mtlPatchTable.h:83\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarParamBuffers\n-std::vector< id< MTLBuffer > > _fvarParamBuffers\n-Definition: mtlPatchTable.h:87\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_fvarPatchArrays\n-std::vector< PatchArrayVector > _fvarPatchArrays\n-Definition: mtlPatchTable.h:85\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_patchParamBuffer\n-id< MTLBuffer > _patchParamBuffer\n-Definition: mtlPatchTable.h:79\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_varyingPatchArrays\n-PatchArrayVector _varyingPatchArrays\n-Definition: mtlPatchTable.h:81\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::Create\n-static MTLPatchTable * Create(Far::PatchTable const *farPatchTable,\n-DEVICE_CONTEXT context)\n-Definition: mtlPatchTable.h:54\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::~MTLPatchTable\n-~MTLPatchTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetFVarPatchIndexBuffer\n-id< MTLBuffer > GetFVarPatchIndexBuffer(int fvarChannel=0) const\n-Definition: mtlPatchTable.h:70\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::GetVaryingPatchArrays\n-PatchArrayVector const & GetVaryingPatchArrays() const\n-Definition: mtlPatchTable.h:65\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::_indexBuffer\n-id< MTLBuffer > _indexBuffer\n-Definition: mtlPatchTable.h:78\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable\n+Definition: mtlLegacyGregoryPatchTable.h:43\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n+GetQuadOffsetsBase\n+int GetQuadOffsetsBase(Far::PatchDescriptor::Type type)\n+Definition: mtlLegacyGregoryPatchTable.h:71\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::Create\n+static MTLLegacyGregoryPatchTable * Create(Far::PatchTable const\n+*farPatchTable, DEVICE_CONTEXT context)\n+Definition: mtlLegacyGregoryPatchTable.h:48\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n+GetVertexValenceBuffer\n+id< MTLBuffer > GetVertexValenceBuffer() const\n+Definition: mtlLegacyGregoryPatchTable.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n+GetQuadOffsetsBuffer\n+id< MTLBuffer > GetQuadOffsetsBuffer() const\n+Definition: mtlLegacyGregoryPatchTable.h:66\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::Create\n+static MTLLegacyGregoryPatchTable * Create(Far::PatchTable const\n+*farPatchTable, MTLContext *context)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n+UpdateVertexBuffer\n+void UpdateVertexBuffer(id< MTLBuffer > vbo, int numVertices, int\n+numVertexElements, MTLContext *context)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n+~MTLLegacyGregoryPatchTable\n+~MTLLegacyGregoryPatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n+GetVertexBuffer\n+id< MTLBuffer > GetVertexBuffer() const\n+Definition: mtlLegacyGregoryPatchTable.h:56\n * opensubdiv\n * osd\n- * mtlPatchTable.h\n+ * mtlLegacyGregoryPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00818.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00818.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/mtlLegacyGregoryPatchTable.h File Reference\n+OpenSubdiv: opensubdiv/osd/mtlMesh.h File Reference\n \n \n \n \n \n \n \n@@ -81,44 +81,43 @@\n \n \n \n
\n
\n-Classes |\n-Namespaces
\n-
mtlLegacyGregoryPatchTable.h File Reference
\n+Namespaces |\n+Typedefs
\n+
mtlMesh.h File Reference
\n \n
\n
#include "../version.h"
\n-#include "../far/patchTable.h"
\n-#include "../osd/nonCopyable.h"
\n-#include "../osd/mtlCommon.h"
\n+#include "../osd/mesh.h"
\n+#include "../osd/mtlPatchTable.h"
\n
\n

Go to the source code of this file.

\n \n-\n-\n-\n-

\n-Classes

class  MTLLegacyGregoryPatchTable
 
\n \n \n \n \n \n \n \n+

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n+\n+\n+\n

\n+Typedefs

typedef MeshInterface< MTLPatchTable > MTLMeshInterface
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,28 +4,27 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-Classes | Namespaces\n-mtlLegacyGregoryPatchTable.h File Reference\n+Namespaces | Typedefs\n+mtlMesh.h File Reference\n #include \"../version.h\"\n-#include \"../far/patchTable.h\"\n-#include \"../osd/nonCopyable.h\"\n-#include \"../osd/mtlCommon.h\"\n+#include \"../osd/mesh.h\"\n+#include \"../osd/mtlPatchTable.h\"\n Go_to_the_source_code_of_this_file.\n- Classes\n-class \u00a0MTLLegacyGregoryPatchTable\n-\u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n+ Typedefs\n+typedef MeshInterface< MTLPatchTable >\u00a0MTLMeshInterface\n+\u00a0\n * opensubdiv\n * osd\n- * mtlLegacyGregoryPatchTable.h\n+ * mtlMesh.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00818.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00818.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00818 = [\n- [\"MTLLegacyGregoryPatchTable\", \"a01289.html\", \"a01289\"]\n+ [\"MTLMeshInterface\", \"a00818.html#aa41e5b43fe46eeb0f97ff6ebdb1935aa\", null]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00818_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00818_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/mtlLegacyGregoryPatchTable.h Source File\n+OpenSubdiv: opensubdiv/osd/mtlMesh.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
mtlLegacyGregoryPatchTable.h
\n+
mtlMesh.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,100 +107,44 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H
\n-
26#define OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H
\n+
25#ifndef OPENSUBDIV3_OSD_MTL_MESH_H
\n+
26#define OPENSUBDIV3_OSD_MTL_MESH_H
\n
27
\n
28#include "../version.h"
\n-
29#include "../far/patchTable.h"
\n-
30#include "../osd/nonCopyable.h"
\n-
31#include "../osd/mtlCommon.h"
\n+
29#include "../osd/mesh.h"
\n+
30#include "../osd/mtlPatchTable.h"
\n+
31
\n
32
\n-
33@protocol MTLDevice;
\n-
34@protocol MTLBuffer;
\n+
33namespace OpenSubdiv {
\n+
34namespace OPENSUBDIV_VERSION {
\n
35
\n-
36namespace OpenSubdiv {
\n-
37namespace OPENSUBDIV_VERSION {
\n-
38
\n-
39namespace Osd {
\n-
40
\n-
41class MTLLegacyGregoryPatchTable
\n-
42: private NonCopyable<MTLLegacyGregoryPatchTable>
\n-
43{
\n-
44public:
\n-
45 ~MTLLegacyGregoryPatchTable();
\n+
36namespace Osd {
\n+
37
\n+
38typedef MeshInterface<MTLPatchTable> MTLMeshInterface;
\n+
39
\n+
40} // end namespace Osd
\n+
41
\n+
42} // end namespace OPENSUBDIV_VERSION
\n+
43using namespace OPENSUBDIV_VERSION;
\n+
44
\n+
45} // end namespace OpenSubdiv
\n
46
\n-
47 template<typename DEVICE_CONTEXT>
\n-
48 static MTLLegacyGregoryPatchTable* Create(Far::PatchTable const* farPatchTable, DEVICE_CONTEXT context) {
\n-
49 return Create(farPatchTable, context);
\n-
50 }
\n-
51
\n-
52 static MTLLegacyGregoryPatchTable* Create(Far::PatchTable const* farPatchTable, MTLContext* context);
\n-
53
\n-
54 void UpdateVertexBuffer(id<MTLBuffer> vbo, int numVertices, int numVertexElements, MTLContext* context);
\n-
55
\n-
56 id<MTLBuffer> GetVertexBuffer() const
\n-
57 {
\n-
58 return _vertexBuffer;
\n-
59 }
\n-
60
\n-
61 id<MTLBuffer> GetVertexValenceBuffer() const
\n-
62 {
\n-
63 return _vertexValenceBuffer;
\n-
64 }
\n-
65
\n-
66 id<MTLBuffer> GetQuadOffsetsBuffer() const
\n-
67 {
\n-
68 return _quadOffsetsBuffer;
\n-
69 }
\n-
70
\n-
71 int GetQuadOffsetsBase(Far::PatchDescriptor::Type type)
\n-
72 {
\n-
73 if(type == Far::PatchDescriptor::GREGORY_BOUNDARY)
\n-
74 return _quadOffsetsBase[1];
\n-
75 return _quadOffsetsBase[0];
\n-
76 }
\n-
77
\n-
78private:
\n-
79 id<MTLBuffer> _vertexBuffer;
\n-
80 id<MTLBuffer> _vertexValenceBuffer;
\n-
81 id<MTLBuffer> _quadOffsetsBuffer;
\n-
82 int _quadOffsetsBase[2];
\n-
83};
\n-
84
\n-
85} //end namespace Osd
\n-
86
\n-
87} //end namespace OPENSUBDIV_VERSION
\n-
88using namespace OPENSUBDIV_VERSION;
\n-
89
\n-
90} //end namespace OpenSuddiv
\n-
91
\n-
92#endif // OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H
\n+
47#endif // OPENSUBDIV3_OSD_MTL_MESH_H
\n \n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::Create
static MTLLegacyGregoryPatchTable * Create(Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context)
Definition: mtlLegacyGregoryPatchTable.h:48
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::Create
static MTLLegacyGregoryPatchTable * Create(Far::PatchTable const *farPatchTable, MTLContext *context)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::UpdateVertexBuffer
void UpdateVertexBuffer(id< MTLBuffer > vbo, int numVertices, int numVertexElements, MTLContext *context)
\n-\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLMeshInterface
MeshInterface< MTLPatchTable > MTLMeshInterface
Definition: mtlMesh.h:38
\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-mtlLegacyGregoryPatchTable.h\n+mtlMesh.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,127 +30,41 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H\n- 26#define OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H\n+ 25#ifndef OPENSUBDIV3_OSD_MTL_MESH_H\n+ 26#define OPENSUBDIV3_OSD_MTL_MESH_H\n 27\n 28#include \"../version.h\"\n- 29#include \"../far/patchTable.h\"\n- 30#include \"../osd/nonCopyable.h\"\n- 31#include \"../osd/mtlCommon.h\"\n+ 29#include \"../osd/mesh.h\"\n+ 30#include \"../osd/mtlPatchTable.h\"\n+ 31\n 32\n- 33@protocol MTLDevice;\n- 34@protocol MTLBuffer;\n+ 33namespace OpenSubdiv {\n+ 34namespace OPENSUBDIV_VERSION {\n 35\n- 36namespace OpenSubdiv {\n- 37namespace OPENSUBDIV_VERSION {\n- 38\n- 39namespace Osd {\n- 40\n-41class MTLLegacyGregoryPatchTable\n- 42: private NonCopyable\n- 43{\n- 44public:\n-45 ~MTLLegacyGregoryPatchTable();\n+ 36namespace Osd {\n+ 37\n+38typedef MeshInterface MTLMeshInterface;\n+ 39\n+ 40} // end namespace Osd\n+ 41\n+ 42} // end namespace OPENSUBDIV_VERSION\n+ 43using namespace OPENSUBDIV_VERSION;\n+ 44\n+ 45} // end namespace OpenSubdiv\n 46\n- 47 template\n-48 static MTLLegacyGregoryPatchTable* Create(Far::PatchTable const*\n-farPatchTable, DEVICE_CONTEXT context) {\n- 49 return Create(farPatchTable, context);\n- 50 }\n- 51\n-52 static MTLLegacyGregoryPatchTable* Create(Far::PatchTable const*\n-farPatchTable, MTLContext* context);\n- 53\n-54 void UpdateVertexBuffer(id vbo, int numVertices, int\n-numVertexElements, MTLContext* context);\n- 55\n-56 id GetVertexBuffer() const\n- 57 {\n- 58 return _vertexBuffer;\n- 59 }\n- 60\n-61 id GetVertexValenceBuffer() const\n- 62 {\n- 63 return _vertexValenceBuffer;\n- 64 }\n- 65\n-66 id GetQuadOffsetsBuffer() const\n- 67 {\n- 68 return _quadOffsetsBuffer;\n- 69 }\n- 70\n-71 int GetQuadOffsetsBase(Far::PatchDescriptor::Type type)\n- 72 {\n- 73 if(type == Far::PatchDescriptor::GREGORY_BOUNDARY)\n- 74 return _quadOffsetsBase[1];\n- 75 return _quadOffsetsBase[0];\n- 76 }\n- 77\n- 78private:\n- 79 id _vertexBuffer;\n- 80 id _vertexValenceBuffer;\n- 81 id _quadOffsetsBuffer;\n- 82 int _quadOffsetsBase[2];\n- 83};\n- 84\n- 85} //end namespace Osd\n- 86\n- 87} //end namespace OPENSUBDIV_VERSION\n- 88using namespace OPENSUBDIV_VERSION;\n- 89\n- 90} //end namespace OpenSuddiv\n- 91\n- 92#endif // OPENSUBDIV3_OSD_MTL_LEGACY_GREGORY_PATCH_TABLE_H\n+ 47#endif // OPENSUBDIV3_OSD_MTL_MESH_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n-Type\n-Definition: patchDescriptor.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_BOUNDARY\n-@ GREGORY_BOUNDARY\n-Definition: patchDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n-Container for arrays of parametric patches.\n-Definition: patchTable.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLContext\n-Definition: mtlCommon.h:40\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable\n-Definition: mtlLegacyGregoryPatchTable.h:43\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n-GetQuadOffsetsBase\n-int GetQuadOffsetsBase(Far::PatchDescriptor::Type type)\n-Definition: mtlLegacyGregoryPatchTable.h:71\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::Create\n-static MTLLegacyGregoryPatchTable * Create(Far::PatchTable const\n-*farPatchTable, DEVICE_CONTEXT context)\n-Definition: mtlLegacyGregoryPatchTable.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n-GetVertexValenceBuffer\n-id< MTLBuffer > GetVertexValenceBuffer() const\n-Definition: mtlLegacyGregoryPatchTable.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n-GetQuadOffsetsBuffer\n-id< MTLBuffer > GetQuadOffsetsBuffer() const\n-Definition: mtlLegacyGregoryPatchTable.h:66\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::Create\n-static MTLLegacyGregoryPatchTable * Create(Far::PatchTable const\n-*farPatchTable, MTLContext *context)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n-UpdateVertexBuffer\n-void UpdateVertexBuffer(id< MTLBuffer > vbo, int numVertices, int\n-numVertexElements, MTLContext *context)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n-~MTLLegacyGregoryPatchTable\n-~MTLLegacyGregoryPatchTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLLegacyGregoryPatchTable::\n-GetVertexBuffer\n-id< MTLBuffer > GetVertexBuffer() const\n-Definition: mtlLegacyGregoryPatchTable.h:56\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLMeshInterface\n+MeshInterface< MTLPatchTable > MTLMeshInterface\n+Definition: mtlMesh.h:38\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface\n+Definition: mesh.h:68\n * opensubdiv\n * osd\n- * mtlLegacyGregoryPatchTable.h\n+ * mtlMesh.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00821.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00821.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/mtlVertexBuffer.h File Reference\n+OpenSubdiv: opensubdiv/osd/mtlComputeEvaluator.h File Reference\n \n \n \n \n \n \n \n@@ -83,40 +83,46 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
mtlVertexBuffer.h File Reference
\n+
mtlComputeEvaluator.h File Reference
\n
\n
\n
#include "../version.h"
\n+#include "../osd/types.h"
\n+#include "../osd/bufferDescriptor.h"
\n #include "../osd/mtlCommon.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n+\n+\n+\n \n

\n Classes

class  CPUMTLVertexBuffer
class  MTLStencilTable
 
class  MTLComputeEvaluator
 
\n \n \n \n \n \n+\n+\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,25 +5,31 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-mtlVertexBuffer.h File Reference\n+mtlComputeEvaluator.h File Reference\n #include \"../version.h\"\n+#include \"../osd/types.h\"\n+#include \"../osd/bufferDescriptor.h\"\n #include \"../osd/mtlCommon.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0CPUMTLVertexBuffer\n+class \u00a0MTLStencilTable\n+\u00a0\n+class \u00a0MTLComputeEvaluator\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n+\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * mtlVertexBuffer.h\n+ * mtlComputeEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00821.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00821.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,4 @@\n var a00821 = [\n- [\"CPUMTLVertexBuffer\", \"a01301.html\", \"a01301\"]\n+ [\"MTLStencilTable\", \"a01281.html\", \"a01281\"],\n+ [\"MTLComputeEvaluator\", \"a01285.html\", \"a01285\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00821_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00821_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/mtlVertexBuffer.h Source File\n+OpenSubdiv: opensubdiv/osd/mtlComputeEvaluator.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
mtlVertexBuffer.h
\n+
mtlComputeEvaluator.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,89 +107,1084 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H
\n-
26#define OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H
\n+
25#ifndef OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H
\n+
26#define OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H
\n
27
\n
28#include "../version.h"
\n-
29#include "../osd/mtlCommon.h"
\n-
30
\n-
31@protocol MTLDevice;
\n-
32@protocol MTLBuffer;
\n+
29
\n+
30#include "../osd/types.h"
\n+
31#include "../osd/bufferDescriptor.h"
\n+
32#include "../osd/mtlCommon.h"
\n
33
\n-
34namespace OpenSubdiv {
\n-
35namespace OPENSUBDIV_VERSION {
\n-
36
\n-
37namespace Osd {
\n+
34@protocol MTLDevice;
\n+
35@protocol MTLBuffer;
\n+
36@protocol MTLLibrary;
\n+
37@protocol MTLComputePipelineState;
\n
38
\n-
39class CPUMTLVertexBuffer {
\n-
40public:
\n-
41 static CPUMTLVertexBuffer* Create(int numElements, int numVertices, MTLContext* context);
\n-
42
\n-
43 void UpdateData(const float* src, int startVertex, int numVertices, MTLContext* context);
\n-
44
\n-
45 int GetNumElements() const
\n-
46 {
\n-
47 return _numElements;
\n-
48 }
\n+
39namespace OpenSubdiv {
\n+
40namespace OPENSUBDIV_VERSION {
\n+
41
\n+
42namespace Far {
\n+
43 class PatchTable;
\n+
44 class StencilTable;
\n+
45 class LimitStencilTable;
\n+
46}
\n+
47
\n+
48namespace Osd {
\n
49
\n-
50 int GetNumVertices() const
\n-
51 {
\n-
52 return _numVertices;
\n-
53 }
\n-
54
\n-
55 float* BindCpuBuffer();
\n-
56 id<MTLBuffer> BindMTLBuffer(MTLContext* context);
\n-
57
\n-
58 id<MTLBuffer> BindVBO(MTLContext* context)
\n-
59 {
\n-
60 return BindMTLBuffer(context);
\n-
61 }
\n-
62
\n-
63protected:
\n+
50class MTLStencilTable
\n+
51{
\n+
52public:
\n+
53 template<typename STENCIL_TABLE, typename DEVICE_CONTEXT>
\n+
54 static MTLStencilTable* Create(STENCIL_TABLE* stencilTable,
\n+
55 DEVICE_CONTEXT context)
\n+
56 {
\n+
57 return new MTLStencilTable(stencilTable, context);
\n+
58 }
\n+
59
\n+
60
\n+
61 MTLStencilTable(Far::StencilTable const* stencilTable, MTLContext* context);
\n+
62 MTLStencilTable(Far::LimitStencilTable const* stencilTable, MTLContext* context);
\n+
63 ~MTLStencilTable();
\n
64
\n-
65 CPUMTLVertexBuffer(int numElements, int numVertices);
\n-
66
\n-
67 bool allocate(MTLContext* context);
\n-
68
\n-
69private:
\n-
70 int _numElements;
\n-
71 int _numVertices;
\n-
72 id<MTLBuffer> _buffer;
\n-
73 bool _dirty;
\n-
74};
\n-
75
\n-
76} //end namespace Osd
\n-
77
\n-
78} //end namespace OPENSUBDIV_VERSION
\n-
79 using namespace OPENSUBDIV_VERSION;
\n-
80
\n-
81} //end namespace OpenSubdiv
\n-
82
\n-
83#endif // OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H
\n+
65 id<MTLBuffer> GetSizesBuffer() const { return _sizesBuffer; }
\n+
66 id<MTLBuffer> GetOffsetsBuffer() const { return _offsetsBuffer; }
\n+
67 id<MTLBuffer> GetIndicesBuffer() const { return _indicesBuffer; }
\n+
68 id<MTLBuffer> GetWeightsBuffer() const { return _weightsBuffer; }
\n+
69 id<MTLBuffer> GetDuWeightsBuffer() const { return _duWeightsBuffer; }
\n+
70 id<MTLBuffer> GetDvWeightsBuffer() const { return _dvWeightsBuffer; }
\n+
71 id<MTLBuffer> GetDuuWeightsBuffer() const { return _duuWeightsBuffer; }
\n+
72 id<MTLBuffer> GetDuvWeightsBuffer() const { return _duvWeightsBuffer; }
\n+
73 id<MTLBuffer> GetDvvWeightsBuffer() const { return _dvvWeightsBuffer; }
\n+
74
\n+
75 int GetNumStencils() const { return _numStencils; }
\n+
76
\n+
77private:
\n+
78 id<MTLBuffer> _sizesBuffer;
\n+
79 id<MTLBuffer> _offsetsBuffer;
\n+
80 id<MTLBuffer> _indicesBuffer;
\n+
81 id<MTLBuffer> _weightsBuffer;
\n+
82 id<MTLBuffer> _duWeightsBuffer;
\n+
83 id<MTLBuffer> _dvWeightsBuffer;
\n+
84 id<MTLBuffer> _duuWeightsBuffer;
\n+
85 id<MTLBuffer> _duvWeightsBuffer;
\n+
86 id<MTLBuffer> _dvvWeightsBuffer;
\n+
87
\n+
88 int _numStencils;
\n+
89};
\n+
90
\n+
91class MTLComputeEvaluator
\n+
92{
\n+
93public:
\n+
94 typedef bool Instantiatable;
\n+
95
\n+
96 static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n+
97 BufferDescriptor const &dstDesc,
\n+
98 BufferDescriptor const &duDesc,
\n+
99 BufferDescriptor const &dvDesc,
\n+
100 MTLContext* context);
\n+
101
\n+
102 static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n+
103 BufferDescriptor const &dstDesc,
\n+
104 BufferDescriptor const &duDesc,
\n+
105 BufferDescriptor const &dvDesc,
\n+
106 BufferDescriptor const &duuDesc,
\n+
107 BufferDescriptor const &duvDesc,
\n+
108 BufferDescriptor const &dvvDesc,
\n+
109 MTLContext* context);
\n+
110
\n+
111 MTLComputeEvaluator();
\n+
112 ~MTLComputeEvaluator();
\n+
113
\n+
119
\n+
148 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
149 static bool EvalStencils(
\n+
150 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
151 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
152 STENCIL_TABLE const *stencilTable,
\n+
153 MTLComputeEvaluator const *instance,
\n+
154 MTLContext* context)
\n+
155 {
\n+
156 if (instance) {
\n+
157 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
158 dstBuffer, dstDesc,
\n+
159 stencilTable,
\n+
160 context);
\n+
161 } else {
\n+
162 // Create an instance on demand (slow)
\n+
163 instance = Create(srcDesc, dstDesc,
\n+
164 BufferDescriptor(),
\n+
165 BufferDescriptor(),
\n+
166 context);
\n+
167 if (instance) {
\n+
168 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
169 dstBuffer, dstDesc,
\n+
170 stencilTable,
\n+
171 context);
\n+
172 delete instance;
\n+
173 return r;
\n+
174 }
\n+
175 return false;
\n+
176 }
\n+
177 }
\n+
178
\n+
219 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
220 static bool EvalStencils(
\n+
221 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
222 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
223 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
224 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
225 STENCIL_TABLE const *stencilTable,
\n+
226 MTLComputeEvaluator const *instance,
\n+
227 MTLContext* context) {
\n+
228
\n+
229 if (instance) {
\n+
230 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
231 dstBuffer, dstDesc,
\n+
232 duBuffer, duDesc,
\n+
233 dvBuffer, dvDesc,
\n+
234 stencilTable,
\n+
235 context);
\n+
236 } else {
\n+
237 // Create an instance on demand (slow)
\n+
238 instance = Create(srcDesc, dstDesc, duDesc, dvDesc, context);
\n+
239 if (instance) {
\n+
240 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
241 dstBuffer, dstDesc,
\n+
242 duBuffer, duDesc,
\n+
243 dvBuffer, dvDesc,
\n+
244 stencilTable,
\n+
245 context);
\n+
246 delete instance;
\n+
247 return r;
\n+
248 }
\n+
249 return false;
\n+
250 }
\n+
251 }
\n+
252
\n+
311 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
312 static bool EvalStencils(
\n+
313 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
314 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
315 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
316 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
317 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
318 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
319 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
320 STENCIL_TABLE const *stencilTable,
\n+
321 MTLComputeEvaluator const *instance,
\n+
322 MTLContext* context) {
\n+
323
\n+
324 if (instance) {
\n+
325 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
326 dstBuffer, dstDesc,
\n+
327 duBuffer, duDesc,
\n+
328 dvBuffer, dvDesc,
\n+
329 duuBuffer, duuDesc,
\n+
330 duvBuffer, duvDesc,
\n+
331 dvvBuffer, dvvDesc,
\n+
332 stencilTable,
\n+
333 context);
\n+
334 } else {
\n+
335 // Create an instance on demand (slow)
\n+
336 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,
\n+
337 duuDesc, duvDesc, dvvDesc, context);
\n+
338 if (instance) {
\n+
339 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
340 dstBuffer, dstDesc,
\n+
341 duBuffer, duDesc,
\n+
342 dvBuffer, dvDesc,
\n+
343 duuBuffer, duuDesc,
\n+
344 duvBuffer, duvDesc,
\n+
345 dvvBuffer, dvvDesc,
\n+
346 stencilTable,
\n+
347 context);
\n+
348 delete instance;
\n+
349 return r;
\n+
350 }
\n+
351 return false;
\n+
352 }
\n+
353 }
\n+
354
\n+
375 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
376 bool EvalStencils(
\n+
377 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
378 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
379 STENCIL_TABLE const *stencilTable,
\n+
380 MTLContext* context) const
\n+
381 {
\n+
382 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,
\n+
383 dstBuffer->BindMTLBuffer(context), dstDesc,
\n+
384 0, BufferDescriptor(),
\n+
385 0, BufferDescriptor(),
\n+
386 stencilTable->GetSizesBuffer(),
\n+
387 stencilTable->GetOffsetsBuffer(),
\n+
388 stencilTable->GetIndicesBuffer(),
\n+
389 stencilTable->GetWeightsBuffer(),
\n+
390 0,
\n+
391 0,
\n+
392 /* start = */ 0,
\n+
393 /* end = */ stencilTable->GetNumStencils(),
\n+
394 context);
\n+
395 }
\n+
396
\n+
429 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
430 bool EvalStencils(
\n+
431 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
432 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
433 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
434 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
435 STENCIL_TABLE const *stencilTable,
\n+
436 MTLContext* context) const
\n+
437 {
\n+
438 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,
\n+
439 dstBuffer->BindMTLBuffer(context), dstDesc,
\n+
440 duBuffer->BindMTLBuffer(context), duDesc,
\n+
441 dvBuffer->BindMTLBuffer(context), dvDesc,
\n+
442 stencilTable->GetSizesBuffer(),
\n+
443 stencilTable->GetOffsetsBuffer(),
\n+
444 stencilTable->GetIndicesBuffer(),
\n+
445 stencilTable->GetWeightsBuffer(),
\n+
446 stencilTable->GetDuWeightsBuffer(),
\n+
447 stencilTable->GetDvWeightsBuffer(),
\n+
448 /* start = */ 0,
\n+
449 /* end = */ stencilTable->GetNumStencils(),
\n+
450 context);
\n+
451 }
\n+
452
\n+
503 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
504 bool EvalStencils(
\n+
505 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
506 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
507 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
508 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
509 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
510 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
511 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
512 STENCIL_TABLE const *stencilTable,
\n+
513 MTLContext* context) const
\n+
514 {
\n+
515 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,
\n+
516 dstBuffer->BindMTLBuffer(context), dstDesc,
\n+
517 duBuffer->BindMTLBuffer(context), duDesc,
\n+
518 dvBuffer->BindMTLBuffer(context), dvDesc,
\n+
519 duuBuffer->BindMTLBuffer(context), duuDesc,
\n+
520 duvBuffer->BindMTLBuffer(context), duvDesc,
\n+
521 dvvBuffer->BindMTLBuffer(context), dvvDesc,
\n+
522 stencilTable->GetSizesBuffer(),
\n+
523 stencilTable->GetOffsetsBuffer(),
\n+
524 stencilTable->GetIndicesBuffer(),
\n+
525 stencilTable->GetWeightsBuffer(),
\n+
526 stencilTable->GetDuWeightsBuffer(),
\n+
527 stencilTable->GetDvWeightsBuffer(),
\n+
528 stencilTable->GetDuuWeightsBuffer(),
\n+
529 stencilTable->GetDuvWeightsBuffer(),
\n+
530 stencilTable->GetDvvWeightsBuffer(),
\n+
531 /* start = */ 0,
\n+
532 /* end = */ stencilTable->GetNumStencils(),
\n+
533 context);
\n+
534 }
\n+
535
\n+
574 bool EvalStencils(id<MTLBuffer> srcBuffer, BufferDescriptor const &srcDesc,
\n+
575 id<MTLBuffer> dstBuffer, BufferDescriptor const &dstDesc,
\n+
576 id<MTLBuffer> duBuffer, BufferDescriptor const &duDesc,
\n+
577 id<MTLBuffer> dvBuffer, BufferDescriptor const &dvDesc,
\n+
578 id<MTLBuffer> sizesBuffer,
\n+
579 id<MTLBuffer> offsetsBuffer,
\n+
580 id<MTLBuffer> indicesBuffer,
\n+
581 id<MTLBuffer> weightsBuffer,
\n+
582 id<MTLBuffer> duWeightsBuffer,
\n+
583 id<MTLBuffer> dvWeightsBuffer,
\n+
584 int start,
\n+
585 int end,
\n+
586 MTLContext* context) const;
\n+
587
\n+
644 bool EvalStencils(id<MTLBuffer> srcBuffer, BufferDescriptor const &srcDesc,
\n+
645 id<MTLBuffer> dstBuffer, BufferDescriptor const &dstDesc,
\n+
646 id<MTLBuffer> duBuffer, BufferDescriptor const &duDesc,
\n+
647 id<MTLBuffer> dvBuffer, BufferDescriptor const &dvDesc,
\n+
648 id<MTLBuffer> duuBuffer, BufferDescriptor const &duuDesc,
\n+
649 id<MTLBuffer> duvBuffer, BufferDescriptor const &duvDesc,
\n+
650 id<MTLBuffer> dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
651 id<MTLBuffer> sizesBuffer,
\n+
652 id<MTLBuffer> offsetsBuffer,
\n+
653 id<MTLBuffer> indicesBuffer,
\n+
654 id<MTLBuffer> weightsBuffer,
\n+
655 id<MTLBuffer> duWeightsBuffer,
\n+
656 id<MTLBuffer> dvWeightsBuffer,
\n+
657 id<MTLBuffer> duuWeightsBuffer,
\n+
658 id<MTLBuffer> duvWeightsBuffer,
\n+
659 id<MTLBuffer> dvvWeightsBuffer,
\n+
660 int start,
\n+
661 int end,
\n+
662 MTLContext* context) const;
\n+
663
\n+
669
\n+
703 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
704 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
705 static bool EvalPatches(
\n+
706 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
707 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
708 int numPatchCoords,
\n+
709 PATCHCOORD_BUFFER *patchCoords,
\n+
710 PATCH_TABLE *patchTable,
\n+
711 MTLComputeEvaluator const *instance,
\n+
712 MTLContext* context) {
\n+
713
\n+
714 if (instance) {
\n+
715 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
716 dstBuffer, dstDesc,
\n+
717 numPatchCoords, patchCoords,
\n+
718 patchTable,
\n+
719 context);
\n+
720 } else {
\n+
721 // Create an instance on demand (slow)
\n+
722 instance = Create(srcDesc, dstDesc,
\n+
723 BufferDescriptor(),
\n+
724 BufferDescriptor(),
\n+
725 context);
\n+
726 if (instance) {
\n+
727 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
728 dstBuffer, dstDesc,
\n+
729 numPatchCoords, patchCoords,
\n+
730 patchTable,
\n+
731 context);
\n+
732 delete instance;
\n+
733 return r;
\n+
734 }
\n+
735 return false;
\n+
736 }
\n+
737 }
\n+
738
\n+
784 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
785 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
786 static bool EvalPatches(
\n+
787 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
788 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
789 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
790 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
791 int numPatchCoords,
\n+
792 PATCHCOORD_BUFFER *patchCoords,
\n+
793 PATCH_TABLE *patchTable,
\n+
794 MTLComputeEvaluator* instance,
\n+
795 MTLContext* context) {
\n+
796
\n+
797 if (instance) {
\n+
798 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
799 dstBuffer, dstDesc,
\n+
800 duBuffer, duDesc,
\n+
801 dvBuffer, dvDesc,
\n+
802 numPatchCoords, patchCoords,
\n+
803 patchTable,
\n+
804 context);
\n+
805 } else {
\n+
806 // Create an instance on demand (slow)
\n+
807 instance = Create(srcDesc, dstDesc,
\n+
808 duDesc, dvDesc, context);
\n+
809 if (instance) {
\n+
810 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
811 dstBuffer, dstDesc,
\n+
812 duBuffer, duDesc,
\n+
813 dvBuffer, dvDesc,
\n+
814 numPatchCoords, patchCoords,
\n+
815 patchTable,
\n+
816 context);
\n+
817 delete instance;
\n+
818 return r;
\n+
819 }
\n+
820 return false;
\n+
821 }
\n+
822 }
\n+
823
\n+
887 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
888 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
889 static bool EvalPatches(
\n+
890 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
891 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
892 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
893 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
894 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
895 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
896 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
897 int numPatchCoords,
\n+
898 PATCHCOORD_BUFFER *patchCoords,
\n+
899 PATCH_TABLE *patchTable,
\n+
900 MTLComputeEvaluator* instance,
\n+
901 MTLContext* context) {
\n+
902
\n+
903 if (instance) {
\n+
904 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
905 dstBuffer, dstDesc,
\n+
906 duBuffer, duDesc,
\n+
907 dvBuffer, dvDesc,
\n+
908 duuBuffer, duuDesc,
\n+
909 duvBuffer, duvDesc,
\n+
910 dvvBuffer, dvvDesc,
\n+
911 numPatchCoords, patchCoords,
\n+
912 patchTable,
\n+
913 context);
\n+
914 } else {
\n+
915 // Create an instance on demand (slow)
\n+
916 instance = Create(srcDesc, dstDesc,
\n+
917 duDesc, dvDesc,
\n+
918 duuDesc, duvDesc, dvvDesc, context);
\n+
919 if (instance) {
\n+
920 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
921 dstBuffer, dstDesc,
\n+
922 duBuffer, duDesc,
\n+
923 dvBuffer, dvDesc,
\n+
924 duuBuffer, duuDesc,
\n+
925 duvBuffer, duvDesc,
\n+
926 dvvBuffer, dvvDesc,
\n+
927 numPatchCoords, patchCoords,
\n+
928 patchTable,
\n+
929 context);
\n+
930 delete instance;
\n+
931 return r;
\n+
932 }
\n+
933 return false;
\n+
934 }
\n+
935 }
\n+
936
\n+
964 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
965 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
966 bool EvalPatches(
\n+
967 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
968 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
969 int numPatchCoords,
\n+
970 PATCHCOORD_BUFFER *patchCoords,
\n+
971 PATCH_TABLE *patchTable,
\n+
972 MTLContext* context) const {
\n+
973
\n+
974 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,
\n+
975 dstBuffer->BindMTLBuffer(context), dstDesc,
\n+
976 0, BufferDescriptor(),
\n+
977 0, BufferDescriptor(),
\n+
978 numPatchCoords,
\n+
979 patchCoords->BindMTLBuffer(context),
\n+
980 patchTable->GetPatchArrays(),
\n+
981 patchTable->GetPatchIndexBuffer(),
\n+
982 patchTable->GetPatchParamBuffer(),
\n+
983 context);
\n+
984 }
\n+
985
\n+
1023 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1024 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1025 bool EvalPatches(
\n+
1026 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1027 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1028 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1029 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1030 int numPatchCoords,
\n+
1031 PATCHCOORD_BUFFER *patchCoords,
\n+
1032 PATCH_TABLE *patchTable,
\n+
1033 MTLContext* context) const {
\n+
1034
\n+
1035 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,
\n+
1036 dstBuffer->BindMTLBuffer(context), dstDesc,
\n+
1037 duBuffer->BindMTLBuffer(context), duDesc,
\n+
1038 dvBuffer->BindMTLBuffer(context), dvDesc,
\n+
1039 numPatchCoords,
\n+
1040 patchCoords->BindMTLBuffer(context),
\n+
1041 patchTable->GetPatchArrays(),
\n+
1042 patchTable->GetPatchIndexBuffer(),
\n+
1043 patchTable->GetPatchParamBuffer(),
\n+
1044 context);
\n+
1045 }
\n+
1046
\n+
1102 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1103 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1104 bool EvalPatches(
\n+
1105 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1106 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1107 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1108 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1109 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1110 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1111 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1112 int numPatchCoords,
\n+
1113 PATCHCOORD_BUFFER *patchCoords,
\n+
1114 PATCH_TABLE *patchTable,
\n+
1115 MTLContext* context) const {
\n+
1116
\n+
1117 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,
\n+
1118 dstBuffer->BindMTLBuffer(context), dstDesc,
\n+
1119 duBuffer->BindMTLBuffer(context), duDesc,
\n+
1120 dvBuffer->BindMTLBuffer(context), dvDesc,
\n+
1121 duuBuffer->BindMTLBuffer(context), duuDesc,
\n+
1122 duvBuffer->BindMTLBuffer(context), duvDesc,
\n+
1123 dvvBuffer->BindMTLBuffer(context), dvvDesc,
\n+
1124 numPatchCoords,
\n+
1125 patchCoords->BindMTLBuffer(context),
\n+
1126 patchTable->GetPatchArrays(),
\n+
1127 patchTable->GetPatchIndexBuffer(),
\n+
1128 patchTable->GetPatchParamBuffer(),
\n+
1129 context);
\n+
1130 }
\n+
1131
\n+
1132 bool EvalPatches(id<MTLBuffer> srcBuffer, BufferDescriptor const &srcDesc,
\n+
1133 id<MTLBuffer> dstBuffer, BufferDescriptor const &dstDesc,
\n+
1134 id<MTLBuffer> duBuffer, BufferDescriptor const &duDesc,
\n+
1135 id<MTLBuffer> dvBuffer, BufferDescriptor const &dvDesc,
\n+
1136 int numPatchCoords,
\n+
1137 id<MTLBuffer> patchCoordsBuffer,
\n+
1138 const PatchArrayVector &patchArrays,
\n+
1139 id<MTLBuffer> patchIndexBuffer,
\n+
1140 id<MTLBuffer> patchParamsBuffer,
\n+
1141 MTLContext* context) const;
\n+
1142
\n+
1143 bool EvalPatches(id<MTLBuffer> srcBuffer, BufferDescriptor const &srcDesc,
\n+
1144 id<MTLBuffer> dstBuffer, BufferDescriptor const &dstDesc,
\n+
1145 id<MTLBuffer> duBuffer, BufferDescriptor const &duDesc,
\n+
1146 id<MTLBuffer> dvBuffer, BufferDescriptor const &dvDesc,
\n+
1147 id<MTLBuffer> duuBuffer, BufferDescriptor const &duuDesc,
\n+
1148 id<MTLBuffer> duvBuffer, BufferDescriptor const &duvDesc,
\n+
1149 id<MTLBuffer> dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1150 int numPatchCoords,
\n+
1151 id<MTLBuffer> patchCoordsBuffer,
\n+
1152 const PatchArrayVector &patchArrays,
\n+
1153 id<MTLBuffer> patchIndexBuffer,
\n+
1154 id<MTLBuffer> patchParamsBuffer,
\n+
1155 MTLContext* context) const;
\n+
1156
\n+
1190 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1191 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1192 static bool EvalPatchesVarying(
\n+
1193 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1194 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1195 int numPatchCoords,
\n+
1196 PATCHCOORD_BUFFER *patchCoords,
\n+
1197 PATCH_TABLE *patchTable,
\n+
1198 MTLComputeEvaluator const *instance,
\n+
1199 MTLContext* deviceContext) {
\n+
1200
\n+
1201 if (instance) {
\n+
1202 return instance->EvalPatchesVarying(
\n+
1203 srcBuffer, srcDesc,
\n+
1204 dstBuffer, dstDesc,
\n+
1205 numPatchCoords, patchCoords,
\n+
1206 patchTable,
\n+
1207 deviceContext);
\n+
1208 } else {
\n+
1209 // Create an instance on demand (slow)
\n+
1210 instance = Create(srcDesc, dstDesc,
\n+
1211 BufferDescriptor(),
\n+
1212 BufferDescriptor(),
\n+
1213 deviceContext);
\n+
1214 if (instance) {
\n+
1215 bool r = instance->EvalPatchesVarying(
\n+
1216 srcBuffer, srcDesc,
\n+
1217 dstBuffer, dstDesc,
\n+
1218 numPatchCoords, patchCoords,
\n+
1219 patchTable,
\n+
1220 deviceContext);
\n+
1221 delete instance;
\n+
1222 return r;
\n+
1223 }
\n+
1224 return false;
\n+
1225 }
\n+
1226 }
\n+
1227
\n+
1255 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1256 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1257 bool EvalPatchesVarying(
\n+
1258 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1259 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1260 int numPatchCoords,
\n+
1261 PATCHCOORD_BUFFER *patchCoords,
\n+
1262 PATCH_TABLE *patchTable,
\n+
1263 MTLContext* deviceContext) const {
\n+
1264
\n+
1265 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n+
1266 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n+
1267 0, BufferDescriptor(),
\n+
1268 0, BufferDescriptor(),
\n+
1269 numPatchCoords,
\n+
1270 patchCoords->BindMTLBuffer(deviceContext),
\n+
1271 patchTable->GetVaryingPatchArrays(),
\n+
1272 patchTable->GetVaryingPatchIndexBuffer(),
\n+
1273 patchTable->GetPatchParamBuffer(),
\n+
1274 deviceContext
\n+
1275 );
\n+
1276 }
\n+
1277
\n+
1323 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1324 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1325 static bool EvalPatchesVarying(
\n+
1326 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1327 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1328 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1329 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1330 int numPatchCoords,
\n+
1331 PATCHCOORD_BUFFER *patchCoords,
\n+
1332 PATCH_TABLE *patchTable,
\n+
1333 MTLComputeEvaluator const *instance,
\n+
1334 MTLContext* deviceContext) {
\n+
1335
\n+
1336 if (instance) {
\n+
1337 return instance->EvalPatchesVarying(
\n+
1338 srcBuffer, srcDesc,
\n+
1339 dstBuffer, dstDesc,
\n+
1340 duBuffer, duDesc,
\n+
1341 dvBuffer, dvDesc,
\n+
1342 numPatchCoords, patchCoords,
\n+
1343 patchTable,
\n+
1344 deviceContext);
\n+
1345 } else {
\n+
1346 // Create an instance on demand (slow)
\n+
1347 instance = Create(srcDesc, dstDesc,
\n+
1348 duDesc, dvDesc,
\n+
1349 deviceContext);
\n+
1350 if (instance) {
\n+
1351 bool r = instance->EvalPatchesVarying(
\n+
1352 srcBuffer, srcDesc,
\n+
1353 dstBuffer, dstDesc,
\n+
1354 duBuffer, duDesc,
\n+
1355 dvBuffer, dvDesc,
\n+
1356 numPatchCoords, patchCoords,
\n+
1357 patchTable,
\n+
1358 deviceContext);
\n+
1359 delete instance;
\n+
1360 return r;
\n+
1361 }
\n+
1362 return false;
\n+
1363 }
\n+
1364 }
\n+
1365
\n+
1405 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1406 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1407 bool EvalPatchesVarying(
\n+
1408 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1409 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1410 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1411 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1412 int numPatchCoords,
\n+
1413 PATCHCOORD_BUFFER *patchCoords,
\n+
1414 PATCH_TABLE *patchTable,
\n+
1415 MTLContext* deviceContext) const {
\n+
1416
\n+
1417 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n+
1418 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n+
1419 duBuffer->BindMTLBuffer(deviceContext), duDesc,
\n+
1420 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,
\n+
1421 numPatchCoords,
\n+
1422 patchCoords->BindMTLBuffer(deviceContext),
\n+
1423 patchTable->GetVaryingPatchArrays(),
\n+
1424 patchTable->GetVaryingPatchIndexBuffer(),
\n+
1425 patchTable->GetPatchParamBuffer(),
\n+
1426 deviceContext
\n+
1427 );
\n+
1428 }
\n+
1429
\n+
1430
\n+
1494 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1495 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1496 static bool EvalPatchesVarying(
\n+
1497 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1498 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1499 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1500 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1501 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1502 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1503 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1504 int numPatchCoords,
\n+
1505 PATCHCOORD_BUFFER *patchCoords,
\n+
1506 PATCH_TABLE *patchTable,
\n+
1507 MTLComputeEvaluator const *instance,
\n+
1508 MTLContext* deviceContext) {
\n+
1509
\n+
1510 if (instance) {
\n+
1511 return instance->EvalPatchesVarying(
\n+
1512 srcBuffer, srcDesc,
\n+
1513 dstBuffer, dstDesc,
\n+
1514 duBuffer, duDesc,
\n+
1515 dvBuffer, dvDesc,
\n+
1516 duuBuffer, duuDesc,
\n+
1517 duvBuffer, duvDesc,
\n+
1518 dvvBuffer, dvvDesc,
\n+
1519 numPatchCoords, patchCoords,
\n+
1520 patchTable,
\n+
1521 deviceContext);
\n+
1522 } else {
\n+
1523 // Create an instance on demand (slow)
\n+
1524 instance = Create(srcDesc, dstDesc,
\n+
1525 duDesc, dvDesc,
\n+
1526 duuDesc, duvDesc, dvvDesc,
\n+
1527 deviceContext);
\n+
1528 if (instance) {
\n+
1529 bool r = instance->EvalPatchesVarying(
\n+
1530 srcBuffer, srcDesc,
\n+
1531 dstBuffer, dstDesc,
\n+
1532 duBuffer, duDesc,
\n+
1533 dvBuffer, dvDesc,
\n+
1534 duuBuffer, duuDesc,
\n+
1535 duvBuffer, duvDesc,
\n+
1536 dvvBuffer, dvvDesc,
\n+
1537 numPatchCoords, patchCoords,
\n+
1538 patchTable,
\n+
1539 deviceContext);
\n+
1540 delete instance;
\n+
1541 return r;
\n+
1542 }
\n+
1543 return false;
\n+
1544 }
\n+
1545 }
\n+
1546
\n+
1604 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1605 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1606 bool EvalPatchesVarying(
\n+
1607 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1608 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1609 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1610 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1611 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1612 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1613 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1614 int numPatchCoords,
\n+
1615 PATCHCOORD_BUFFER *patchCoords,
\n+
1616 PATCH_TABLE *patchTable,
\n+
1617 MTLContext* deviceContext) const {
\n+
1618
\n+
1619 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n+
1620 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n+
1621 duBuffer->BindMTLBuffer(deviceContext), duDesc,
\n+
1622 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,
\n+
1623 duuBuffer->BindMTLBuffer(deviceContext), duuDesc,
\n+
1624 duvBuffer->BindMTLBuffer(deviceContext), duvDesc,
\n+
1625 dvvBuffer->BindMTLBuffer(deviceContext), dvvDesc,
\n+
1626 numPatchCoords,
\n+
1627 patchCoords->BindMTLBuffer(deviceContext),
\n+
1628 patchTable->GetVaryingPatchArrays(),
\n+
1629 patchTable->GetVaryingPatchIndexBuffer(),
\n+
1630 patchTable->GetPatchParamBuffer(),
\n+
1631 deviceContext
\n+
1632 );
\n+
1633 }
\n+
1634
\n+
1670 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1671 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1672 static bool EvalPatchesFaceVarying(
\n+
1673 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1674 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1675 int numPatchCoords,
\n+
1676 PATCHCOORD_BUFFER *patchCoords,
\n+
1677 PATCH_TABLE *patchTable,
\n+
1678 int fvarChannel,
\n+
1679 MTLComputeEvaluator const *instance,
\n+
1680 MTLContext* deviceContext) {
\n+
1681
\n+
1682 if (instance) {
\n+
1683 return instance->EvalPatchesFaceVarying(
\n+
1684 srcBuffer, srcDesc,
\n+
1685 dstBuffer, dstDesc,
\n+
1686 numPatchCoords, patchCoords,
\n+
1687 patchTable,
\n+
1688 fvarChannel,
\n+
1689 deviceContext);
\n+
1690 } else {
\n+
1691 // Create an instance on demand (slow)
\n+
1692 instance = Create(srcDesc, dstDesc,
\n+
1693 BufferDescriptor(),
\n+
1694 BufferDescriptor(),
\n+
1695 deviceContext);
\n+
1696 if (instance) {
\n+
1697 bool r = instance->EvalPatchesFaceVarying(
\n+
1698 srcBuffer, srcDesc,
\n+
1699 dstBuffer, dstDesc,
\n+
1700 numPatchCoords, patchCoords,
\n+
1701 patchTable,
\n+
1702 fvarChannel,
\n+
1703 deviceContext);
\n+
1704 delete instance;
\n+
1705 return r;
\n+
1706 }
\n+
1707 return false;
\n+
1708 }
\n+
1709 }
\n+
1710
\n+
1740 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1741 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1742 bool EvalPatchesFaceVarying(
\n+
1743 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1744 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1745 int numPatchCoords,
\n+
1746 PATCHCOORD_BUFFER *patchCoords,
\n+
1747 PATCH_TABLE *patchTable,
\n+
1748 int fvarChannel,
\n+
1749 MTLContext* deviceContext) const {
\n+
1750
\n+
1751 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n+
1752 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n+
1753 0, BufferDescriptor(),
\n+
1754 0, BufferDescriptor(),
\n+
1755 numPatchCoords,
\n+
1756 patchCoords->BindMTLBuffer(deviceContext),
\n+
1757 patchTable->GetFVarPatchArrays(fvarChannel),
\n+
1758 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1759 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n+
1760 deviceContext
\n+
1761 );
\n+
1762 }
\n+
1763
\n+
1811 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1812 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1813 static bool EvalPatchesFaceVarying(
\n+
1814 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1815 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1816 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1817 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1818 int numPatchCoords,
\n+
1819 PATCHCOORD_BUFFER *patchCoords,
\n+
1820 PATCH_TABLE *patchTable,
\n+
1821 int fvarChannel,
\n+
1822 MTLComputeEvaluator const *instance,
\n+
1823 MTLContext* deviceContext) {
\n+
1824
\n+
1825 if (instance) {
\n+
1826 return instance->EvalPatchesFaceVarying(
\n+
1827 srcBuffer, srcDesc,
\n+
1828 dstBuffer, dstDesc,
\n+
1829 duBuffer, duDesc,
\n+
1830 dvBuffer, dvDesc,
\n+
1831 numPatchCoords, patchCoords,
\n+
1832 patchTable,
\n+
1833 fvarChannel,
\n+
1834 deviceContext);
\n+
1835 } else {
\n+
1836 // Create an instance on demand (slow)
\n+
1837 instance = Create(srcDesc, dstDesc,
\n+
1838 duDesc, dvDesc,
\n+
1839 deviceContext);
\n+
1840 if (instance) {
\n+
1841 bool r = instance->EvalPatchesFaceVarying(
\n+
1842 srcBuffer, srcDesc,
\n+
1843 dstBuffer, dstDesc,
\n+
1844 duBuffer, duDesc,
\n+
1845 dvBuffer, dvDesc,
\n+
1846 numPatchCoords, patchCoords,
\n+
1847 patchTable,
\n+
1848 fvarChannel,
\n+
1849 deviceContext);
\n+
1850 delete instance;
\n+
1851 return r;
\n+
1852 }
\n+
1853 return false;
\n+
1854 }
\n+
1855 }
\n+
1856
\n+
1898 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1899 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1900 bool EvalPatchesFaceVarying(
\n+
1901 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1902 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1903 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1904 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1905 int numPatchCoords,
\n+
1906 PATCHCOORD_BUFFER *patchCoords,
\n+
1907 PATCH_TABLE *patchTable,
\n+
1908 int fvarChannel,
\n+
1909 MTLContext* deviceContext) const {
\n+
1910
\n+
1911 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n+
1912 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n+
1913 duBuffer->BindMTLBuffer(deviceContext), duDesc,
\n+
1914 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,
\n+
1915 numPatchCoords,
\n+
1916 patchCoords->BindMTLBuffer(deviceContext),
\n+
1917 patchTable->GetFVarPatchArrays(fvarChannel),
\n+
1918 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1919 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n+
1920 deviceContext
\n+
1921 );
\n+
1922 }
\n+
1923
\n+
1989 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1990 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1991 static bool EvalPatchesFaceVarying(
\n+
1992 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1993 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1994 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1995 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1996 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1997 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1998 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1999 int numPatchCoords,
\n+
2000 PATCHCOORD_BUFFER *patchCoords,
\n+
2001 PATCH_TABLE *patchTable,
\n+
2002 int fvarChannel,
\n+
2003 MTLComputeEvaluator const *instance,
\n+
2004 MTLContext* deviceContext) {
\n+
2005
\n+
2006 if (instance) {
\n+
2007 return instance->EvalPatchesFaceVarying(
\n+
2008 srcBuffer, srcDesc,
\n+
2009 dstBuffer, dstDesc,
\n+
2010 duBuffer, duDesc,
\n+
2011 dvBuffer, dvDesc,
\n+
2012 duuBuffer, duuDesc,
\n+
2013 duvBuffer, duvDesc,
\n+
2014 dvvBuffer, dvvDesc,
\n+
2015 numPatchCoords, patchCoords,
\n+
2016 patchTable,
\n+
2017 fvarChannel,
\n+
2018 deviceContext);
\n+
2019 } else {
\n+
2020 // Create an instance on demand (slow)
\n+
2021 instance = Create(srcDesc, dstDesc,
\n+
2022 duDesc, dvDesc,
\n+
2023 duuDesc, duvDesc, dvvDesc,
\n+
2024 deviceContext);
\n+
2025 if (instance) {
\n+
2026 bool r = instance->EvalPatchesFaceVarying(
\n+
2027 srcBuffer, srcDesc,
\n+
2028 dstBuffer, dstDesc,
\n+
2029 duBuffer, duDesc,
\n+
2030 dvBuffer, dvDesc,
\n+
2031 duuBuffer, duuDesc,
\n+
2032 duvBuffer, duvDesc,
\n+
2033 dvvBuffer, dvvDesc,
\n+
2034 numPatchCoords, patchCoords,
\n+
2035 patchTable,
\n+
2036 fvarChannel,
\n+
2037 deviceContext);
\n+
2038 delete instance;
\n+
2039 return r;
\n+
2040 }
\n+
2041 return false;
\n+
2042 }
\n+
2043 }
\n+
2044
\n+
2104 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
2105 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
2106 bool EvalPatchesFaceVarying(
\n+
2107 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
2108 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
2109 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
2110 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
2111 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
2112 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
2113 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
2114 int numPatchCoords,
\n+
2115 PATCHCOORD_BUFFER *patchCoords,
\n+
2116 PATCH_TABLE *patchTable,
\n+
2117 int fvarChannel,
\n+
2118 MTLContext* deviceContext) const {
\n+
2119
\n+
2120 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n+
2121 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n+
2122 duBuffer->BindMTLBuffer(deviceContext), duDesc,
\n+
2123 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,
\n+
2124 duuBuffer->BindMTLBuffer(deviceContext), duuDesc,
\n+
2125 duvBuffer->BindMTLBuffer(deviceContext), duvDesc,
\n+
2126 dvvBuffer->BindMTLBuffer(deviceContext), dvvDesc,
\n+
2127 numPatchCoords,
\n+
2128 patchCoords->BindMTLBuffer(deviceContext),
\n+
2129 patchTable->GetFVarPatchArrays(fvarChannel),
\n+
2130 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
2131 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n+
2132 fvarChannel,
\n+
2133 deviceContext
\n+
2134 );
\n+
2135 }
\n+
2136
\n+
2138 bool Compile(BufferDescriptor const &srcDesc,
\n+
2139 BufferDescriptor const &dstDesc,
\n+
2140 BufferDescriptor const &duDesc,
\n+
2141 BufferDescriptor const &dvDesc,
\n+
2142 BufferDescriptor const &duuDesc,
\n+
2143 BufferDescriptor const &duvDesc,
\n+
2144 BufferDescriptor const &dvvDesc,
\n+
2145 MTLContext* context);
\n+
2146
\n+
2148 static void Synchronize(MTLContext* context);
\n+
2149
\n+
2150 private:
\n+
2151
\n+
2152 id<MTLLibrary> _computeLibrary;
\n+
2153 id<MTLComputePipelineState> _evalStencils;
\n+
2154 id<MTLComputePipelineState> _evalPatches;
\n+
2155 id<MTLBuffer> _parameterBuffer;
\n+
2156
\n+
2157 int _workGroupSize;
\n+
2158};
\n+
2159
\n+
2160} //end namespace Osd
\n+
2161
\n+
2162} //end namespace OPENSUBDIV_VERSION
\n+
2163using namespace OPENSUBDIV_VERSION;
\n+
2164
\n+
2165} //end namespace OpenSubdiv
\n+
2166
\n+
2167#endif // OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H
\n \n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n \n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::BindVBO
id< MTLBuffer > BindVBO(MTLContext *context)
Definition: mtlVertexBuffer.h:58
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::CPUMTLVertexBuffer
CPUMTLVertexBuffer(int numElements, int numVertices)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::Create
static CPUMTLVertexBuffer * Create(int numElements, int numVertices, MTLContext *context)
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::UpdateData
void UpdateData(const float *src, int startVertex, int numVertices, MTLContext *context)
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::BindMTLBuffer
id< MTLBuffer > BindMTLBuffer(MTLContext *context)
\n-\n-\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::MTLStencilTable
MTLStencilTable(Far::LimitStencilTable const *stencilTable, MTLContext *context)
\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::MTLStencilTable
MTLStencilTable(Far::StencilTable const *stencilTable, MTLContext *context)
\n+\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::Create
static MTLStencilTable * Create(STENCIL_TABLE *stencilTable, DEVICE_CONTEXT context)
Definition: mtlComputeEvaluator.h:54
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
bool EvalStencils(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc, id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer > duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer, BufferDescriptor const &dvDesc, id< MTLBuffer > duuBuffer, BufferDescriptor const &duuDesc, id< MTLBuffer > duvBuffer, BufferDescriptor const &duvDesc, id< MTLBuffer > dvvBuffer, BufferDescriptor const &dvvDesc, id< MTLBuffer > sizesBuffer, id< MTLBuffer > offsetsBuffer, id< MTLBuffer > indicesBuffer, id< MTLBuffer > weightsBuffer, id< MTLBuffer > duWeightsBuffer, id< MTLBuffer > dvWeightsBuffer, id< MTLBuffer > duuWeightsBuffer, id< MTLBuffer > duvWeightsBuffer, id< MTLBuffer > dvvWeightsBuffer, int start, int end, MTLContext *context) const
Dispatch the MTL compute kernel on GPU asynchronously returns false if the kernel hasn't been compile...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator *instance, MTLContext *context)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:786
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, MTLComputeEvaluator const *instance, MTLContext *context)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: mtlComputeEvaluator.h:312
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1991
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator const *instance, MTLContext *context)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:705
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1606
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, MTLComputeEvaluator const *instance, MTLContext *context)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: mtlComputeEvaluator.h:220
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1900
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1672
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1496
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, MTLContext *context) const
Generic stencil function.
Definition: mtlComputeEvaluator.h:376
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
bool EvalPatches(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc, id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer > duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer, BufferDescriptor const &dvDesc, id< MTLBuffer > duuBuffer, BufferDescriptor const &duuDesc, id< MTLBuffer > duvBuffer, BufferDescriptor const &duvDesc, id< MTLBuffer > dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, id< MTLBuffer > patchCoordsBuffer, const PatchArrayVector &patchArrays, id< MTLBuffer > patchIndexBuffer, id< MTLBuffer > patchParamsBuffer, MTLContext *context) const
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:2106
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Synchronize
static void Synchronize(MTLContext *context)
Wait for the dispatched kernel to finish.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, MTLContext *context) const
Generic stencil function.
Definition: mtlComputeEvaluator.h:430
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator *instance, MTLContext *context)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:889
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1742
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1192
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, MTLContext *context) const
Generic stencil function.
Definition: mtlComputeEvaluator.h:504
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
bool EvalStencils(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc, id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer > duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer, BufferDescriptor const &dvDesc, id< MTLBuffer > sizesBuffer, id< MTLBuffer > offsetsBuffer, id< MTLBuffer > indicesBuffer, id< MTLBuffer > weightsBuffer, id< MTLBuffer > duWeightsBuffer, id< MTLBuffer > dvWeightsBuffer, int start, int end, MTLContext *context) const
Dispatch the MTL compute kernel on GPU asynchronously returns false if the kernel hasn't been compile...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1813
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *context) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:966
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, MTLComputeEvaluator const *instance, MTLContext *context)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: mtlComputeEvaluator.h:149
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1325
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *context) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: mtlComputeEvaluator.h:1104
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *context) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: mtlComputeEvaluator.h:1025
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1407
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1257
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Create
static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, MTLContext *context)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Create
static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, MTLContext *context)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Compile
bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, MTLContext *context)
Configure compute pipline state. Returns false if it fails to create the pipeline state.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
bool EvalPatches(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc, id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer > duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, id< MTLBuffer > patchCoordsBuffer, const PatchArrayVector &patchArrays, id< MTLBuffer > patchIndexBuffer, id< MTLBuffer > patchParamsBuffer, MTLContext *context) const
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-mtlVertexBuffer.h\n+mtlComputeEvaluator.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,101 +30,1388 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H\n- 26#define OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H\n+ 25#ifndef OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H\n+ 26#define OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H\n 27\n 28#include \"../version.h\"\n- 29#include \"../osd/mtlCommon.h\"\n- 30\n- 31@protocol MTLDevice;\n- 32@protocol MTLBuffer;\n+ 29\n+ 30#include \"../osd/types.h\"\n+ 31#include \"../osd/bufferDescriptor.h\"\n+ 32#include \"../osd/mtlCommon.h\"\n 33\n- 34namespace OpenSubdiv {\n- 35namespace OPENSUBDIV_VERSION {\n- 36\n- 37namespace Osd {\n+ 34@protocol MTLDevice;\n+ 35@protocol MTLBuffer;\n+ 36@protocol MTLLibrary;\n+ 37@protocol MTLComputePipelineState;\n 38\n-39class CPUMTLVertexBuffer {\n- 40public:\n-41 static CPUMTLVertexBuffer* Create(int numElements, int numVertices,\n-MTLContext* context);\n- 42\n-43 void UpdateData(const float* src, int startVertex, int numVertices,\n-MTLContext* context);\n- 44\n-45 int GetNumElements() const\n- 46 {\n- 47 return _numElements;\n- 48 }\n+ 39namespace OpenSubdiv {\n+ 40namespace OPENSUBDIV_VERSION {\n+ 41\n+ 42namespace Far {\n+ 43 class PatchTable;\n+ 44 class StencilTable;\n+ 45 class LimitStencilTable;\n+ 46}\n+ 47\n+ 48namespace Osd {\n 49\n-50 int GetNumVertices() const\n- 51 {\n- 52 return _numVertices;\n- 53 }\n- 54\n-55 float* BindCpuBuffer();\n-56 id BindMTLBuffer(MTLContext* context);\n- 57\n-58 id BindVBO(MTLContext* context)\n- 59 {\n- 60 return BindMTLBuffer(context);\n- 61 }\n- 62\n- 63protected:\n+50class MTLStencilTable\n+ 51{\n+ 52public:\n+ 53 template\n+54 static MTLStencilTable* Create(STENCIL_TABLE* stencilTable,\n+ 55 DEVICE_CONTEXT context)\n+ 56 {\n+ 57 return new MTLStencilTable(stencilTable, context);\n+ 58 }\n+ 59\n+ 60\n+61 MTLStencilTable(Far::StencilTable const* stencilTable, MTLContext* context);\n+62 MTLStencilTable(Far::LimitStencilTable const* stencilTable, MTLContext*\n+context);\n+63 ~MTLStencilTable();\n 64\n-65 CPUMTLVertexBuffer(int numElements, int numVertices);\n- 66\n-67 bool allocate(MTLContext* context);\n- 68\n- 69private:\n- 70 int _numElements;\n- 71 int _numVertices;\n- 72 id _buffer;\n- 73 bool _dirty;\n- 74};\n- 75\n- 76} //end namespace Osd\n- 77\n- 78} //end namespace OPENSUBDIV_VERSION\n- 79 using namespace OPENSUBDIV_VERSION;\n- 80\n- 81} //end namespace OpenSubdiv\n- 82\n- 83#endif // OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H\n+65 id GetSizesBuffer() const { return _sizesBuffer; }\n+66 id GetOffsetsBuffer() const { return _offsetsBuffer; }\n+67 id GetIndicesBuffer() const { return _indicesBuffer; }\n+68 id GetWeightsBuffer() const { return _weightsBuffer; }\n+69 id GetDuWeightsBuffer() const { return _duWeightsBuffer; }\n+70 id GetDvWeightsBuffer() const { return _dvWeightsBuffer; }\n+71 id GetDuuWeightsBuffer() const { return _duuWeightsBuffer; }\n+72 id GetDuvWeightsBuffer() const { return _duvWeightsBuffer; }\n+73 id GetDvvWeightsBuffer() const { return _dvvWeightsBuffer; }\n+ 74\n+75 int GetNumStencils() const { return _numStencils; }\n+ 76\n+ 77private:\n+ 78 id _sizesBuffer;\n+ 79 id _offsetsBuffer;\n+ 80 id _indicesBuffer;\n+ 81 id _weightsBuffer;\n+ 82 id _duWeightsBuffer;\n+ 83 id _dvWeightsBuffer;\n+ 84 id _duuWeightsBuffer;\n+ 85 id _duvWeightsBuffer;\n+ 86 id _dvvWeightsBuffer;\n+ 87\n+ 88 int _numStencils;\n+ 89};\n+ 90\n+91class MTLComputeEvaluator\n+ 92{\n+ 93public:\n+94 typedef bool Instantiatable;\n+ 95\n+96 static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n+ 97 BufferDescriptor const &dstDesc,\n+ 98 BufferDescriptor const &duDesc,\n+ 99 BufferDescriptor const &dvDesc,\n+ 100 MTLContext* context);\n+ 101\n+102 static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n+ 103 BufferDescriptor const &dstDesc,\n+ 104 BufferDescriptor const &duDesc,\n+ 105 BufferDescriptor const &dvDesc,\n+ 106 BufferDescriptor const &duuDesc,\n+ 107 BufferDescriptor const &duvDesc,\n+ 108 BufferDescriptor const &dvvDesc,\n+ 109 MTLContext* context);\n+ 110\n+111 MTLComputeEvaluator();\n+112 ~MTLComputeEvaluator();\n+ 113\n+ 119\n+ 148 template \n+149 static bool EvalStencils(\n+ 150 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 151 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 152 STENCIL_TABLE const *stencilTable,\n+ 153 MTLComputeEvaluator const *instance,\n+ 154 MTLContext* context)\n+ 155 {\n+ 156 if (instance) {\n+ 157 return instance->EvalStencils(srcBuffer, srcDesc,\n+ 158 dstBuffer, dstDesc,\n+ 159 stencilTable,\n+ 160 context);\n+ 161 } else {\n+ 162 // Create an instance on demand (slow)\n+ 163 instance = Create(srcDesc, dstDesc,\n+ 164 BufferDescriptor(),\n+ 165 BufferDescriptor(),\n+ 166 context);\n+ 167 if (instance) {\n+ 168 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 169 dstBuffer, dstDesc,\n+ 170 stencilTable,\n+ 171 context);\n+ 172 delete instance;\n+ 173 return r;\n+ 174 }\n+ 175 return false;\n+ 176 }\n+ 177 }\n+ 178\n+ 219 template \n+220 static bool EvalStencils(\n+ 221 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 222 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 223 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 224 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 225 STENCIL_TABLE const *stencilTable,\n+ 226 MTLComputeEvaluator const *instance,\n+ 227 MTLContext* context) {\n+ 228\n+ 229 if (instance) {\n+ 230 return instance->EvalStencils(srcBuffer, srcDesc,\n+ 231 dstBuffer, dstDesc,\n+ 232 duBuffer, duDesc,\n+ 233 dvBuffer, dvDesc,\n+ 234 stencilTable,\n+ 235 context);\n+ 236 } else {\n+ 237 // Create an instance on demand (slow)\n+ 238 instance = Create(srcDesc, dstDesc, duDesc, dvDesc, context);\n+ 239 if (instance) {\n+ 240 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 241 dstBuffer, dstDesc,\n+ 242 duBuffer, duDesc,\n+ 243 dvBuffer, dvDesc,\n+ 244 stencilTable,\n+ 245 context);\n+ 246 delete instance;\n+ 247 return r;\n+ 248 }\n+ 249 return false;\n+ 250 }\n+ 251 }\n+ 252\n+ 311 template \n+312 static bool EvalStencils(\n+ 313 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 314 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 315 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 316 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 317 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 318 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 319 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 320 STENCIL_TABLE const *stencilTable,\n+ 321 MTLComputeEvaluator const *instance,\n+ 322 MTLContext* context) {\n+ 323\n+ 324 if (instance) {\n+ 325 return instance->EvalStencils(srcBuffer, srcDesc,\n+ 326 dstBuffer, dstDesc,\n+ 327 duBuffer, duDesc,\n+ 328 dvBuffer, dvDesc,\n+ 329 duuBuffer, duuDesc,\n+ 330 duvBuffer, duvDesc,\n+ 331 dvvBuffer, dvvDesc,\n+ 332 stencilTable,\n+ 333 context);\n+ 334 } else {\n+ 335 // Create an instance on demand (slow)\n+ 336 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,\n+ 337 duuDesc, duvDesc, dvvDesc, context);\n+ 338 if (instance) {\n+ 339 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 340 dstBuffer, dstDesc,\n+ 341 duBuffer, duDesc,\n+ 342 dvBuffer, dvDesc,\n+ 343 duuBuffer, duuDesc,\n+ 344 duvBuffer, duvDesc,\n+ 345 dvvBuffer, dvvDesc,\n+ 346 stencilTable,\n+ 347 context);\n+ 348 delete instance;\n+ 349 return r;\n+ 350 }\n+ 351 return false;\n+ 352 }\n+ 353 }\n+ 354\n+ 375 template \n+376 bool EvalStencils(\n+ 377 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 378 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 379 STENCIL_TABLE const *stencilTable,\n+ 380 MTLContext* context) const\n+ 381 {\n+ 382 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,\n+ 383 dstBuffer->BindMTLBuffer(context), dstDesc,\n+ 384 0, BufferDescriptor(),\n+ 385 0, BufferDescriptor(),\n+ 386 stencilTable->GetSizesBuffer(),\n+ 387 stencilTable->GetOffsetsBuffer(),\n+ 388 stencilTable->GetIndicesBuffer(),\n+ 389 stencilTable->GetWeightsBuffer(),\n+ 390 0,\n+ 391 0,\n+ 392 /* start = */ 0,\n+ 393 /* end = */ stencilTable->GetNumStencils(),\n+ 394 context);\n+ 395 }\n+ 396\n+ 429 template \n+430 bool EvalStencils(\n+ 431 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 432 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 433 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 434 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 435 STENCIL_TABLE const *stencilTable,\n+ 436 MTLContext* context) const\n+ 437 {\n+ 438 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,\n+ 439 dstBuffer->BindMTLBuffer(context), dstDesc,\n+ 440 duBuffer->BindMTLBuffer(context), duDesc,\n+ 441 dvBuffer->BindMTLBuffer(context), dvDesc,\n+ 442 stencilTable->GetSizesBuffer(),\n+ 443 stencilTable->GetOffsetsBuffer(),\n+ 444 stencilTable->GetIndicesBuffer(),\n+ 445 stencilTable->GetWeightsBuffer(),\n+ 446 stencilTable->GetDuWeightsBuffer(),\n+ 447 stencilTable->GetDvWeightsBuffer(),\n+ 448 /* start = */ 0,\n+ 449 /* end = */ stencilTable->GetNumStencils(),\n+ 450 context);\n+ 451 }\n+ 452\n+ 503 template \n+504 bool EvalStencils(\n+ 505 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 506 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 507 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 508 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 509 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 510 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 511 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 512 STENCIL_TABLE const *stencilTable,\n+ 513 MTLContext* context) const\n+ 514 {\n+ 515 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,\n+ 516 dstBuffer->BindMTLBuffer(context), dstDesc,\n+ 517 duBuffer->BindMTLBuffer(context), duDesc,\n+ 518 dvBuffer->BindMTLBuffer(context), dvDesc,\n+ 519 duuBuffer->BindMTLBuffer(context), duuDesc,\n+ 520 duvBuffer->BindMTLBuffer(context), duvDesc,\n+ 521 dvvBuffer->BindMTLBuffer(context), dvvDesc,\n+ 522 stencilTable->GetSizesBuffer(),\n+ 523 stencilTable->GetOffsetsBuffer(),\n+ 524 stencilTable->GetIndicesBuffer(),\n+ 525 stencilTable->GetWeightsBuffer(),\n+ 526 stencilTable->GetDuWeightsBuffer(),\n+ 527 stencilTable->GetDvWeightsBuffer(),\n+ 528 stencilTable->GetDuuWeightsBuffer(),\n+ 529 stencilTable->GetDuvWeightsBuffer(),\n+ 530 stencilTable->GetDvvWeightsBuffer(),\n+ 531 /* start = */ 0,\n+ 532 /* end = */ stencilTable->GetNumStencils(),\n+ 533 context);\n+ 534 }\n+ 535\n+574 bool EvalStencils(id srcBuffer, BufferDescriptor const &srcDesc,\n+ 575 id dstBuffer, BufferDescriptor const &dstDesc,\n+ 576 id duBuffer, BufferDescriptor const &duDesc,\n+ 577 id dvBuffer, BufferDescriptor const &dvDesc,\n+ 578 id sizesBuffer,\n+ 579 id offsetsBuffer,\n+ 580 id indicesBuffer,\n+ 581 id weightsBuffer,\n+ 582 id duWeightsBuffer,\n+ 583 id dvWeightsBuffer,\n+ 584 int start,\n+ 585 int end,\n+ 586 MTLContext* context) const;\n+ 587\n+644 bool EvalStencils(id srcBuffer, BufferDescriptor const &srcDesc,\n+ 645 id dstBuffer, BufferDescriptor const &dstDesc,\n+ 646 id duBuffer, BufferDescriptor const &duDesc,\n+ 647 id dvBuffer, BufferDescriptor const &dvDesc,\n+ 648 id duuBuffer, BufferDescriptor const &duuDesc,\n+ 649 id duvBuffer, BufferDescriptor const &duvDesc,\n+ 650 id dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 651 id sizesBuffer,\n+ 652 id offsetsBuffer,\n+ 653 id indicesBuffer,\n+ 654 id weightsBuffer,\n+ 655 id duWeightsBuffer,\n+ 656 id dvWeightsBuffer,\n+ 657 id duuWeightsBuffer,\n+ 658 id duvWeightsBuffer,\n+ 659 id dvvWeightsBuffer,\n+ 660 int start,\n+ 661 int end,\n+ 662 MTLContext* context) const;\n+ 663\n+ 669\n+ 703 template \n+705 static bool EvalPatches(\n+ 706 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 707 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 708 int numPatchCoords,\n+ 709 PATCHCOORD_BUFFER *patchCoords,\n+ 710 PATCH_TABLE *patchTable,\n+ 711 MTLComputeEvaluator const *instance,\n+ 712 MTLContext* context) {\n+ 713\n+ 714 if (instance) {\n+ 715 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 716 dstBuffer, dstDesc,\n+ 717 numPatchCoords, patchCoords,\n+ 718 patchTable,\n+ 719 context);\n+ 720 } else {\n+ 721 // Create an instance on demand (slow)\n+ 722 instance = Create(srcDesc, dstDesc,\n+ 723 BufferDescriptor(),\n+ 724 BufferDescriptor(),\n+ 725 context);\n+ 726 if (instance) {\n+ 727 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 728 dstBuffer, dstDesc,\n+ 729 numPatchCoords, patchCoords,\n+ 730 patchTable,\n+ 731 context);\n+ 732 delete instance;\n+ 733 return r;\n+ 734 }\n+ 735 return false;\n+ 736 }\n+ 737 }\n+ 738\n+ 784 template \n+786 static bool EvalPatches(\n+ 787 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 788 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 789 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 790 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 791 int numPatchCoords,\n+ 792 PATCHCOORD_BUFFER *patchCoords,\n+ 793 PATCH_TABLE *patchTable,\n+ 794 MTLComputeEvaluator* instance,\n+ 795 MTLContext* context) {\n+ 796\n+ 797 if (instance) {\n+ 798 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 799 dstBuffer, dstDesc,\n+ 800 duBuffer, duDesc,\n+ 801 dvBuffer, dvDesc,\n+ 802 numPatchCoords, patchCoords,\n+ 803 patchTable,\n+ 804 context);\n+ 805 } else {\n+ 806 // Create an instance on demand (slow)\n+ 807 instance = Create(srcDesc, dstDesc,\n+ 808 duDesc, dvDesc, context);\n+ 809 if (instance) {\n+ 810 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 811 dstBuffer, dstDesc,\n+ 812 duBuffer, duDesc,\n+ 813 dvBuffer, dvDesc,\n+ 814 numPatchCoords, patchCoords,\n+ 815 patchTable,\n+ 816 context);\n+ 817 delete instance;\n+ 818 return r;\n+ 819 }\n+ 820 return false;\n+ 821 }\n+ 822 }\n+ 823\n+ 887 template \n+889 static bool EvalPatches(\n+ 890 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 891 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 892 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 893 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 894 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 895 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 896 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 897 int numPatchCoords,\n+ 898 PATCHCOORD_BUFFER *patchCoords,\n+ 899 PATCH_TABLE *patchTable,\n+ 900 MTLComputeEvaluator* instance,\n+ 901 MTLContext* context) {\n+ 902\n+ 903 if (instance) {\n+ 904 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 905 dstBuffer, dstDesc,\n+ 906 duBuffer, duDesc,\n+ 907 dvBuffer, dvDesc,\n+ 908 duuBuffer, duuDesc,\n+ 909 duvBuffer, duvDesc,\n+ 910 dvvBuffer, dvvDesc,\n+ 911 numPatchCoords, patchCoords,\n+ 912 patchTable,\n+ 913 context);\n+ 914 } else {\n+ 915 // Create an instance on demand (slow)\n+ 916 instance = Create(srcDesc, dstDesc,\n+ 917 duDesc, dvDesc,\n+ 918 duuDesc, duvDesc, dvvDesc, context);\n+ 919 if (instance) {\n+ 920 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 921 dstBuffer, dstDesc,\n+ 922 duBuffer, duDesc,\n+ 923 dvBuffer, dvDesc,\n+ 924 duuBuffer, duuDesc,\n+ 925 duvBuffer, duvDesc,\n+ 926 dvvBuffer, dvvDesc,\n+ 927 numPatchCoords, patchCoords,\n+ 928 patchTable,\n+ 929 context);\n+ 930 delete instance;\n+ 931 return r;\n+ 932 }\n+ 933 return false;\n+ 934 }\n+ 935 }\n+ 936\n+ 964 template \n+966 bool EvalPatches(\n+ 967 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 968 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 969 int numPatchCoords,\n+ 970 PATCHCOORD_BUFFER *patchCoords,\n+ 971 PATCH_TABLE *patchTable,\n+ 972 MTLContext* context) const {\n+ 973\n+ 974 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,\n+ 975 dstBuffer->BindMTLBuffer(context), dstDesc,\n+ 976 0, BufferDescriptor(),\n+ 977 0, BufferDescriptor(),\n+ 978 numPatchCoords,\n+ 979 patchCoords->BindMTLBuffer(context),\n+ 980 patchTable->GetPatchArrays(),\n+ 981 patchTable->GetPatchIndexBuffer(),\n+ 982 patchTable->GetPatchParamBuffer(),\n+ 983 context);\n+ 984 }\n+ 985\n+ 1023 template \n+1025 bool EvalPatches(\n+ 1026 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1027 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1028 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1029 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1030 int numPatchCoords,\n+ 1031 PATCHCOORD_BUFFER *patchCoords,\n+ 1032 PATCH_TABLE *patchTable,\n+ 1033 MTLContext* context) const {\n+ 1034\n+ 1035 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,\n+ 1036 dstBuffer->BindMTLBuffer(context), dstDesc,\n+ 1037 duBuffer->BindMTLBuffer(context), duDesc,\n+ 1038 dvBuffer->BindMTLBuffer(context), dvDesc,\n+ 1039 numPatchCoords,\n+ 1040 patchCoords->BindMTLBuffer(context),\n+ 1041 patchTable->GetPatchArrays(),\n+ 1042 patchTable->GetPatchIndexBuffer(),\n+ 1043 patchTable->GetPatchParamBuffer(),\n+ 1044 context);\n+ 1045 }\n+ 1046\n+ 1102 template \n+1104 bool EvalPatches(\n+ 1105 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1106 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1107 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1108 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1109 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1110 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1111 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1112 int numPatchCoords,\n+ 1113 PATCHCOORD_BUFFER *patchCoords,\n+ 1114 PATCH_TABLE *patchTable,\n+ 1115 MTLContext* context) const {\n+ 1116\n+ 1117 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,\n+ 1118 dstBuffer->BindMTLBuffer(context), dstDesc,\n+ 1119 duBuffer->BindMTLBuffer(context), duDesc,\n+ 1120 dvBuffer->BindMTLBuffer(context), dvDesc,\n+ 1121 duuBuffer->BindMTLBuffer(context), duuDesc,\n+ 1122 duvBuffer->BindMTLBuffer(context), duvDesc,\n+ 1123 dvvBuffer->BindMTLBuffer(context), dvvDesc,\n+ 1124 numPatchCoords,\n+ 1125 patchCoords->BindMTLBuffer(context),\n+ 1126 patchTable->GetPatchArrays(),\n+ 1127 patchTable->GetPatchIndexBuffer(),\n+ 1128 patchTable->GetPatchParamBuffer(),\n+ 1129 context);\n+ 1130 }\n+ 1131\n+1132 bool EvalPatches(id srcBuffer, BufferDescriptor const &srcDesc,\n+ 1133 id dstBuffer, BufferDescriptor const &dstDesc,\n+ 1134 id duBuffer, BufferDescriptor const &duDesc,\n+ 1135 id dvBuffer, BufferDescriptor const &dvDesc,\n+ 1136 int numPatchCoords,\n+ 1137 id patchCoordsBuffer,\n+ 1138 const PatchArrayVector &patchArrays,\n+ 1139 id patchIndexBuffer,\n+ 1140 id patchParamsBuffer,\n+ 1141 MTLContext* context) const;\n+ 1142\n+1143 bool EvalPatches(id srcBuffer, BufferDescriptor const &srcDesc,\n+ 1144 id dstBuffer, BufferDescriptor const &dstDesc,\n+ 1145 id duBuffer, BufferDescriptor const &duDesc,\n+ 1146 id dvBuffer, BufferDescriptor const &dvDesc,\n+ 1147 id duuBuffer, BufferDescriptor const &duuDesc,\n+ 1148 id duvBuffer, BufferDescriptor const &duvDesc,\n+ 1149 id dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1150 int numPatchCoords,\n+ 1151 id patchCoordsBuffer,\n+ 1152 const PatchArrayVector &patchArrays,\n+ 1153 id patchIndexBuffer,\n+ 1154 id patchParamsBuffer,\n+ 1155 MTLContext* context) const;\n+ 1156\n+ 1190 template \n+1192 static bool EvalPatchesVarying(\n+ 1193 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1194 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1195 int numPatchCoords,\n+ 1196 PATCHCOORD_BUFFER *patchCoords,\n+ 1197 PATCH_TABLE *patchTable,\n+ 1198 MTLComputeEvaluator const *instance,\n+ 1199 MTLContext* deviceContext) {\n+ 1200\n+ 1201 if (instance) {\n+ 1202 return instance->EvalPatchesVarying(\n+ 1203 srcBuffer, srcDesc,\n+ 1204 dstBuffer, dstDesc,\n+ 1205 numPatchCoords, patchCoords,\n+ 1206 patchTable,\n+ 1207 deviceContext);\n+ 1208 } else {\n+ 1209 // Create an instance on demand (slow)\n+ 1210 instance = Create(srcDesc, dstDesc,\n+ 1211 BufferDescriptor(),\n+ 1212 BufferDescriptor(),\n+ 1213 deviceContext);\n+ 1214 if (instance) {\n+ 1215 bool r = instance->EvalPatchesVarying(\n+ 1216 srcBuffer, srcDesc,\n+ 1217 dstBuffer, dstDesc,\n+ 1218 numPatchCoords, patchCoords,\n+ 1219 patchTable,\n+ 1220 deviceContext);\n+ 1221 delete instance;\n+ 1222 return r;\n+ 1223 }\n+ 1224 return false;\n+ 1225 }\n+ 1226 }\n+ 1227\n+ 1255 template \n+1257 bool EvalPatchesVarying(\n+ 1258 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1259 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1260 int numPatchCoords,\n+ 1261 PATCHCOORD_BUFFER *patchCoords,\n+ 1262 PATCH_TABLE *patchTable,\n+ 1263 MTLContext* deviceContext) const {\n+ 1264\n+ 1265 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n+ 1266 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n+ 1267 0, BufferDescriptor(),\n+ 1268 0, BufferDescriptor(),\n+ 1269 numPatchCoords,\n+ 1270 patchCoords->BindMTLBuffer(deviceContext),\n+ 1271 patchTable->GetVaryingPatchArrays(),\n+ 1272 patchTable->GetVaryingPatchIndexBuffer(),\n+ 1273 patchTable->GetPatchParamBuffer(),\n+ 1274 deviceContext\n+ 1275 );\n+ 1276 }\n+ 1277\n+ 1323 template \n+1325 static bool EvalPatchesVarying(\n+ 1326 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1327 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1328 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1329 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1330 int numPatchCoords,\n+ 1331 PATCHCOORD_BUFFER *patchCoords,\n+ 1332 PATCH_TABLE *patchTable,\n+ 1333 MTLComputeEvaluator const *instance,\n+ 1334 MTLContext* deviceContext) {\n+ 1335\n+ 1336 if (instance) {\n+ 1337 return instance->EvalPatchesVarying(\n+ 1338 srcBuffer, srcDesc,\n+ 1339 dstBuffer, dstDesc,\n+ 1340 duBuffer, duDesc,\n+ 1341 dvBuffer, dvDesc,\n+ 1342 numPatchCoords, patchCoords,\n+ 1343 patchTable,\n+ 1344 deviceContext);\n+ 1345 } else {\n+ 1346 // Create an instance on demand (slow)\n+ 1347 instance = Create(srcDesc, dstDesc,\n+ 1348 duDesc, dvDesc,\n+ 1349 deviceContext);\n+ 1350 if (instance) {\n+ 1351 bool r = instance->EvalPatchesVarying(\n+ 1352 srcBuffer, srcDesc,\n+ 1353 dstBuffer, dstDesc,\n+ 1354 duBuffer, duDesc,\n+ 1355 dvBuffer, dvDesc,\n+ 1356 numPatchCoords, patchCoords,\n+ 1357 patchTable,\n+ 1358 deviceContext);\n+ 1359 delete instance;\n+ 1360 return r;\n+ 1361 }\n+ 1362 return false;\n+ 1363 }\n+ 1364 }\n+ 1365\n+ 1405 template \n+1407 bool EvalPatchesVarying(\n+ 1408 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1409 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1410 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1411 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1412 int numPatchCoords,\n+ 1413 PATCHCOORD_BUFFER *patchCoords,\n+ 1414 PATCH_TABLE *patchTable,\n+ 1415 MTLContext* deviceContext) const {\n+ 1416\n+ 1417 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n+ 1418 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n+ 1419 duBuffer->BindMTLBuffer(deviceContext), duDesc,\n+ 1420 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,\n+ 1421 numPatchCoords,\n+ 1422 patchCoords->BindMTLBuffer(deviceContext),\n+ 1423 patchTable->GetVaryingPatchArrays(),\n+ 1424 patchTable->GetVaryingPatchIndexBuffer(),\n+ 1425 patchTable->GetPatchParamBuffer(),\n+ 1426 deviceContext\n+ 1427 );\n+ 1428 }\n+ 1429\n+ 1430\n+ 1494 template \n+1496 static bool EvalPatchesVarying(\n+ 1497 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1498 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1499 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1500 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1501 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1502 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1503 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1504 int numPatchCoords,\n+ 1505 PATCHCOORD_BUFFER *patchCoords,\n+ 1506 PATCH_TABLE *patchTable,\n+ 1507 MTLComputeEvaluator const *instance,\n+ 1508 MTLContext* deviceContext) {\n+ 1509\n+ 1510 if (instance) {\n+ 1511 return instance->EvalPatchesVarying(\n+ 1512 srcBuffer, srcDesc,\n+ 1513 dstBuffer, dstDesc,\n+ 1514 duBuffer, duDesc,\n+ 1515 dvBuffer, dvDesc,\n+ 1516 duuBuffer, duuDesc,\n+ 1517 duvBuffer, duvDesc,\n+ 1518 dvvBuffer, dvvDesc,\n+ 1519 numPatchCoords, patchCoords,\n+ 1520 patchTable,\n+ 1521 deviceContext);\n+ 1522 } else {\n+ 1523 // Create an instance on demand (slow)\n+ 1524 instance = Create(srcDesc, dstDesc,\n+ 1525 duDesc, dvDesc,\n+ 1526 duuDesc, duvDesc, dvvDesc,\n+ 1527 deviceContext);\n+ 1528 if (instance) {\n+ 1529 bool r = instance->EvalPatchesVarying(\n+ 1530 srcBuffer, srcDesc,\n+ 1531 dstBuffer, dstDesc,\n+ 1532 duBuffer, duDesc,\n+ 1533 dvBuffer, dvDesc,\n+ 1534 duuBuffer, duuDesc,\n+ 1535 duvBuffer, duvDesc,\n+ 1536 dvvBuffer, dvvDesc,\n+ 1537 numPatchCoords, patchCoords,\n+ 1538 patchTable,\n+ 1539 deviceContext);\n+ 1540 delete instance;\n+ 1541 return r;\n+ 1542 }\n+ 1543 return false;\n+ 1544 }\n+ 1545 }\n+ 1546\n+ 1604 template \n+1606 bool EvalPatchesVarying(\n+ 1607 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1608 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1609 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1610 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1611 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1612 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1613 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1614 int numPatchCoords,\n+ 1615 PATCHCOORD_BUFFER *patchCoords,\n+ 1616 PATCH_TABLE *patchTable,\n+ 1617 MTLContext* deviceContext) const {\n+ 1618\n+ 1619 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n+ 1620 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n+ 1621 duBuffer->BindMTLBuffer(deviceContext), duDesc,\n+ 1622 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,\n+ 1623 duuBuffer->BindMTLBuffer(deviceContext), duuDesc,\n+ 1624 duvBuffer->BindMTLBuffer(deviceContext), duvDesc,\n+ 1625 dvvBuffer->BindMTLBuffer(deviceContext), dvvDesc,\n+ 1626 numPatchCoords,\n+ 1627 patchCoords->BindMTLBuffer(deviceContext),\n+ 1628 patchTable->GetVaryingPatchArrays(),\n+ 1629 patchTable->GetVaryingPatchIndexBuffer(),\n+ 1630 patchTable->GetPatchParamBuffer(),\n+ 1631 deviceContext\n+ 1632 );\n+ 1633 }\n+ 1634\n+ 1670 template \n+1672 static bool EvalPatchesFaceVarying(\n+ 1673 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1674 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1675 int numPatchCoords,\n+ 1676 PATCHCOORD_BUFFER *patchCoords,\n+ 1677 PATCH_TABLE *patchTable,\n+ 1678 int fvarChannel,\n+ 1679 MTLComputeEvaluator const *instance,\n+ 1680 MTLContext* deviceContext) {\n+ 1681\n+ 1682 if (instance) {\n+ 1683 return instance->EvalPatchesFaceVarying(\n+ 1684 srcBuffer, srcDesc,\n+ 1685 dstBuffer, dstDesc,\n+ 1686 numPatchCoords, patchCoords,\n+ 1687 patchTable,\n+ 1688 fvarChannel,\n+ 1689 deviceContext);\n+ 1690 } else {\n+ 1691 // Create an instance on demand (slow)\n+ 1692 instance = Create(srcDesc, dstDesc,\n+ 1693 BufferDescriptor(),\n+ 1694 BufferDescriptor(),\n+ 1695 deviceContext);\n+ 1696 if (instance) {\n+ 1697 bool r = instance->EvalPatchesFaceVarying(\n+ 1698 srcBuffer, srcDesc,\n+ 1699 dstBuffer, dstDesc,\n+ 1700 numPatchCoords, patchCoords,\n+ 1701 patchTable,\n+ 1702 fvarChannel,\n+ 1703 deviceContext);\n+ 1704 delete instance;\n+ 1705 return r;\n+ 1706 }\n+ 1707 return false;\n+ 1708 }\n+ 1709 }\n+ 1710\n+ 1740 template \n+1742 bool EvalPatchesFaceVarying(\n+ 1743 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1744 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1745 int numPatchCoords,\n+ 1746 PATCHCOORD_BUFFER *patchCoords,\n+ 1747 PATCH_TABLE *patchTable,\n+ 1748 int fvarChannel,\n+ 1749 MTLContext* deviceContext) const {\n+ 1750\n+ 1751 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n+ 1752 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n+ 1753 0, BufferDescriptor(),\n+ 1754 0, BufferDescriptor(),\n+ 1755 numPatchCoords,\n+ 1756 patchCoords->BindMTLBuffer(deviceContext),\n+ 1757 patchTable->GetFVarPatchArrays(fvarChannel),\n+ 1758 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1759 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n+ 1760 deviceContext\n+ 1761 );\n+ 1762 }\n+ 1763\n+ 1811 template \n+1813 static bool EvalPatchesFaceVarying(\n+ 1814 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1815 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1816 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1817 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1818 int numPatchCoords,\n+ 1819 PATCHCOORD_BUFFER *patchCoords,\n+ 1820 PATCH_TABLE *patchTable,\n+ 1821 int fvarChannel,\n+ 1822 MTLComputeEvaluator const *instance,\n+ 1823 MTLContext* deviceContext) {\n+ 1824\n+ 1825 if (instance) {\n+ 1826 return instance->EvalPatchesFaceVarying(\n+ 1827 srcBuffer, srcDesc,\n+ 1828 dstBuffer, dstDesc,\n+ 1829 duBuffer, duDesc,\n+ 1830 dvBuffer, dvDesc,\n+ 1831 numPatchCoords, patchCoords,\n+ 1832 patchTable,\n+ 1833 fvarChannel,\n+ 1834 deviceContext);\n+ 1835 } else {\n+ 1836 // Create an instance on demand (slow)\n+ 1837 instance = Create(srcDesc, dstDesc,\n+ 1838 duDesc, dvDesc,\n+ 1839 deviceContext);\n+ 1840 if (instance) {\n+ 1841 bool r = instance->EvalPatchesFaceVarying(\n+ 1842 srcBuffer, srcDesc,\n+ 1843 dstBuffer, dstDesc,\n+ 1844 duBuffer, duDesc,\n+ 1845 dvBuffer, dvDesc,\n+ 1846 numPatchCoords, patchCoords,\n+ 1847 patchTable,\n+ 1848 fvarChannel,\n+ 1849 deviceContext);\n+ 1850 delete instance;\n+ 1851 return r;\n+ 1852 }\n+ 1853 return false;\n+ 1854 }\n+ 1855 }\n+ 1856\n+ 1898 template \n+1900 bool EvalPatchesFaceVarying(\n+ 1901 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1902 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1903 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1904 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1905 int numPatchCoords,\n+ 1906 PATCHCOORD_BUFFER *patchCoords,\n+ 1907 PATCH_TABLE *patchTable,\n+ 1908 int fvarChannel,\n+ 1909 MTLContext* deviceContext) const {\n+ 1910\n+ 1911 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n+ 1912 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n+ 1913 duBuffer->BindMTLBuffer(deviceContext), duDesc,\n+ 1914 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,\n+ 1915 numPatchCoords,\n+ 1916 patchCoords->BindMTLBuffer(deviceContext),\n+ 1917 patchTable->GetFVarPatchArrays(fvarChannel),\n+ 1918 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1919 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n+ 1920 deviceContext\n+ 1921 );\n+ 1922 }\n+ 1923\n+ 1989 template \n+1991 static bool EvalPatchesFaceVarying(\n+ 1992 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1993 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1994 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1995 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1996 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1997 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1998 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1999 int numPatchCoords,\n+ 2000 PATCHCOORD_BUFFER *patchCoords,\n+ 2001 PATCH_TABLE *patchTable,\n+ 2002 int fvarChannel,\n+ 2003 MTLComputeEvaluator const *instance,\n+ 2004 MTLContext* deviceContext) {\n+ 2005\n+ 2006 if (instance) {\n+ 2007 return instance->EvalPatchesFaceVarying(\n+ 2008 srcBuffer, srcDesc,\n+ 2009 dstBuffer, dstDesc,\n+ 2010 duBuffer, duDesc,\n+ 2011 dvBuffer, dvDesc,\n+ 2012 duuBuffer, duuDesc,\n+ 2013 duvBuffer, duvDesc,\n+ 2014 dvvBuffer, dvvDesc,\n+ 2015 numPatchCoords, patchCoords,\n+ 2016 patchTable,\n+ 2017 fvarChannel,\n+ 2018 deviceContext);\n+ 2019 } else {\n+ 2020 // Create an instance on demand (slow)\n+ 2021 instance = Create(srcDesc, dstDesc,\n+ 2022 duDesc, dvDesc,\n+ 2023 duuDesc, duvDesc, dvvDesc,\n+ 2024 deviceContext);\n+ 2025 if (instance) {\n+ 2026 bool r = instance->EvalPatchesFaceVarying(\n+ 2027 srcBuffer, srcDesc,\n+ 2028 dstBuffer, dstDesc,\n+ 2029 duBuffer, duDesc,\n+ 2030 dvBuffer, dvDesc,\n+ 2031 duuBuffer, duuDesc,\n+ 2032 duvBuffer, duvDesc,\n+ 2033 dvvBuffer, dvvDesc,\n+ 2034 numPatchCoords, patchCoords,\n+ 2035 patchTable,\n+ 2036 fvarChannel,\n+ 2037 deviceContext);\n+ 2038 delete instance;\n+ 2039 return r;\n+ 2040 }\n+ 2041 return false;\n+ 2042 }\n+ 2043 }\n+ 2044\n+ 2104 template \n+2106 bool EvalPatchesFaceVarying(\n+ 2107 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 2108 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 2109 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 2110 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 2111 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 2112 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 2113 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 2114 int numPatchCoords,\n+ 2115 PATCHCOORD_BUFFER *patchCoords,\n+ 2116 PATCH_TABLE *patchTable,\n+ 2117 int fvarChannel,\n+ 2118 MTLContext* deviceContext) const {\n+ 2119\n+ 2120 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n+ 2121 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n+ 2122 duBuffer->BindMTLBuffer(deviceContext), duDesc,\n+ 2123 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,\n+ 2124 duuBuffer->BindMTLBuffer(deviceContext), duuDesc,\n+ 2125 duvBuffer->BindMTLBuffer(deviceContext), duvDesc,\n+ 2126 dvvBuffer->BindMTLBuffer(deviceContext), dvvDesc,\n+ 2127 numPatchCoords,\n+ 2128 patchCoords->BindMTLBuffer(deviceContext),\n+ 2129 patchTable->GetFVarPatchArrays(fvarChannel),\n+ 2130 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 2131 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n+ 2132 fvarChannel,\n+ 2133 deviceContext\n+ 2134 );\n+ 2135 }\n+ 2136\n+2138 bool Compile(BufferDescriptor const &srcDesc,\n+ 2139 BufferDescriptor const &dstDesc,\n+ 2140 BufferDescriptor const &duDesc,\n+ 2141 BufferDescriptor const &dvDesc,\n+ 2142 BufferDescriptor const &duuDesc,\n+ 2143 BufferDescriptor const &duvDesc,\n+ 2144 BufferDescriptor const &dvvDesc,\n+ 2145 MTLContext* context);\n+ 2146\n+2148 static void Synchronize(MTLContext* context);\n+ 2149\n+ 2150 private:\n+ 2151\n+ 2152 id _computeLibrary;\n+ 2153 id _evalStencils;\n+ 2154 id _evalPatches;\n+ 2155 id _parameterBuffer;\n+ 2156\n+ 2157 int _workGroupSize;\n+ 2158};\n+ 2159\n+ 2160} //end namespace Osd\n+ 2161\n+ 2162} //end namespace OPENSUBDIV_VERSION\n+ 2163using namespace OPENSUBDIV_VERSION;\n+ 2164\n+ 2165} //end namespace OpenSubdiv\n+ 2166\n+ 2167#endif // OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H\n OpenSubdiv\n Definition: limits.h:32\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n+std::vector< PatchArray > PatchArrayVector\n+Definition: types.h:132\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n+Stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:273\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n+Limit stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:583\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n+BufferDescriptor is a struct which describes buffer elements in interleaved\n+data buffers....\n+Definition: bufferDescriptor.h:61\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLContext\n Definition: mtlCommon.h:40\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer\n-Definition: mtlVertexBuffer.h:39\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::BindVBO\n-id< MTLBuffer > BindVBO(MTLContext *context)\n-Definition: mtlVertexBuffer.h:58\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::CPUMTLVertexBuffer\n-CPUMTLVertexBuffer(int numElements, int numVertices)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::Create\n-static CPUMTLVertexBuffer * Create(int numElements, int numVertices, MTLContext\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable\n+Definition: mtlComputeEvaluator.h:51\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::MTLStencilTable\n+MTLStencilTable(Far::LimitStencilTable const *stencilTable, MTLContext\n *context)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::GetNumElements\n-int GetNumElements() const\n-Definition: mtlVertexBuffer.h:45\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::UpdateData\n-void UpdateData(const float *src, int startVertex, int numVertices, MTLContext\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetDuWeightsBuffer\n+id< MTLBuffer > GetDuWeightsBuffer() const\n+Definition: mtlComputeEvaluator.h:69\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::~MTLStencilTable\n+~MTLStencilTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetDvvWeightsBuffer\n+id< MTLBuffer > GetDvvWeightsBuffer() const\n+Definition: mtlComputeEvaluator.h:73\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::MTLStencilTable\n+MTLStencilTable(Far::StencilTable const *stencilTable, MTLContext *context)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetDuuWeightsBuffer\n+id< MTLBuffer > GetDuuWeightsBuffer() const\n+Definition: mtlComputeEvaluator.h:71\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetDuvWeightsBuffer\n+id< MTLBuffer > GetDuvWeightsBuffer() const\n+Definition: mtlComputeEvaluator.h:72\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetNumStencils\n+int GetNumStencils() const\n+Definition: mtlComputeEvaluator.h:75\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetIndicesBuffer\n+id< MTLBuffer > GetIndicesBuffer() const\n+Definition: mtlComputeEvaluator.h:67\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetSizesBuffer\n+id< MTLBuffer > GetSizesBuffer() const\n+Definition: mtlComputeEvaluator.h:65\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetOffsetsBuffer\n+id< MTLBuffer > GetOffsetsBuffer() const\n+Definition: mtlComputeEvaluator.h:66\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetWeightsBuffer\n+id< MTLBuffer > GetWeightsBuffer() const\n+Definition: mtlComputeEvaluator.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::Create\n+static MTLStencilTable * Create(STENCIL_TABLE *stencilTable, DEVICE_CONTEXT\n+context)\n+Definition: mtlComputeEvaluator.h:54\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetDvWeightsBuffer\n+id< MTLBuffer > GetDvWeightsBuffer() const\n+Definition: mtlComputeEvaluator.h:70\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator\n+Definition: mtlComputeEvaluator.h:92\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n+bool EvalStencils(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc,\n+id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer >\n+duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer,\n+BufferDescriptor const &dvDesc, id< MTLBuffer > duuBuffer, BufferDescriptor\n+const &duuDesc, id< MTLBuffer > duvBuffer, BufferDescriptor const &duvDesc, id<\n+MTLBuffer > dvvBuffer, BufferDescriptor const &dvvDesc, id< MTLBuffer >\n+sizesBuffer, id< MTLBuffer > offsetsBuffer, id< MTLBuffer > indicesBuffer, id<\n+MTLBuffer > weightsBuffer, id< MTLBuffer > duWeightsBuffer, id< MTLBuffer >\n+dvWeightsBuffer, id< MTLBuffer > duuWeightsBuffer, id< MTLBuffer >\n+duvWeightsBuffer, id< MTLBuffer > dvvWeightsBuffer, int start, int end,\n+MTLContext *context) const\n+Dispatch the MTL compute kernel on GPU asynchronously returns false if the\n+kernel hasn't been compile...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable, MTLComputeEvaluator *instance, MTLContext *context)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:786\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n+MTLComputeEvaluator const *instance, MTLContext *context)\n+Generic static stencil function. This function has a same signature as other\n+device kernels have so t...\n+Definition: mtlComputeEvaluator.h:312\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n+EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n+MTLComputeEvaluator const *instance, MTLContext *deviceContext)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:1991\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::MTLComputeEvaluator\n+MTLComputeEvaluator()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::~MTLComputeEvaluator\n+~MTLComputeEvaluator()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator\n+const *instance, MTLContext *context)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:705\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Instantiatable\n+bool Instantiatable\n+Definition: mtlComputeEvaluator.h:94\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n+bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, MTLContext *deviceContext) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:1606\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable,\n+MTLComputeEvaluator const *instance, MTLContext *context)\n+Generic static stencil function. This function has a same signature as other\n+device kernels have so t...\n+Definition: mtlComputeEvaluator.h:220\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n+EvalPatchesFaceVarying\n+bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLContext\n+*deviceContext) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:1900\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n+EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n+fvarChannel, MTLComputeEvaluator const *instance, MTLContext *deviceContext)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:1672\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator\n+const *instance, MTLContext *deviceContext)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:1496\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n+bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const\n+*stencilTable, MTLContext *context) const\n+Generic stencil function.\n+Definition: mtlComputeEvaluator.h:376\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n+bool EvalPatches(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc,\n+id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer >\n+duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer,\n+BufferDescriptor const &dvDesc, id< MTLBuffer > duuBuffer, BufferDescriptor\n+const &duuDesc, id< MTLBuffer > duvBuffer, BufferDescriptor const &duvDesc, id<\n+MTLBuffer > dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, id<\n+MTLBuffer > patchCoordsBuffer, const PatchArrayVector &patchArrays, id<\n+MTLBuffer > patchIndexBuffer, id< MTLBuffer > patchParamsBuffer, MTLContext\n+*context) const\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n+EvalPatchesFaceVarying\n+bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n+MTLContext *deviceContext) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:2106\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Synchronize\n+static void Synchronize(MTLContext *context)\n+Wait for the dispatched kernel to finish.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n+bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, STENCIL_TABLE const *stencilTable, MTLContext *context) const\n+Generic stencil function.\n+Definition: mtlComputeEvaluator.h:430\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator *instance,\n+MTLContext *context)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:889\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n+EvalPatchesFaceVarying\n+bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n+fvarChannel, MTLContext *deviceContext) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:1742\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n+MTLComputeEvaluator const *instance, MTLContext *deviceContext)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:1192\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n+bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, MTLContext\n+*context) const\n+Generic stencil function.\n+Definition: mtlComputeEvaluator.h:504\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n+bool EvalStencils(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc,\n+id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer >\n+duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer,\n+BufferDescriptor const &dvDesc, id< MTLBuffer > sizesBuffer, id< MTLBuffer >\n+offsetsBuffer, id< MTLBuffer > indicesBuffer, id< MTLBuffer > weightsBuffer,\n+id< MTLBuffer > duWeightsBuffer, id< MTLBuffer > dvWeightsBuffer, int start,\n+int end, MTLContext *context) const\n+Dispatch the MTL compute kernel on GPU asynchronously returns false if the\n+kernel hasn't been compile...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n+EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLComputeEvaluator\n+const *instance, MTLContext *deviceContext)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:1813\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n+bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *context)\n+const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:966\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n+const *stencilTable, MTLComputeEvaluator const *instance, MTLContext *context)\n+Generic static stencil function. This function has a same signature as other\n+device kernels have so t...\n+Definition: mtlComputeEvaluator.h:149\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator const *instance,\n+MTLContext *deviceContext)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:1325\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n+bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, MTLContext *context) const\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: mtlComputeEvaluator.h:1104\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n+bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable, MTLContext *context) const\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: mtlComputeEvaluator.h:1025\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n+bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable, MTLContext *deviceContext) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:1407\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n+bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext\n+*deviceContext) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: mtlComputeEvaluator.h:1257\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Create\n+static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, MTLContext *context)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Create\n+static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n+BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, MTLContext\n *context)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::GetNumVertices\n-int GetNumVertices() const\n-Definition: mtlVertexBuffer.h:50\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::BindMTLBuffer\n-id< MTLBuffer > BindMTLBuffer(MTLContext *context)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::BindCpuBuffer\n-float * BindCpuBuffer()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::allocate\n-bool allocate(MTLContext *context)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Compile\n+bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc,\n+BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc,\n+BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc,\n+BufferDescriptor const &dvvDesc, MTLContext *context)\n+Configure compute pipline state. Returns false if it fails to create the\n+pipeline state.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n+bool EvalPatches(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc,\n+id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer >\n+duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, id< MTLBuffer >\n+patchCoordsBuffer, const PatchArrayVector &patchArrays, id< MTLBuffer >\n+patchIndexBuffer, id< MTLBuffer > patchParamsBuffer, MTLContext *context) const\n * opensubdiv\n * osd\n- * mtlVertexBuffer.h\n+ * mtlComputeEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00824.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00824.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/mtlComputeEvaluator.h File Reference\n+OpenSubdiv: opensubdiv/osd/mtlVertexBuffer.h File Reference\n \n \n \n \n \n \n \n@@ -83,46 +83,40 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
mtlComputeEvaluator.h File Reference
\n+
mtlVertexBuffer.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../osd/types.h"
\n-#include "../osd/bufferDescriptor.h"
\n #include "../osd/mtlCommon.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n-\n+\n \n

\n Classes

class  MTLStencilTable
 
class  MTLComputeEvaluator
class  CPUMTLVertexBuffer
 
\n \n \n \n \n \n-\n-\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,31 +5,25 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-mtlComputeEvaluator.h File Reference\n+mtlVertexBuffer.h File Reference\n #include \"../version.h\"\n-#include \"../osd/types.h\"\n-#include \"../osd/bufferDescriptor.h\"\n #include \"../osd/mtlCommon.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0MTLStencilTable\n-\u00a0\n-class \u00a0MTLComputeEvaluator\n+class \u00a0CPUMTLVertexBuffer\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n-\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * mtlComputeEvaluator.h\n+ * mtlVertexBuffer.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00824.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00824.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,3 @@\n var a00824 = [\n- [\"MTLStencilTable\", \"a01281.html\", \"a01281\"],\n- [\"MTLComputeEvaluator\", \"a01285.html\", \"a01285\"]\n+ [\"CPUMTLVertexBuffer\", \"a01301.html\", \"a01301\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00824_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00824_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/mtlComputeEvaluator.h Source File\n+OpenSubdiv: opensubdiv/osd/mtlVertexBuffer.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
mtlComputeEvaluator.h
\n+
mtlVertexBuffer.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,1084 +107,89 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H
\n-
26#define OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H
\n+
25#ifndef OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H
\n+
26#define OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H
\n
27
\n
28#include "../version.h"
\n-
29
\n-
30#include "../osd/types.h"
\n-
31#include "../osd/bufferDescriptor.h"
\n-
32#include "../osd/mtlCommon.h"
\n+
29#include "../osd/mtlCommon.h"
\n+
30
\n+
31@protocol MTLDevice;
\n+
32@protocol MTLBuffer;
\n
33
\n-
34@protocol MTLDevice;
\n-
35@protocol MTLBuffer;
\n-
36@protocol MTLLibrary;
\n-
37@protocol MTLComputePipelineState;
\n+
34namespace OpenSubdiv {
\n+
35namespace OPENSUBDIV_VERSION {
\n+
36
\n+
37namespace Osd {
\n
38
\n-
39namespace OpenSubdiv {
\n-
40namespace OPENSUBDIV_VERSION {
\n-
41
\n-
42namespace Far {
\n-
43 class PatchTable;
\n-
44 class StencilTable;
\n-
45 class LimitStencilTable;
\n-
46}
\n-
47
\n-
48namespace Osd {
\n+
39class CPUMTLVertexBuffer {
\n+
40public:
\n+
41 static CPUMTLVertexBuffer* Create(int numElements, int numVertices, MTLContext* context);
\n+
42
\n+
43 void UpdateData(const float* src, int startVertex, int numVertices, MTLContext* context);
\n+
44
\n+
45 int GetNumElements() const
\n+
46 {
\n+
47 return _numElements;
\n+
48 }
\n
49
\n-
50class MTLStencilTable
\n-
51{
\n-
52public:
\n-
53 template<typename STENCIL_TABLE, typename DEVICE_CONTEXT>
\n-
54 static MTLStencilTable* Create(STENCIL_TABLE* stencilTable,
\n-
55 DEVICE_CONTEXT context)
\n-
56 {
\n-
57 return new MTLStencilTable(stencilTable, context);
\n-
58 }
\n-
59
\n-
60
\n-
61 MTLStencilTable(Far::StencilTable const* stencilTable, MTLContext* context);
\n-
62 MTLStencilTable(Far::LimitStencilTable const* stencilTable, MTLContext* context);
\n-
63 ~MTLStencilTable();
\n+
50 int GetNumVertices() const
\n+
51 {
\n+
52 return _numVertices;
\n+
53 }
\n+
54
\n+
55 float* BindCpuBuffer();
\n+
56 id<MTLBuffer> BindMTLBuffer(MTLContext* context);
\n+
57
\n+
58 id<MTLBuffer> BindVBO(MTLContext* context)
\n+
59 {
\n+
60 return BindMTLBuffer(context);
\n+
61 }
\n+
62
\n+
63protected:
\n
64
\n-
65 id<MTLBuffer> GetSizesBuffer() const { return _sizesBuffer; }
\n-
66 id<MTLBuffer> GetOffsetsBuffer() const { return _offsetsBuffer; }
\n-
67 id<MTLBuffer> GetIndicesBuffer() const { return _indicesBuffer; }
\n-
68 id<MTLBuffer> GetWeightsBuffer() const { return _weightsBuffer; }
\n-
69 id<MTLBuffer> GetDuWeightsBuffer() const { return _duWeightsBuffer; }
\n-
70 id<MTLBuffer> GetDvWeightsBuffer() const { return _dvWeightsBuffer; }
\n-
71 id<MTLBuffer> GetDuuWeightsBuffer() const { return _duuWeightsBuffer; }
\n-
72 id<MTLBuffer> GetDuvWeightsBuffer() const { return _duvWeightsBuffer; }
\n-
73 id<MTLBuffer> GetDvvWeightsBuffer() const { return _dvvWeightsBuffer; }
\n-
74
\n-
75 int GetNumStencils() const { return _numStencils; }
\n-
76
\n-
77private:
\n-
78 id<MTLBuffer> _sizesBuffer;
\n-
79 id<MTLBuffer> _offsetsBuffer;
\n-
80 id<MTLBuffer> _indicesBuffer;
\n-
81 id<MTLBuffer> _weightsBuffer;
\n-
82 id<MTLBuffer> _duWeightsBuffer;
\n-
83 id<MTLBuffer> _dvWeightsBuffer;
\n-
84 id<MTLBuffer> _duuWeightsBuffer;
\n-
85 id<MTLBuffer> _duvWeightsBuffer;
\n-
86 id<MTLBuffer> _dvvWeightsBuffer;
\n-
87
\n-
88 int _numStencils;
\n-
89};
\n-
90
\n-
91class MTLComputeEvaluator
\n-
92{
\n-
93public:
\n-
94 typedef bool Instantiatable;
\n-
95
\n-
96 static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n-
97 BufferDescriptor const &dstDesc,
\n-
98 BufferDescriptor const &duDesc,
\n-
99 BufferDescriptor const &dvDesc,
\n-
100 MTLContext* context);
\n-
101
\n-
102 static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n-
103 BufferDescriptor const &dstDesc,
\n-
104 BufferDescriptor const &duDesc,
\n-
105 BufferDescriptor const &dvDesc,
\n-
106 BufferDescriptor const &duuDesc,
\n-
107 BufferDescriptor const &duvDesc,
\n-
108 BufferDescriptor const &dvvDesc,
\n-
109 MTLContext* context);
\n-
110
\n-
111 MTLComputeEvaluator();
\n-
112 ~MTLComputeEvaluator();
\n-
113
\n-
119
\n-
148 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
149 static bool EvalStencils(
\n-
150 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
151 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
152 STENCIL_TABLE const *stencilTable,
\n-
153 MTLComputeEvaluator const *instance,
\n-
154 MTLContext* context)
\n-
155 {
\n-
156 if (instance) {
\n-
157 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
158 dstBuffer, dstDesc,
\n-
159 stencilTable,
\n-
160 context);
\n-
161 } else {
\n-
162 // Create an instance on demand (slow)
\n-
163 instance = Create(srcDesc, dstDesc,
\n-
164 BufferDescriptor(),
\n-
165 BufferDescriptor(),
\n-
166 context);
\n-
167 if (instance) {
\n-
168 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
169 dstBuffer, dstDesc,
\n-
170 stencilTable,
\n-
171 context);
\n-
172 delete instance;
\n-
173 return r;
\n-
174 }
\n-
175 return false;
\n-
176 }
\n-
177 }
\n-
178
\n-
219 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
220 static bool EvalStencils(
\n-
221 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
222 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
223 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
224 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
225 STENCIL_TABLE const *stencilTable,
\n-
226 MTLComputeEvaluator const *instance,
\n-
227 MTLContext* context) {
\n-
228
\n-
229 if (instance) {
\n-
230 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
231 dstBuffer, dstDesc,
\n-
232 duBuffer, duDesc,
\n-
233 dvBuffer, dvDesc,
\n-
234 stencilTable,
\n-
235 context);
\n-
236 } else {
\n-
237 // Create an instance on demand (slow)
\n-
238 instance = Create(srcDesc, dstDesc, duDesc, dvDesc, context);
\n-
239 if (instance) {
\n-
240 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
241 dstBuffer, dstDesc,
\n-
242 duBuffer, duDesc,
\n-
243 dvBuffer, dvDesc,
\n-
244 stencilTable,
\n-
245 context);
\n-
246 delete instance;
\n-
247 return r;
\n-
248 }
\n-
249 return false;
\n-
250 }
\n-
251 }
\n-
252
\n-
311 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
312 static bool EvalStencils(
\n-
313 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
314 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
315 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
316 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
317 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
318 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
319 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
320 STENCIL_TABLE const *stencilTable,
\n-
321 MTLComputeEvaluator const *instance,
\n-
322 MTLContext* context) {
\n-
323
\n-
324 if (instance) {
\n-
325 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
326 dstBuffer, dstDesc,
\n-
327 duBuffer, duDesc,
\n-
328 dvBuffer, dvDesc,
\n-
329 duuBuffer, duuDesc,
\n-
330 duvBuffer, duvDesc,
\n-
331 dvvBuffer, dvvDesc,
\n-
332 stencilTable,
\n-
333 context);
\n-
334 } else {
\n-
335 // Create an instance on demand (slow)
\n-
336 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,
\n-
337 duuDesc, duvDesc, dvvDesc, context);
\n-
338 if (instance) {
\n-
339 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
340 dstBuffer, dstDesc,
\n-
341 duBuffer, duDesc,
\n-
342 dvBuffer, dvDesc,
\n-
343 duuBuffer, duuDesc,
\n-
344 duvBuffer, duvDesc,
\n-
345 dvvBuffer, dvvDesc,
\n-
346 stencilTable,
\n-
347 context);
\n-
348 delete instance;
\n-
349 return r;
\n-
350 }
\n-
351 return false;
\n-
352 }
\n-
353 }
\n-
354
\n-
375 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
376 bool EvalStencils(
\n-
377 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
378 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
379 STENCIL_TABLE const *stencilTable,
\n-
380 MTLContext* context) const
\n-
381 {
\n-
382 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,
\n-
383 dstBuffer->BindMTLBuffer(context), dstDesc,
\n-
384 0, BufferDescriptor(),
\n-
385 0, BufferDescriptor(),
\n-
386 stencilTable->GetSizesBuffer(),
\n-
387 stencilTable->GetOffsetsBuffer(),
\n-
388 stencilTable->GetIndicesBuffer(),
\n-
389 stencilTable->GetWeightsBuffer(),
\n-
390 0,
\n-
391 0,
\n-
392 /* start = */ 0,
\n-
393 /* end = */ stencilTable->GetNumStencils(),
\n-
394 context);
\n-
395 }
\n-
396
\n-
429 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
430 bool EvalStencils(
\n-
431 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
432 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
433 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
434 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
435 STENCIL_TABLE const *stencilTable,
\n-
436 MTLContext* context) const
\n-
437 {
\n-
438 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,
\n-
439 dstBuffer->BindMTLBuffer(context), dstDesc,
\n-
440 duBuffer->BindMTLBuffer(context), duDesc,
\n-
441 dvBuffer->BindMTLBuffer(context), dvDesc,
\n-
442 stencilTable->GetSizesBuffer(),
\n-
443 stencilTable->GetOffsetsBuffer(),
\n-
444 stencilTable->GetIndicesBuffer(),
\n-
445 stencilTable->GetWeightsBuffer(),
\n-
446 stencilTable->GetDuWeightsBuffer(),
\n-
447 stencilTable->GetDvWeightsBuffer(),
\n-
448 /* start = */ 0,
\n-
449 /* end = */ stencilTable->GetNumStencils(),
\n-
450 context);
\n-
451 }
\n-
452
\n-
503 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
504 bool EvalStencils(
\n-
505 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
506 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
507 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
508 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
509 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
510 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
511 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
512 STENCIL_TABLE const *stencilTable,
\n-
513 MTLContext* context) const
\n-
514 {
\n-
515 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,
\n-
516 dstBuffer->BindMTLBuffer(context), dstDesc,
\n-
517 duBuffer->BindMTLBuffer(context), duDesc,
\n-
518 dvBuffer->BindMTLBuffer(context), dvDesc,
\n-
519 duuBuffer->BindMTLBuffer(context), duuDesc,
\n-
520 duvBuffer->BindMTLBuffer(context), duvDesc,
\n-
521 dvvBuffer->BindMTLBuffer(context), dvvDesc,
\n-
522 stencilTable->GetSizesBuffer(),
\n-
523 stencilTable->GetOffsetsBuffer(),
\n-
524 stencilTable->GetIndicesBuffer(),
\n-
525 stencilTable->GetWeightsBuffer(),
\n-
526 stencilTable->GetDuWeightsBuffer(),
\n-
527 stencilTable->GetDvWeightsBuffer(),
\n-
528 stencilTable->GetDuuWeightsBuffer(),
\n-
529 stencilTable->GetDuvWeightsBuffer(),
\n-
530 stencilTable->GetDvvWeightsBuffer(),
\n-
531 /* start = */ 0,
\n-
532 /* end = */ stencilTable->GetNumStencils(),
\n-
533 context);
\n-
534 }
\n-
535
\n-
574 bool EvalStencils(id<MTLBuffer> srcBuffer, BufferDescriptor const &srcDesc,
\n-
575 id<MTLBuffer> dstBuffer, BufferDescriptor const &dstDesc,
\n-
576 id<MTLBuffer> duBuffer, BufferDescriptor const &duDesc,
\n-
577 id<MTLBuffer> dvBuffer, BufferDescriptor const &dvDesc,
\n-
578 id<MTLBuffer> sizesBuffer,
\n-
579 id<MTLBuffer> offsetsBuffer,
\n-
580 id<MTLBuffer> indicesBuffer,
\n-
581 id<MTLBuffer> weightsBuffer,
\n-
582 id<MTLBuffer> duWeightsBuffer,
\n-
583 id<MTLBuffer> dvWeightsBuffer,
\n-
584 int start,
\n-
585 int end,
\n-
586 MTLContext* context) const;
\n-
587
\n-
644 bool EvalStencils(id<MTLBuffer> srcBuffer, BufferDescriptor const &srcDesc,
\n-
645 id<MTLBuffer> dstBuffer, BufferDescriptor const &dstDesc,
\n-
646 id<MTLBuffer> duBuffer, BufferDescriptor const &duDesc,
\n-
647 id<MTLBuffer> dvBuffer, BufferDescriptor const &dvDesc,
\n-
648 id<MTLBuffer> duuBuffer, BufferDescriptor const &duuDesc,
\n-
649 id<MTLBuffer> duvBuffer, BufferDescriptor const &duvDesc,
\n-
650 id<MTLBuffer> dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
651 id<MTLBuffer> sizesBuffer,
\n-
652 id<MTLBuffer> offsetsBuffer,
\n-
653 id<MTLBuffer> indicesBuffer,
\n-
654 id<MTLBuffer> weightsBuffer,
\n-
655 id<MTLBuffer> duWeightsBuffer,
\n-
656 id<MTLBuffer> dvWeightsBuffer,
\n-
657 id<MTLBuffer> duuWeightsBuffer,
\n-
658 id<MTLBuffer> duvWeightsBuffer,
\n-
659 id<MTLBuffer> dvvWeightsBuffer,
\n-
660 int start,
\n-
661 int end,
\n-
662 MTLContext* context) const;
\n-
663
\n-
669
\n-
703 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
704 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
705 static bool EvalPatches(
\n-
706 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
707 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
708 int numPatchCoords,
\n-
709 PATCHCOORD_BUFFER *patchCoords,
\n-
710 PATCH_TABLE *patchTable,
\n-
711 MTLComputeEvaluator const *instance,
\n-
712 MTLContext* context) {
\n-
713
\n-
714 if (instance) {
\n-
715 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
716 dstBuffer, dstDesc,
\n-
717 numPatchCoords, patchCoords,
\n-
718 patchTable,
\n-
719 context);
\n-
720 } else {
\n-
721 // Create an instance on demand (slow)
\n-
722 instance = Create(srcDesc, dstDesc,
\n-
723 BufferDescriptor(),
\n-
724 BufferDescriptor(),
\n-
725 context);
\n-
726 if (instance) {
\n-
727 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
728 dstBuffer, dstDesc,
\n-
729 numPatchCoords, patchCoords,
\n-
730 patchTable,
\n-
731 context);
\n-
732 delete instance;
\n-
733 return r;
\n-
734 }
\n-
735 return false;
\n-
736 }
\n-
737 }
\n-
738
\n-
784 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
785 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
786 static bool EvalPatches(
\n-
787 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
788 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
789 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
790 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
791 int numPatchCoords,
\n-
792 PATCHCOORD_BUFFER *patchCoords,
\n-
793 PATCH_TABLE *patchTable,
\n-
794 MTLComputeEvaluator* instance,
\n-
795 MTLContext* context) {
\n-
796
\n-
797 if (instance) {
\n-
798 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
799 dstBuffer, dstDesc,
\n-
800 duBuffer, duDesc,
\n-
801 dvBuffer, dvDesc,
\n-
802 numPatchCoords, patchCoords,
\n-
803 patchTable,
\n-
804 context);
\n-
805 } else {
\n-
806 // Create an instance on demand (slow)
\n-
807 instance = Create(srcDesc, dstDesc,
\n-
808 duDesc, dvDesc, context);
\n-
809 if (instance) {
\n-
810 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
811 dstBuffer, dstDesc,
\n-
812 duBuffer, duDesc,
\n-
813 dvBuffer, dvDesc,
\n-
814 numPatchCoords, patchCoords,
\n-
815 patchTable,
\n-
816 context);
\n-
817 delete instance;
\n-
818 return r;
\n-
819 }
\n-
820 return false;
\n-
821 }
\n-
822 }
\n-
823
\n-
887 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
888 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
889 static bool EvalPatches(
\n-
890 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
891 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
892 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
893 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
894 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
895 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
896 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
897 int numPatchCoords,
\n-
898 PATCHCOORD_BUFFER *patchCoords,
\n-
899 PATCH_TABLE *patchTable,
\n-
900 MTLComputeEvaluator* instance,
\n-
901 MTLContext* context) {
\n-
902
\n-
903 if (instance) {
\n-
904 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
905 dstBuffer, dstDesc,
\n-
906 duBuffer, duDesc,
\n-
907 dvBuffer, dvDesc,
\n-
908 duuBuffer, duuDesc,
\n-
909 duvBuffer, duvDesc,
\n-
910 dvvBuffer, dvvDesc,
\n-
911 numPatchCoords, patchCoords,
\n-
912 patchTable,
\n-
913 context);
\n-
914 } else {
\n-
915 // Create an instance on demand (slow)
\n-
916 instance = Create(srcDesc, dstDesc,
\n-
917 duDesc, dvDesc,
\n-
918 duuDesc, duvDesc, dvvDesc, context);
\n-
919 if (instance) {
\n-
920 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
921 dstBuffer, dstDesc,
\n-
922 duBuffer, duDesc,
\n-
923 dvBuffer, dvDesc,
\n-
924 duuBuffer, duuDesc,
\n-
925 duvBuffer, duvDesc,
\n-
926 dvvBuffer, dvvDesc,
\n-
927 numPatchCoords, patchCoords,
\n-
928 patchTable,
\n-
929 context);
\n-
930 delete instance;
\n-
931 return r;
\n-
932 }
\n-
933 return false;
\n-
934 }
\n-
935 }
\n-
936
\n-
964 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
965 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
966 bool EvalPatches(
\n-
967 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
968 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
969 int numPatchCoords,
\n-
970 PATCHCOORD_BUFFER *patchCoords,
\n-
971 PATCH_TABLE *patchTable,
\n-
972 MTLContext* context) const {
\n-
973
\n-
974 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,
\n-
975 dstBuffer->BindMTLBuffer(context), dstDesc,
\n-
976 0, BufferDescriptor(),
\n-
977 0, BufferDescriptor(),
\n-
978 numPatchCoords,
\n-
979 patchCoords->BindMTLBuffer(context),
\n-
980 patchTable->GetPatchArrays(),
\n-
981 patchTable->GetPatchIndexBuffer(),
\n-
982 patchTable->GetPatchParamBuffer(),
\n-
983 context);
\n-
984 }
\n-
985
\n-
1023 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1024 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1025 bool EvalPatches(
\n-
1026 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1027 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1028 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1029 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1030 int numPatchCoords,
\n-
1031 PATCHCOORD_BUFFER *patchCoords,
\n-
1032 PATCH_TABLE *patchTable,
\n-
1033 MTLContext* context) const {
\n-
1034
\n-
1035 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,
\n-
1036 dstBuffer->BindMTLBuffer(context), dstDesc,
\n-
1037 duBuffer->BindMTLBuffer(context), duDesc,
\n-
1038 dvBuffer->BindMTLBuffer(context), dvDesc,
\n-
1039 numPatchCoords,
\n-
1040 patchCoords->BindMTLBuffer(context),
\n-
1041 patchTable->GetPatchArrays(),
\n-
1042 patchTable->GetPatchIndexBuffer(),
\n-
1043 patchTable->GetPatchParamBuffer(),
\n-
1044 context);
\n-
1045 }
\n-
1046
\n-
1102 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1103 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1104 bool EvalPatches(
\n-
1105 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1106 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1107 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1108 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1109 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1110 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1111 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1112 int numPatchCoords,
\n-
1113 PATCHCOORD_BUFFER *patchCoords,
\n-
1114 PATCH_TABLE *patchTable,
\n-
1115 MTLContext* context) const {
\n-
1116
\n-
1117 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,
\n-
1118 dstBuffer->BindMTLBuffer(context), dstDesc,
\n-
1119 duBuffer->BindMTLBuffer(context), duDesc,
\n-
1120 dvBuffer->BindMTLBuffer(context), dvDesc,
\n-
1121 duuBuffer->BindMTLBuffer(context), duuDesc,
\n-
1122 duvBuffer->BindMTLBuffer(context), duvDesc,
\n-
1123 dvvBuffer->BindMTLBuffer(context), dvvDesc,
\n-
1124 numPatchCoords,
\n-
1125 patchCoords->BindMTLBuffer(context),
\n-
1126 patchTable->GetPatchArrays(),
\n-
1127 patchTable->GetPatchIndexBuffer(),
\n-
1128 patchTable->GetPatchParamBuffer(),
\n-
1129 context);
\n-
1130 }
\n-
1131
\n-
1132 bool EvalPatches(id<MTLBuffer> srcBuffer, BufferDescriptor const &srcDesc,
\n-
1133 id<MTLBuffer> dstBuffer, BufferDescriptor const &dstDesc,
\n-
1134 id<MTLBuffer> duBuffer, BufferDescriptor const &duDesc,
\n-
1135 id<MTLBuffer> dvBuffer, BufferDescriptor const &dvDesc,
\n-
1136 int numPatchCoords,
\n-
1137 id<MTLBuffer> patchCoordsBuffer,
\n-
1138 const PatchArrayVector &patchArrays,
\n-
1139 id<MTLBuffer> patchIndexBuffer,
\n-
1140 id<MTLBuffer> patchParamsBuffer,
\n-
1141 MTLContext* context) const;
\n-
1142
\n-
1143 bool EvalPatches(id<MTLBuffer> srcBuffer, BufferDescriptor const &srcDesc,
\n-
1144 id<MTLBuffer> dstBuffer, BufferDescriptor const &dstDesc,
\n-
1145 id<MTLBuffer> duBuffer, BufferDescriptor const &duDesc,
\n-
1146 id<MTLBuffer> dvBuffer, BufferDescriptor const &dvDesc,
\n-
1147 id<MTLBuffer> duuBuffer, BufferDescriptor const &duuDesc,
\n-
1148 id<MTLBuffer> duvBuffer, BufferDescriptor const &duvDesc,
\n-
1149 id<MTLBuffer> dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1150 int numPatchCoords,
\n-
1151 id<MTLBuffer> patchCoordsBuffer,
\n-
1152 const PatchArrayVector &patchArrays,
\n-
1153 id<MTLBuffer> patchIndexBuffer,
\n-
1154 id<MTLBuffer> patchParamsBuffer,
\n-
1155 MTLContext* context) const;
\n-
1156
\n-
1190 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1191 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1192 static bool EvalPatchesVarying(
\n-
1193 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1194 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1195 int numPatchCoords,
\n-
1196 PATCHCOORD_BUFFER *patchCoords,
\n-
1197 PATCH_TABLE *patchTable,
\n-
1198 MTLComputeEvaluator const *instance,
\n-
1199 MTLContext* deviceContext) {
\n-
1200
\n-
1201 if (instance) {
\n-
1202 return instance->EvalPatchesVarying(
\n-
1203 srcBuffer, srcDesc,
\n-
1204 dstBuffer, dstDesc,
\n-
1205 numPatchCoords, patchCoords,
\n-
1206 patchTable,
\n-
1207 deviceContext);
\n-
1208 } else {
\n-
1209 // Create an instance on demand (slow)
\n-
1210 instance = Create(srcDesc, dstDesc,
\n-
1211 BufferDescriptor(),
\n-
1212 BufferDescriptor(),
\n-
1213 deviceContext);
\n-
1214 if (instance) {
\n-
1215 bool r = instance->EvalPatchesVarying(
\n-
1216 srcBuffer, srcDesc,
\n-
1217 dstBuffer, dstDesc,
\n-
1218 numPatchCoords, patchCoords,
\n-
1219 patchTable,
\n-
1220 deviceContext);
\n-
1221 delete instance;
\n-
1222 return r;
\n-
1223 }
\n-
1224 return false;
\n-
1225 }
\n-
1226 }
\n-
1227
\n-
1255 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1256 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1257 bool EvalPatchesVarying(
\n-
1258 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1259 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1260 int numPatchCoords,
\n-
1261 PATCHCOORD_BUFFER *patchCoords,
\n-
1262 PATCH_TABLE *patchTable,
\n-
1263 MTLContext* deviceContext) const {
\n-
1264
\n-
1265 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n-
1266 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n-
1267 0, BufferDescriptor(),
\n-
1268 0, BufferDescriptor(),
\n-
1269 numPatchCoords,
\n-
1270 patchCoords->BindMTLBuffer(deviceContext),
\n-
1271 patchTable->GetVaryingPatchArrays(),
\n-
1272 patchTable->GetVaryingPatchIndexBuffer(),
\n-
1273 patchTable->GetPatchParamBuffer(),
\n-
1274 deviceContext
\n-
1275 );
\n-
1276 }
\n-
1277
\n-
1323 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1324 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1325 static bool EvalPatchesVarying(
\n-
1326 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1327 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1328 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1329 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1330 int numPatchCoords,
\n-
1331 PATCHCOORD_BUFFER *patchCoords,
\n-
1332 PATCH_TABLE *patchTable,
\n-
1333 MTLComputeEvaluator const *instance,
\n-
1334 MTLContext* deviceContext) {
\n-
1335
\n-
1336 if (instance) {
\n-
1337 return instance->EvalPatchesVarying(
\n-
1338 srcBuffer, srcDesc,
\n-
1339 dstBuffer, dstDesc,
\n-
1340 duBuffer, duDesc,
\n-
1341 dvBuffer, dvDesc,
\n-
1342 numPatchCoords, patchCoords,
\n-
1343 patchTable,
\n-
1344 deviceContext);
\n-
1345 } else {
\n-
1346 // Create an instance on demand (slow)
\n-
1347 instance = Create(srcDesc, dstDesc,
\n-
1348 duDesc, dvDesc,
\n-
1349 deviceContext);
\n-
1350 if (instance) {
\n-
1351 bool r = instance->EvalPatchesVarying(
\n-
1352 srcBuffer, srcDesc,
\n-
1353 dstBuffer, dstDesc,
\n-
1354 duBuffer, duDesc,
\n-
1355 dvBuffer, dvDesc,
\n-
1356 numPatchCoords, patchCoords,
\n-
1357 patchTable,
\n-
1358 deviceContext);
\n-
1359 delete instance;
\n-
1360 return r;
\n-
1361 }
\n-
1362 return false;
\n-
1363 }
\n-
1364 }
\n-
1365
\n-
1405 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1406 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1407 bool EvalPatchesVarying(
\n-
1408 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1409 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1410 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1411 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1412 int numPatchCoords,
\n-
1413 PATCHCOORD_BUFFER *patchCoords,
\n-
1414 PATCH_TABLE *patchTable,
\n-
1415 MTLContext* deviceContext) const {
\n-
1416
\n-
1417 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n-
1418 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n-
1419 duBuffer->BindMTLBuffer(deviceContext), duDesc,
\n-
1420 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,
\n-
1421 numPatchCoords,
\n-
1422 patchCoords->BindMTLBuffer(deviceContext),
\n-
1423 patchTable->GetVaryingPatchArrays(),
\n-
1424 patchTable->GetVaryingPatchIndexBuffer(),
\n-
1425 patchTable->GetPatchParamBuffer(),
\n-
1426 deviceContext
\n-
1427 );
\n-
1428 }
\n-
1429
\n-
1430
\n-
1494 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1495 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1496 static bool EvalPatchesVarying(
\n-
1497 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1498 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1499 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1500 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1501 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1502 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1503 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1504 int numPatchCoords,
\n-
1505 PATCHCOORD_BUFFER *patchCoords,
\n-
1506 PATCH_TABLE *patchTable,
\n-
1507 MTLComputeEvaluator const *instance,
\n-
1508 MTLContext* deviceContext) {
\n-
1509
\n-
1510 if (instance) {
\n-
1511 return instance->EvalPatchesVarying(
\n-
1512 srcBuffer, srcDesc,
\n-
1513 dstBuffer, dstDesc,
\n-
1514 duBuffer, duDesc,
\n-
1515 dvBuffer, dvDesc,
\n-
1516 duuBuffer, duuDesc,
\n-
1517 duvBuffer, duvDesc,
\n-
1518 dvvBuffer, dvvDesc,
\n-
1519 numPatchCoords, patchCoords,
\n-
1520 patchTable,
\n-
1521 deviceContext);
\n-
1522 } else {
\n-
1523 // Create an instance on demand (slow)
\n-
1524 instance = Create(srcDesc, dstDesc,
\n-
1525 duDesc, dvDesc,
\n-
1526 duuDesc, duvDesc, dvvDesc,
\n-
1527 deviceContext);
\n-
1528 if (instance) {
\n-
1529 bool r = instance->EvalPatchesVarying(
\n-
1530 srcBuffer, srcDesc,
\n-
1531 dstBuffer, dstDesc,
\n-
1532 duBuffer, duDesc,
\n-
1533 dvBuffer, dvDesc,
\n-
1534 duuBuffer, duuDesc,
\n-
1535 duvBuffer, duvDesc,
\n-
1536 dvvBuffer, dvvDesc,
\n-
1537 numPatchCoords, patchCoords,
\n-
1538 patchTable,
\n-
1539 deviceContext);
\n-
1540 delete instance;
\n-
1541 return r;
\n-
1542 }
\n-
1543 return false;
\n-
1544 }
\n-
1545 }
\n-
1546
\n-
1604 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1605 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1606 bool EvalPatchesVarying(
\n-
1607 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1608 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1609 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1610 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1611 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1612 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1613 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1614 int numPatchCoords,
\n-
1615 PATCHCOORD_BUFFER *patchCoords,
\n-
1616 PATCH_TABLE *patchTable,
\n-
1617 MTLContext* deviceContext) const {
\n-
1618
\n-
1619 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n-
1620 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n-
1621 duBuffer->BindMTLBuffer(deviceContext), duDesc,
\n-
1622 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,
\n-
1623 duuBuffer->BindMTLBuffer(deviceContext), duuDesc,
\n-
1624 duvBuffer->BindMTLBuffer(deviceContext), duvDesc,
\n-
1625 dvvBuffer->BindMTLBuffer(deviceContext), dvvDesc,
\n-
1626 numPatchCoords,
\n-
1627 patchCoords->BindMTLBuffer(deviceContext),
\n-
1628 patchTable->GetVaryingPatchArrays(),
\n-
1629 patchTable->GetVaryingPatchIndexBuffer(),
\n-
1630 patchTable->GetPatchParamBuffer(),
\n-
1631 deviceContext
\n-
1632 );
\n-
1633 }
\n-
1634
\n-
1670 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1671 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1672 static bool EvalPatchesFaceVarying(
\n-
1673 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1674 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1675 int numPatchCoords,
\n-
1676 PATCHCOORD_BUFFER *patchCoords,
\n-
1677 PATCH_TABLE *patchTable,
\n-
1678 int fvarChannel,
\n-
1679 MTLComputeEvaluator const *instance,
\n-
1680 MTLContext* deviceContext) {
\n-
1681
\n-
1682 if (instance) {
\n-
1683 return instance->EvalPatchesFaceVarying(
\n-
1684 srcBuffer, srcDesc,
\n-
1685 dstBuffer, dstDesc,
\n-
1686 numPatchCoords, patchCoords,
\n-
1687 patchTable,
\n-
1688 fvarChannel,
\n-
1689 deviceContext);
\n-
1690 } else {
\n-
1691 // Create an instance on demand (slow)
\n-
1692 instance = Create(srcDesc, dstDesc,
\n-
1693 BufferDescriptor(),
\n-
1694 BufferDescriptor(),
\n-
1695 deviceContext);
\n-
1696 if (instance) {
\n-
1697 bool r = instance->EvalPatchesFaceVarying(
\n-
1698 srcBuffer, srcDesc,
\n-
1699 dstBuffer, dstDesc,
\n-
1700 numPatchCoords, patchCoords,
\n-
1701 patchTable,
\n-
1702 fvarChannel,
\n-
1703 deviceContext);
\n-
1704 delete instance;
\n-
1705 return r;
\n-
1706 }
\n-
1707 return false;
\n-
1708 }
\n-
1709 }
\n-
1710
\n-
1740 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1741 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1742 bool EvalPatchesFaceVarying(
\n-
1743 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1744 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1745 int numPatchCoords,
\n-
1746 PATCHCOORD_BUFFER *patchCoords,
\n-
1747 PATCH_TABLE *patchTable,
\n-
1748 int fvarChannel,
\n-
1749 MTLContext* deviceContext) const {
\n-
1750
\n-
1751 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n-
1752 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n-
1753 0, BufferDescriptor(),
\n-
1754 0, BufferDescriptor(),
\n-
1755 numPatchCoords,
\n-
1756 patchCoords->BindMTLBuffer(deviceContext),
\n-
1757 patchTable->GetFVarPatchArrays(fvarChannel),
\n-
1758 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1759 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n-
1760 deviceContext
\n-
1761 );
\n-
1762 }
\n-
1763
\n-
1811 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1812 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1813 static bool EvalPatchesFaceVarying(
\n-
1814 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1815 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1816 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1817 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1818 int numPatchCoords,
\n-
1819 PATCHCOORD_BUFFER *patchCoords,
\n-
1820 PATCH_TABLE *patchTable,
\n-
1821 int fvarChannel,
\n-
1822 MTLComputeEvaluator const *instance,
\n-
1823 MTLContext* deviceContext) {
\n-
1824
\n-
1825 if (instance) {
\n-
1826 return instance->EvalPatchesFaceVarying(
\n-
1827 srcBuffer, srcDesc,
\n-
1828 dstBuffer, dstDesc,
\n-
1829 duBuffer, duDesc,
\n-
1830 dvBuffer, dvDesc,
\n-
1831 numPatchCoords, patchCoords,
\n-
1832 patchTable,
\n-
1833 fvarChannel,
\n-
1834 deviceContext);
\n-
1835 } else {
\n-
1836 // Create an instance on demand (slow)
\n-
1837 instance = Create(srcDesc, dstDesc,
\n-
1838 duDesc, dvDesc,
\n-
1839 deviceContext);
\n-
1840 if (instance) {
\n-
1841 bool r = instance->EvalPatchesFaceVarying(
\n-
1842 srcBuffer, srcDesc,
\n-
1843 dstBuffer, dstDesc,
\n-
1844 duBuffer, duDesc,
\n-
1845 dvBuffer, dvDesc,
\n-
1846 numPatchCoords, patchCoords,
\n-
1847 patchTable,
\n-
1848 fvarChannel,
\n-
1849 deviceContext);
\n-
1850 delete instance;
\n-
1851 return r;
\n-
1852 }
\n-
1853 return false;
\n-
1854 }
\n-
1855 }
\n-
1856
\n-
1898 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1899 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1900 bool EvalPatchesFaceVarying(
\n-
1901 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1902 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1903 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1904 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1905 int numPatchCoords,
\n-
1906 PATCHCOORD_BUFFER *patchCoords,
\n-
1907 PATCH_TABLE *patchTable,
\n-
1908 int fvarChannel,
\n-
1909 MTLContext* deviceContext) const {
\n-
1910
\n-
1911 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n-
1912 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n-
1913 duBuffer->BindMTLBuffer(deviceContext), duDesc,
\n-
1914 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,
\n-
1915 numPatchCoords,
\n-
1916 patchCoords->BindMTLBuffer(deviceContext),
\n-
1917 patchTable->GetFVarPatchArrays(fvarChannel),
\n-
1918 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1919 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n-
1920 deviceContext
\n-
1921 );
\n-
1922 }
\n-
1923
\n-
1989 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1990 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1991 static bool EvalPatchesFaceVarying(
\n-
1992 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1993 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1994 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1995 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1996 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1997 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1998 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1999 int numPatchCoords,
\n-
2000 PATCHCOORD_BUFFER *patchCoords,
\n-
2001 PATCH_TABLE *patchTable,
\n-
2002 int fvarChannel,
\n-
2003 MTLComputeEvaluator const *instance,
\n-
2004 MTLContext* deviceContext) {
\n-
2005
\n-
2006 if (instance) {
\n-
2007 return instance->EvalPatchesFaceVarying(
\n-
2008 srcBuffer, srcDesc,
\n-
2009 dstBuffer, dstDesc,
\n-
2010 duBuffer, duDesc,
\n-
2011 dvBuffer, dvDesc,
\n-
2012 duuBuffer, duuDesc,
\n-
2013 duvBuffer, duvDesc,
\n-
2014 dvvBuffer, dvvDesc,
\n-
2015 numPatchCoords, patchCoords,
\n-
2016 patchTable,
\n-
2017 fvarChannel,
\n-
2018 deviceContext);
\n-
2019 } else {
\n-
2020 // Create an instance on demand (slow)
\n-
2021 instance = Create(srcDesc, dstDesc,
\n-
2022 duDesc, dvDesc,
\n-
2023 duuDesc, duvDesc, dvvDesc,
\n-
2024 deviceContext);
\n-
2025 if (instance) {
\n-
2026 bool r = instance->EvalPatchesFaceVarying(
\n-
2027 srcBuffer, srcDesc,
\n-
2028 dstBuffer, dstDesc,
\n-
2029 duBuffer, duDesc,
\n-
2030 dvBuffer, dvDesc,
\n-
2031 duuBuffer, duuDesc,
\n-
2032 duvBuffer, duvDesc,
\n-
2033 dvvBuffer, dvvDesc,
\n-
2034 numPatchCoords, patchCoords,
\n-
2035 patchTable,
\n-
2036 fvarChannel,
\n-
2037 deviceContext);
\n-
2038 delete instance;
\n-
2039 return r;
\n-
2040 }
\n-
2041 return false;
\n-
2042 }
\n-
2043 }
\n-
2044
\n-
2104 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
2105 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
2106 bool EvalPatchesFaceVarying(
\n-
2107 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
2108 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
2109 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
2110 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
2111 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
2112 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
2113 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
2114 int numPatchCoords,
\n-
2115 PATCHCOORD_BUFFER *patchCoords,
\n-
2116 PATCH_TABLE *patchTable,
\n-
2117 int fvarChannel,
\n-
2118 MTLContext* deviceContext) const {
\n-
2119
\n-
2120 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,
\n-
2121 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,
\n-
2122 duBuffer->BindMTLBuffer(deviceContext), duDesc,
\n-
2123 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,
\n-
2124 duuBuffer->BindMTLBuffer(deviceContext), duuDesc,
\n-
2125 duvBuffer->BindMTLBuffer(deviceContext), duvDesc,
\n-
2126 dvvBuffer->BindMTLBuffer(deviceContext), dvvDesc,
\n-
2127 numPatchCoords,
\n-
2128 patchCoords->BindMTLBuffer(deviceContext),
\n-
2129 patchTable->GetFVarPatchArrays(fvarChannel),
\n-
2130 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
2131 patchTable->GetFVarPatchParamBuffer(fvarChannel),
\n-
2132 fvarChannel,
\n-
2133 deviceContext
\n-
2134 );
\n-
2135 }
\n-
2136
\n-
2138 bool Compile(BufferDescriptor const &srcDesc,
\n-
2139 BufferDescriptor const &dstDesc,
\n-
2140 BufferDescriptor const &duDesc,
\n-
2141 BufferDescriptor const &dvDesc,
\n-
2142 BufferDescriptor const &duuDesc,
\n-
2143 BufferDescriptor const &duvDesc,
\n-
2144 BufferDescriptor const &dvvDesc,
\n-
2145 MTLContext* context);
\n-
2146
\n-
2148 static void Synchronize(MTLContext* context);
\n-
2149
\n-
2150 private:
\n-
2151
\n-
2152 id<MTLLibrary> _computeLibrary;
\n-
2153 id<MTLComputePipelineState> _evalStencils;
\n-
2154 id<MTLComputePipelineState> _evalPatches;
\n-
2155 id<MTLBuffer> _parameterBuffer;
\n-
2156
\n-
2157 int _workGroupSize;
\n-
2158};
\n-
2159
\n-
2160} //end namespace Osd
\n-
2161
\n-
2162} //end namespace OPENSUBDIV_VERSION
\n-
2163using namespace OPENSUBDIV_VERSION;
\n-
2164
\n-
2165} //end namespace OpenSubdiv
\n-
2166
\n-
2167#endif // OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H
\n+
65 CPUMTLVertexBuffer(int numElements, int numVertices);
\n+
66
\n+
67 bool allocate(MTLContext* context);
\n+
68
\n+
69private:
\n+
70 int _numElements;
\n+
71 int _numVertices;
\n+
72 id<MTLBuffer> _buffer;
\n+
73 bool _dirty;
\n+
74};
\n+
75
\n+
76} //end namespace Osd
\n+
77
\n+
78} //end namespace OPENSUBDIV_VERSION
\n+
79 using namespace OPENSUBDIV_VERSION;
\n+
80
\n+
81} //end namespace OpenSubdiv
\n+
82
\n+
83#endif // OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n \n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::MTLStencilTable
MTLStencilTable(Far::LimitStencilTable const *stencilTable, MTLContext *context)
\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::MTLStencilTable
MTLStencilTable(Far::StencilTable const *stencilTable, MTLContext *context)
\n-\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::Create
static MTLStencilTable * Create(STENCIL_TABLE *stencilTable, DEVICE_CONTEXT context)
Definition: mtlComputeEvaluator.h:54
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
bool EvalStencils(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc, id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer > duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer, BufferDescriptor const &dvDesc, id< MTLBuffer > duuBuffer, BufferDescriptor const &duuDesc, id< MTLBuffer > duvBuffer, BufferDescriptor const &duvDesc, id< MTLBuffer > dvvBuffer, BufferDescriptor const &dvvDesc, id< MTLBuffer > sizesBuffer, id< MTLBuffer > offsetsBuffer, id< MTLBuffer > indicesBuffer, id< MTLBuffer > weightsBuffer, id< MTLBuffer > duWeightsBuffer, id< MTLBuffer > dvWeightsBuffer, id< MTLBuffer > duuWeightsBuffer, id< MTLBuffer > duvWeightsBuffer, id< MTLBuffer > dvvWeightsBuffer, int start, int end, MTLContext *context) const
Dispatch the MTL compute kernel on GPU asynchronously returns false if the kernel hasn't been compile...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator *instance, MTLContext *context)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:786
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, MTLComputeEvaluator const *instance, MTLContext *context)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: mtlComputeEvaluator.h:312
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1991
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator const *instance, MTLContext *context)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:705
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1606
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, MTLComputeEvaluator const *instance, MTLContext *context)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: mtlComputeEvaluator.h:220
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1900
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1672
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1496
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, MTLContext *context) const
Generic stencil function.
Definition: mtlComputeEvaluator.h:376
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
bool EvalPatches(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc, id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer > duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer, BufferDescriptor const &dvDesc, id< MTLBuffer > duuBuffer, BufferDescriptor const &duuDesc, id< MTLBuffer > duvBuffer, BufferDescriptor const &duvDesc, id< MTLBuffer > dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, id< MTLBuffer > patchCoordsBuffer, const PatchArrayVector &patchArrays, id< MTLBuffer > patchIndexBuffer, id< MTLBuffer > patchParamsBuffer, MTLContext *context) const
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:2106
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Synchronize
static void Synchronize(MTLContext *context)
Wait for the dispatched kernel to finish.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, MTLContext *context) const
Generic stencil function.
Definition: mtlComputeEvaluator.h:430
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator *instance, MTLContext *context)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:889
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1742
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1192
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, MTLContext *context) const
Generic stencil function.
Definition: mtlComputeEvaluator.h:504
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
bool EvalStencils(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc, id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer > duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer, BufferDescriptor const &dvDesc, id< MTLBuffer > sizesBuffer, id< MTLBuffer > offsetsBuffer, id< MTLBuffer > indicesBuffer, id< MTLBuffer > weightsBuffer, id< MTLBuffer > duWeightsBuffer, id< MTLBuffer > dvWeightsBuffer, int start, int end, MTLContext *context) const
Dispatch the MTL compute kernel on GPU asynchronously returns false if the kernel hasn't been compile...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1813
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *context) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:966
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, MTLComputeEvaluator const *instance, MTLContext *context)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: mtlComputeEvaluator.h:149
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator const *instance, MTLContext *deviceContext)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1325
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *context) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: mtlComputeEvaluator.h:1104
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *context) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: mtlComputeEvaluator.h:1025
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1407
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *deviceContext) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: mtlComputeEvaluator.h:1257
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Create
static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, MTLContext *context)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Create
static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, MTLContext *context)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Compile
bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, MTLContext *context)
Configure compute pipline state. Returns false if it fails to create the pipeline state.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches
bool EvalPatches(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc, id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer > duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, id< MTLBuffer > patchCoordsBuffer, const PatchArrayVector &patchArrays, id< MTLBuffer > patchIndexBuffer, id< MTLBuffer > patchParamsBuffer, MTLContext *context) const
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::BindVBO
id< MTLBuffer > BindVBO(MTLContext *context)
Definition: mtlVertexBuffer.h:58
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::CPUMTLVertexBuffer
CPUMTLVertexBuffer(int numElements, int numVertices)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::Create
static CPUMTLVertexBuffer * Create(int numElements, int numVertices, MTLContext *context)
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::UpdateData
void UpdateData(const float *src, int startVertex, int numVertices, MTLContext *context)
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::BindMTLBuffer
id< MTLBuffer > BindMTLBuffer(MTLContext *context)
\n+\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-mtlComputeEvaluator.h\n+mtlVertexBuffer.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,1388 +30,101 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H\n- 26#define OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H\n+ 25#ifndef OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H\n+ 26#define OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H\n 27\n 28#include \"../version.h\"\n- 29\n- 30#include \"../osd/types.h\"\n- 31#include \"../osd/bufferDescriptor.h\"\n- 32#include \"../osd/mtlCommon.h\"\n+ 29#include \"../osd/mtlCommon.h\"\n+ 30\n+ 31@protocol MTLDevice;\n+ 32@protocol MTLBuffer;\n 33\n- 34@protocol MTLDevice;\n- 35@protocol MTLBuffer;\n- 36@protocol MTLLibrary;\n- 37@protocol MTLComputePipelineState;\n+ 34namespace OpenSubdiv {\n+ 35namespace OPENSUBDIV_VERSION {\n+ 36\n+ 37namespace Osd {\n 38\n- 39namespace OpenSubdiv {\n- 40namespace OPENSUBDIV_VERSION {\n- 41\n- 42namespace Far {\n- 43 class PatchTable;\n- 44 class StencilTable;\n- 45 class LimitStencilTable;\n- 46}\n- 47\n- 48namespace Osd {\n+39class CPUMTLVertexBuffer {\n+ 40public:\n+41 static CPUMTLVertexBuffer* Create(int numElements, int numVertices,\n+MTLContext* context);\n+ 42\n+43 void UpdateData(const float* src, int startVertex, int numVertices,\n+MTLContext* context);\n+ 44\n+45 int GetNumElements() const\n+ 46 {\n+ 47 return _numElements;\n+ 48 }\n 49\n-50class MTLStencilTable\n- 51{\n- 52public:\n- 53 template\n-54 static MTLStencilTable* Create(STENCIL_TABLE* stencilTable,\n- 55 DEVICE_CONTEXT context)\n- 56 {\n- 57 return new MTLStencilTable(stencilTable, context);\n- 58 }\n- 59\n- 60\n-61 MTLStencilTable(Far::StencilTable const* stencilTable, MTLContext* context);\n-62 MTLStencilTable(Far::LimitStencilTable const* stencilTable, MTLContext*\n-context);\n-63 ~MTLStencilTable();\n+50 int GetNumVertices() const\n+ 51 {\n+ 52 return _numVertices;\n+ 53 }\n+ 54\n+55 float* BindCpuBuffer();\n+56 id BindMTLBuffer(MTLContext* context);\n+ 57\n+58 id BindVBO(MTLContext* context)\n+ 59 {\n+ 60 return BindMTLBuffer(context);\n+ 61 }\n+ 62\n+ 63protected:\n 64\n-65 id GetSizesBuffer() const { return _sizesBuffer; }\n-66 id GetOffsetsBuffer() const { return _offsetsBuffer; }\n-67 id GetIndicesBuffer() const { return _indicesBuffer; }\n-68 id GetWeightsBuffer() const { return _weightsBuffer; }\n-69 id GetDuWeightsBuffer() const { return _duWeightsBuffer; }\n-70 id GetDvWeightsBuffer() const { return _dvWeightsBuffer; }\n-71 id GetDuuWeightsBuffer() const { return _duuWeightsBuffer; }\n-72 id GetDuvWeightsBuffer() const { return _duvWeightsBuffer; }\n-73 id GetDvvWeightsBuffer() const { return _dvvWeightsBuffer; }\n- 74\n-75 int GetNumStencils() const { return _numStencils; }\n- 76\n- 77private:\n- 78 id _sizesBuffer;\n- 79 id _offsetsBuffer;\n- 80 id _indicesBuffer;\n- 81 id _weightsBuffer;\n- 82 id _duWeightsBuffer;\n- 83 id _dvWeightsBuffer;\n- 84 id _duuWeightsBuffer;\n- 85 id _duvWeightsBuffer;\n- 86 id _dvvWeightsBuffer;\n- 87\n- 88 int _numStencils;\n- 89};\n- 90\n-91class MTLComputeEvaluator\n- 92{\n- 93public:\n-94 typedef bool Instantiatable;\n- 95\n-96 static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n- 97 BufferDescriptor const &dstDesc,\n- 98 BufferDescriptor const &duDesc,\n- 99 BufferDescriptor const &dvDesc,\n- 100 MTLContext* context);\n- 101\n-102 static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n- 103 BufferDescriptor const &dstDesc,\n- 104 BufferDescriptor const &duDesc,\n- 105 BufferDescriptor const &dvDesc,\n- 106 BufferDescriptor const &duuDesc,\n- 107 BufferDescriptor const &duvDesc,\n- 108 BufferDescriptor const &dvvDesc,\n- 109 MTLContext* context);\n- 110\n-111 MTLComputeEvaluator();\n-112 ~MTLComputeEvaluator();\n- 113\n- 119\n- 148 template \n-149 static bool EvalStencils(\n- 150 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 151 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 152 STENCIL_TABLE const *stencilTable,\n- 153 MTLComputeEvaluator const *instance,\n- 154 MTLContext* context)\n- 155 {\n- 156 if (instance) {\n- 157 return instance->EvalStencils(srcBuffer, srcDesc,\n- 158 dstBuffer, dstDesc,\n- 159 stencilTable,\n- 160 context);\n- 161 } else {\n- 162 // Create an instance on demand (slow)\n- 163 instance = Create(srcDesc, dstDesc,\n- 164 BufferDescriptor(),\n- 165 BufferDescriptor(),\n- 166 context);\n- 167 if (instance) {\n- 168 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 169 dstBuffer, dstDesc,\n- 170 stencilTable,\n- 171 context);\n- 172 delete instance;\n- 173 return r;\n- 174 }\n- 175 return false;\n- 176 }\n- 177 }\n- 178\n- 219 template \n-220 static bool EvalStencils(\n- 221 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 222 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 223 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 224 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 225 STENCIL_TABLE const *stencilTable,\n- 226 MTLComputeEvaluator const *instance,\n- 227 MTLContext* context) {\n- 228\n- 229 if (instance) {\n- 230 return instance->EvalStencils(srcBuffer, srcDesc,\n- 231 dstBuffer, dstDesc,\n- 232 duBuffer, duDesc,\n- 233 dvBuffer, dvDesc,\n- 234 stencilTable,\n- 235 context);\n- 236 } else {\n- 237 // Create an instance on demand (slow)\n- 238 instance = Create(srcDesc, dstDesc, duDesc, dvDesc, context);\n- 239 if (instance) {\n- 240 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 241 dstBuffer, dstDesc,\n- 242 duBuffer, duDesc,\n- 243 dvBuffer, dvDesc,\n- 244 stencilTable,\n- 245 context);\n- 246 delete instance;\n- 247 return r;\n- 248 }\n- 249 return false;\n- 250 }\n- 251 }\n- 252\n- 311 template \n-312 static bool EvalStencils(\n- 313 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 314 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 315 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 316 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 317 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 318 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 319 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 320 STENCIL_TABLE const *stencilTable,\n- 321 MTLComputeEvaluator const *instance,\n- 322 MTLContext* context) {\n- 323\n- 324 if (instance) {\n- 325 return instance->EvalStencils(srcBuffer, srcDesc,\n- 326 dstBuffer, dstDesc,\n- 327 duBuffer, duDesc,\n- 328 dvBuffer, dvDesc,\n- 329 duuBuffer, duuDesc,\n- 330 duvBuffer, duvDesc,\n- 331 dvvBuffer, dvvDesc,\n- 332 stencilTable,\n- 333 context);\n- 334 } else {\n- 335 // Create an instance on demand (slow)\n- 336 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,\n- 337 duuDesc, duvDesc, dvvDesc, context);\n- 338 if (instance) {\n- 339 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 340 dstBuffer, dstDesc,\n- 341 duBuffer, duDesc,\n- 342 dvBuffer, dvDesc,\n- 343 duuBuffer, duuDesc,\n- 344 duvBuffer, duvDesc,\n- 345 dvvBuffer, dvvDesc,\n- 346 stencilTable,\n- 347 context);\n- 348 delete instance;\n- 349 return r;\n- 350 }\n- 351 return false;\n- 352 }\n- 353 }\n- 354\n- 375 template \n-376 bool EvalStencils(\n- 377 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 378 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 379 STENCIL_TABLE const *stencilTable,\n- 380 MTLContext* context) const\n- 381 {\n- 382 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,\n- 383 dstBuffer->BindMTLBuffer(context), dstDesc,\n- 384 0, BufferDescriptor(),\n- 385 0, BufferDescriptor(),\n- 386 stencilTable->GetSizesBuffer(),\n- 387 stencilTable->GetOffsetsBuffer(),\n- 388 stencilTable->GetIndicesBuffer(),\n- 389 stencilTable->GetWeightsBuffer(),\n- 390 0,\n- 391 0,\n- 392 /* start = */ 0,\n- 393 /* end = */ stencilTable->GetNumStencils(),\n- 394 context);\n- 395 }\n- 396\n- 429 template \n-430 bool EvalStencils(\n- 431 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 432 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 433 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 434 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 435 STENCIL_TABLE const *stencilTable,\n- 436 MTLContext* context) const\n- 437 {\n- 438 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,\n- 439 dstBuffer->BindMTLBuffer(context), dstDesc,\n- 440 duBuffer->BindMTLBuffer(context), duDesc,\n- 441 dvBuffer->BindMTLBuffer(context), dvDesc,\n- 442 stencilTable->GetSizesBuffer(),\n- 443 stencilTable->GetOffsetsBuffer(),\n- 444 stencilTable->GetIndicesBuffer(),\n- 445 stencilTable->GetWeightsBuffer(),\n- 446 stencilTable->GetDuWeightsBuffer(),\n- 447 stencilTable->GetDvWeightsBuffer(),\n- 448 /* start = */ 0,\n- 449 /* end = */ stencilTable->GetNumStencils(),\n- 450 context);\n- 451 }\n- 452\n- 503 template \n-504 bool EvalStencils(\n- 505 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 506 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 507 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 508 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 509 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 510 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 511 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 512 STENCIL_TABLE const *stencilTable,\n- 513 MTLContext* context) const\n- 514 {\n- 515 return EvalStencils(srcBuffer->BindMTLBuffer(context), srcDesc,\n- 516 dstBuffer->BindMTLBuffer(context), dstDesc,\n- 517 duBuffer->BindMTLBuffer(context), duDesc,\n- 518 dvBuffer->BindMTLBuffer(context), dvDesc,\n- 519 duuBuffer->BindMTLBuffer(context), duuDesc,\n- 520 duvBuffer->BindMTLBuffer(context), duvDesc,\n- 521 dvvBuffer->BindMTLBuffer(context), dvvDesc,\n- 522 stencilTable->GetSizesBuffer(),\n- 523 stencilTable->GetOffsetsBuffer(),\n- 524 stencilTable->GetIndicesBuffer(),\n- 525 stencilTable->GetWeightsBuffer(),\n- 526 stencilTable->GetDuWeightsBuffer(),\n- 527 stencilTable->GetDvWeightsBuffer(),\n- 528 stencilTable->GetDuuWeightsBuffer(),\n- 529 stencilTable->GetDuvWeightsBuffer(),\n- 530 stencilTable->GetDvvWeightsBuffer(),\n- 531 /* start = */ 0,\n- 532 /* end = */ stencilTable->GetNumStencils(),\n- 533 context);\n- 534 }\n- 535\n-574 bool EvalStencils(id srcBuffer, BufferDescriptor const &srcDesc,\n- 575 id dstBuffer, BufferDescriptor const &dstDesc,\n- 576 id duBuffer, BufferDescriptor const &duDesc,\n- 577 id dvBuffer, BufferDescriptor const &dvDesc,\n- 578 id sizesBuffer,\n- 579 id offsetsBuffer,\n- 580 id indicesBuffer,\n- 581 id weightsBuffer,\n- 582 id duWeightsBuffer,\n- 583 id dvWeightsBuffer,\n- 584 int start,\n- 585 int end,\n- 586 MTLContext* context) const;\n- 587\n-644 bool EvalStencils(id srcBuffer, BufferDescriptor const &srcDesc,\n- 645 id dstBuffer, BufferDescriptor const &dstDesc,\n- 646 id duBuffer, BufferDescriptor const &duDesc,\n- 647 id dvBuffer, BufferDescriptor const &dvDesc,\n- 648 id duuBuffer, BufferDescriptor const &duuDesc,\n- 649 id duvBuffer, BufferDescriptor const &duvDesc,\n- 650 id dvvBuffer, BufferDescriptor const &dvvDesc,\n- 651 id sizesBuffer,\n- 652 id offsetsBuffer,\n- 653 id indicesBuffer,\n- 654 id weightsBuffer,\n- 655 id duWeightsBuffer,\n- 656 id dvWeightsBuffer,\n- 657 id duuWeightsBuffer,\n- 658 id duvWeightsBuffer,\n- 659 id dvvWeightsBuffer,\n- 660 int start,\n- 661 int end,\n- 662 MTLContext* context) const;\n- 663\n- 669\n- 703 template \n-705 static bool EvalPatches(\n- 706 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 707 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 708 int numPatchCoords,\n- 709 PATCHCOORD_BUFFER *patchCoords,\n- 710 PATCH_TABLE *patchTable,\n- 711 MTLComputeEvaluator const *instance,\n- 712 MTLContext* context) {\n- 713\n- 714 if (instance) {\n- 715 return instance->EvalPatches(srcBuffer, srcDesc,\n- 716 dstBuffer, dstDesc,\n- 717 numPatchCoords, patchCoords,\n- 718 patchTable,\n- 719 context);\n- 720 } else {\n- 721 // Create an instance on demand (slow)\n- 722 instance = Create(srcDesc, dstDesc,\n- 723 BufferDescriptor(),\n- 724 BufferDescriptor(),\n- 725 context);\n- 726 if (instance) {\n- 727 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 728 dstBuffer, dstDesc,\n- 729 numPatchCoords, patchCoords,\n- 730 patchTable,\n- 731 context);\n- 732 delete instance;\n- 733 return r;\n- 734 }\n- 735 return false;\n- 736 }\n- 737 }\n- 738\n- 784 template \n-786 static bool EvalPatches(\n- 787 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 788 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 789 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 790 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 791 int numPatchCoords,\n- 792 PATCHCOORD_BUFFER *patchCoords,\n- 793 PATCH_TABLE *patchTable,\n- 794 MTLComputeEvaluator* instance,\n- 795 MTLContext* context) {\n- 796\n- 797 if (instance) {\n- 798 return instance->EvalPatches(srcBuffer, srcDesc,\n- 799 dstBuffer, dstDesc,\n- 800 duBuffer, duDesc,\n- 801 dvBuffer, dvDesc,\n- 802 numPatchCoords, patchCoords,\n- 803 patchTable,\n- 804 context);\n- 805 } else {\n- 806 // Create an instance on demand (slow)\n- 807 instance = Create(srcDesc, dstDesc,\n- 808 duDesc, dvDesc, context);\n- 809 if (instance) {\n- 810 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 811 dstBuffer, dstDesc,\n- 812 duBuffer, duDesc,\n- 813 dvBuffer, dvDesc,\n- 814 numPatchCoords, patchCoords,\n- 815 patchTable,\n- 816 context);\n- 817 delete instance;\n- 818 return r;\n- 819 }\n- 820 return false;\n- 821 }\n- 822 }\n- 823\n- 887 template \n-889 static bool EvalPatches(\n- 890 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 891 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 892 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 893 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 894 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 895 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 896 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 897 int numPatchCoords,\n- 898 PATCHCOORD_BUFFER *patchCoords,\n- 899 PATCH_TABLE *patchTable,\n- 900 MTLComputeEvaluator* instance,\n- 901 MTLContext* context) {\n- 902\n- 903 if (instance) {\n- 904 return instance->EvalPatches(srcBuffer, srcDesc,\n- 905 dstBuffer, dstDesc,\n- 906 duBuffer, duDesc,\n- 907 dvBuffer, dvDesc,\n- 908 duuBuffer, duuDesc,\n- 909 duvBuffer, duvDesc,\n- 910 dvvBuffer, dvvDesc,\n- 911 numPatchCoords, patchCoords,\n- 912 patchTable,\n- 913 context);\n- 914 } else {\n- 915 // Create an instance on demand (slow)\n- 916 instance = Create(srcDesc, dstDesc,\n- 917 duDesc, dvDesc,\n- 918 duuDesc, duvDesc, dvvDesc, context);\n- 919 if (instance) {\n- 920 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 921 dstBuffer, dstDesc,\n- 922 duBuffer, duDesc,\n- 923 dvBuffer, dvDesc,\n- 924 duuBuffer, duuDesc,\n- 925 duvBuffer, duvDesc,\n- 926 dvvBuffer, dvvDesc,\n- 927 numPatchCoords, patchCoords,\n- 928 patchTable,\n- 929 context);\n- 930 delete instance;\n- 931 return r;\n- 932 }\n- 933 return false;\n- 934 }\n- 935 }\n- 936\n- 964 template \n-966 bool EvalPatches(\n- 967 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 968 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 969 int numPatchCoords,\n- 970 PATCHCOORD_BUFFER *patchCoords,\n- 971 PATCH_TABLE *patchTable,\n- 972 MTLContext* context) const {\n- 973\n- 974 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,\n- 975 dstBuffer->BindMTLBuffer(context), dstDesc,\n- 976 0, BufferDescriptor(),\n- 977 0, BufferDescriptor(),\n- 978 numPatchCoords,\n- 979 patchCoords->BindMTLBuffer(context),\n- 980 patchTable->GetPatchArrays(),\n- 981 patchTable->GetPatchIndexBuffer(),\n- 982 patchTable->GetPatchParamBuffer(),\n- 983 context);\n- 984 }\n- 985\n- 1023 template \n-1025 bool EvalPatches(\n- 1026 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1027 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1028 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1029 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1030 int numPatchCoords,\n- 1031 PATCHCOORD_BUFFER *patchCoords,\n- 1032 PATCH_TABLE *patchTable,\n- 1033 MTLContext* context) const {\n- 1034\n- 1035 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,\n- 1036 dstBuffer->BindMTLBuffer(context), dstDesc,\n- 1037 duBuffer->BindMTLBuffer(context), duDesc,\n- 1038 dvBuffer->BindMTLBuffer(context), dvDesc,\n- 1039 numPatchCoords,\n- 1040 patchCoords->BindMTLBuffer(context),\n- 1041 patchTable->GetPatchArrays(),\n- 1042 patchTable->GetPatchIndexBuffer(),\n- 1043 patchTable->GetPatchParamBuffer(),\n- 1044 context);\n- 1045 }\n- 1046\n- 1102 template \n-1104 bool EvalPatches(\n- 1105 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1106 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1107 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1108 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1109 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1110 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1111 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1112 int numPatchCoords,\n- 1113 PATCHCOORD_BUFFER *patchCoords,\n- 1114 PATCH_TABLE *patchTable,\n- 1115 MTLContext* context) const {\n- 1116\n- 1117 return EvalPatches(srcBuffer->BindMTLBuffer(context), srcDesc,\n- 1118 dstBuffer->BindMTLBuffer(context), dstDesc,\n- 1119 duBuffer->BindMTLBuffer(context), duDesc,\n- 1120 dvBuffer->BindMTLBuffer(context), dvDesc,\n- 1121 duuBuffer->BindMTLBuffer(context), duuDesc,\n- 1122 duvBuffer->BindMTLBuffer(context), duvDesc,\n- 1123 dvvBuffer->BindMTLBuffer(context), dvvDesc,\n- 1124 numPatchCoords,\n- 1125 patchCoords->BindMTLBuffer(context),\n- 1126 patchTable->GetPatchArrays(),\n- 1127 patchTable->GetPatchIndexBuffer(),\n- 1128 patchTable->GetPatchParamBuffer(),\n- 1129 context);\n- 1130 }\n- 1131\n-1132 bool EvalPatches(id srcBuffer, BufferDescriptor const &srcDesc,\n- 1133 id dstBuffer, BufferDescriptor const &dstDesc,\n- 1134 id duBuffer, BufferDescriptor const &duDesc,\n- 1135 id dvBuffer, BufferDescriptor const &dvDesc,\n- 1136 int numPatchCoords,\n- 1137 id patchCoordsBuffer,\n- 1138 const PatchArrayVector &patchArrays,\n- 1139 id patchIndexBuffer,\n- 1140 id patchParamsBuffer,\n- 1141 MTLContext* context) const;\n- 1142\n-1143 bool EvalPatches(id srcBuffer, BufferDescriptor const &srcDesc,\n- 1144 id dstBuffer, BufferDescriptor const &dstDesc,\n- 1145 id duBuffer, BufferDescriptor const &duDesc,\n- 1146 id dvBuffer, BufferDescriptor const &dvDesc,\n- 1147 id duuBuffer, BufferDescriptor const &duuDesc,\n- 1148 id duvBuffer, BufferDescriptor const &duvDesc,\n- 1149 id dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1150 int numPatchCoords,\n- 1151 id patchCoordsBuffer,\n- 1152 const PatchArrayVector &patchArrays,\n- 1153 id patchIndexBuffer,\n- 1154 id patchParamsBuffer,\n- 1155 MTLContext* context) const;\n- 1156\n- 1190 template \n-1192 static bool EvalPatchesVarying(\n- 1193 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1194 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1195 int numPatchCoords,\n- 1196 PATCHCOORD_BUFFER *patchCoords,\n- 1197 PATCH_TABLE *patchTable,\n- 1198 MTLComputeEvaluator const *instance,\n- 1199 MTLContext* deviceContext) {\n- 1200\n- 1201 if (instance) {\n- 1202 return instance->EvalPatchesVarying(\n- 1203 srcBuffer, srcDesc,\n- 1204 dstBuffer, dstDesc,\n- 1205 numPatchCoords, patchCoords,\n- 1206 patchTable,\n- 1207 deviceContext);\n- 1208 } else {\n- 1209 // Create an instance on demand (slow)\n- 1210 instance = Create(srcDesc, dstDesc,\n- 1211 BufferDescriptor(),\n- 1212 BufferDescriptor(),\n- 1213 deviceContext);\n- 1214 if (instance) {\n- 1215 bool r = instance->EvalPatchesVarying(\n- 1216 srcBuffer, srcDesc,\n- 1217 dstBuffer, dstDesc,\n- 1218 numPatchCoords, patchCoords,\n- 1219 patchTable,\n- 1220 deviceContext);\n- 1221 delete instance;\n- 1222 return r;\n- 1223 }\n- 1224 return false;\n- 1225 }\n- 1226 }\n- 1227\n- 1255 template \n-1257 bool EvalPatchesVarying(\n- 1258 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1259 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1260 int numPatchCoords,\n- 1261 PATCHCOORD_BUFFER *patchCoords,\n- 1262 PATCH_TABLE *patchTable,\n- 1263 MTLContext* deviceContext) const {\n- 1264\n- 1265 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n- 1266 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n- 1267 0, BufferDescriptor(),\n- 1268 0, BufferDescriptor(),\n- 1269 numPatchCoords,\n- 1270 patchCoords->BindMTLBuffer(deviceContext),\n- 1271 patchTable->GetVaryingPatchArrays(),\n- 1272 patchTable->GetVaryingPatchIndexBuffer(),\n- 1273 patchTable->GetPatchParamBuffer(),\n- 1274 deviceContext\n- 1275 );\n- 1276 }\n- 1277\n- 1323 template \n-1325 static bool EvalPatchesVarying(\n- 1326 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1327 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1328 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1329 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1330 int numPatchCoords,\n- 1331 PATCHCOORD_BUFFER *patchCoords,\n- 1332 PATCH_TABLE *patchTable,\n- 1333 MTLComputeEvaluator const *instance,\n- 1334 MTLContext* deviceContext) {\n- 1335\n- 1336 if (instance) {\n- 1337 return instance->EvalPatchesVarying(\n- 1338 srcBuffer, srcDesc,\n- 1339 dstBuffer, dstDesc,\n- 1340 duBuffer, duDesc,\n- 1341 dvBuffer, dvDesc,\n- 1342 numPatchCoords, patchCoords,\n- 1343 patchTable,\n- 1344 deviceContext);\n- 1345 } else {\n- 1346 // Create an instance on demand (slow)\n- 1347 instance = Create(srcDesc, dstDesc,\n- 1348 duDesc, dvDesc,\n- 1349 deviceContext);\n- 1350 if (instance) {\n- 1351 bool r = instance->EvalPatchesVarying(\n- 1352 srcBuffer, srcDesc,\n- 1353 dstBuffer, dstDesc,\n- 1354 duBuffer, duDesc,\n- 1355 dvBuffer, dvDesc,\n- 1356 numPatchCoords, patchCoords,\n- 1357 patchTable,\n- 1358 deviceContext);\n- 1359 delete instance;\n- 1360 return r;\n- 1361 }\n- 1362 return false;\n- 1363 }\n- 1364 }\n- 1365\n- 1405 template \n-1407 bool EvalPatchesVarying(\n- 1408 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1409 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1410 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1411 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1412 int numPatchCoords,\n- 1413 PATCHCOORD_BUFFER *patchCoords,\n- 1414 PATCH_TABLE *patchTable,\n- 1415 MTLContext* deviceContext) const {\n- 1416\n- 1417 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n- 1418 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n- 1419 duBuffer->BindMTLBuffer(deviceContext), duDesc,\n- 1420 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,\n- 1421 numPatchCoords,\n- 1422 patchCoords->BindMTLBuffer(deviceContext),\n- 1423 patchTable->GetVaryingPatchArrays(),\n- 1424 patchTable->GetVaryingPatchIndexBuffer(),\n- 1425 patchTable->GetPatchParamBuffer(),\n- 1426 deviceContext\n- 1427 );\n- 1428 }\n- 1429\n- 1430\n- 1494 template \n-1496 static bool EvalPatchesVarying(\n- 1497 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1498 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1499 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1500 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1501 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1502 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1503 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1504 int numPatchCoords,\n- 1505 PATCHCOORD_BUFFER *patchCoords,\n- 1506 PATCH_TABLE *patchTable,\n- 1507 MTLComputeEvaluator const *instance,\n- 1508 MTLContext* deviceContext) {\n- 1509\n- 1510 if (instance) {\n- 1511 return instance->EvalPatchesVarying(\n- 1512 srcBuffer, srcDesc,\n- 1513 dstBuffer, dstDesc,\n- 1514 duBuffer, duDesc,\n- 1515 dvBuffer, dvDesc,\n- 1516 duuBuffer, duuDesc,\n- 1517 duvBuffer, duvDesc,\n- 1518 dvvBuffer, dvvDesc,\n- 1519 numPatchCoords, patchCoords,\n- 1520 patchTable,\n- 1521 deviceContext);\n- 1522 } else {\n- 1523 // Create an instance on demand (slow)\n- 1524 instance = Create(srcDesc, dstDesc,\n- 1525 duDesc, dvDesc,\n- 1526 duuDesc, duvDesc, dvvDesc,\n- 1527 deviceContext);\n- 1528 if (instance) {\n- 1529 bool r = instance->EvalPatchesVarying(\n- 1530 srcBuffer, srcDesc,\n- 1531 dstBuffer, dstDesc,\n- 1532 duBuffer, duDesc,\n- 1533 dvBuffer, dvDesc,\n- 1534 duuBuffer, duuDesc,\n- 1535 duvBuffer, duvDesc,\n- 1536 dvvBuffer, dvvDesc,\n- 1537 numPatchCoords, patchCoords,\n- 1538 patchTable,\n- 1539 deviceContext);\n- 1540 delete instance;\n- 1541 return r;\n- 1542 }\n- 1543 return false;\n- 1544 }\n- 1545 }\n- 1546\n- 1604 template \n-1606 bool EvalPatchesVarying(\n- 1607 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1608 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1609 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1610 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1611 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1612 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1613 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1614 int numPatchCoords,\n- 1615 PATCHCOORD_BUFFER *patchCoords,\n- 1616 PATCH_TABLE *patchTable,\n- 1617 MTLContext* deviceContext) const {\n- 1618\n- 1619 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n- 1620 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n- 1621 duBuffer->BindMTLBuffer(deviceContext), duDesc,\n- 1622 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,\n- 1623 duuBuffer->BindMTLBuffer(deviceContext), duuDesc,\n- 1624 duvBuffer->BindMTLBuffer(deviceContext), duvDesc,\n- 1625 dvvBuffer->BindMTLBuffer(deviceContext), dvvDesc,\n- 1626 numPatchCoords,\n- 1627 patchCoords->BindMTLBuffer(deviceContext),\n- 1628 patchTable->GetVaryingPatchArrays(),\n- 1629 patchTable->GetVaryingPatchIndexBuffer(),\n- 1630 patchTable->GetPatchParamBuffer(),\n- 1631 deviceContext\n- 1632 );\n- 1633 }\n- 1634\n- 1670 template \n-1672 static bool EvalPatchesFaceVarying(\n- 1673 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1674 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1675 int numPatchCoords,\n- 1676 PATCHCOORD_BUFFER *patchCoords,\n- 1677 PATCH_TABLE *patchTable,\n- 1678 int fvarChannel,\n- 1679 MTLComputeEvaluator const *instance,\n- 1680 MTLContext* deviceContext) {\n- 1681\n- 1682 if (instance) {\n- 1683 return instance->EvalPatchesFaceVarying(\n- 1684 srcBuffer, srcDesc,\n- 1685 dstBuffer, dstDesc,\n- 1686 numPatchCoords, patchCoords,\n- 1687 patchTable,\n- 1688 fvarChannel,\n- 1689 deviceContext);\n- 1690 } else {\n- 1691 // Create an instance on demand (slow)\n- 1692 instance = Create(srcDesc, dstDesc,\n- 1693 BufferDescriptor(),\n- 1694 BufferDescriptor(),\n- 1695 deviceContext);\n- 1696 if (instance) {\n- 1697 bool r = instance->EvalPatchesFaceVarying(\n- 1698 srcBuffer, srcDesc,\n- 1699 dstBuffer, dstDesc,\n- 1700 numPatchCoords, patchCoords,\n- 1701 patchTable,\n- 1702 fvarChannel,\n- 1703 deviceContext);\n- 1704 delete instance;\n- 1705 return r;\n- 1706 }\n- 1707 return false;\n- 1708 }\n- 1709 }\n- 1710\n- 1740 template \n-1742 bool EvalPatchesFaceVarying(\n- 1743 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1744 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1745 int numPatchCoords,\n- 1746 PATCHCOORD_BUFFER *patchCoords,\n- 1747 PATCH_TABLE *patchTable,\n- 1748 int fvarChannel,\n- 1749 MTLContext* deviceContext) const {\n- 1750\n- 1751 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n- 1752 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n- 1753 0, BufferDescriptor(),\n- 1754 0, BufferDescriptor(),\n- 1755 numPatchCoords,\n- 1756 patchCoords->BindMTLBuffer(deviceContext),\n- 1757 patchTable->GetFVarPatchArrays(fvarChannel),\n- 1758 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1759 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n- 1760 deviceContext\n- 1761 );\n- 1762 }\n- 1763\n- 1811 template \n-1813 static bool EvalPatchesFaceVarying(\n- 1814 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1815 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1816 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1817 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1818 int numPatchCoords,\n- 1819 PATCHCOORD_BUFFER *patchCoords,\n- 1820 PATCH_TABLE *patchTable,\n- 1821 int fvarChannel,\n- 1822 MTLComputeEvaluator const *instance,\n- 1823 MTLContext* deviceContext) {\n- 1824\n- 1825 if (instance) {\n- 1826 return instance->EvalPatchesFaceVarying(\n- 1827 srcBuffer, srcDesc,\n- 1828 dstBuffer, dstDesc,\n- 1829 duBuffer, duDesc,\n- 1830 dvBuffer, dvDesc,\n- 1831 numPatchCoords, patchCoords,\n- 1832 patchTable,\n- 1833 fvarChannel,\n- 1834 deviceContext);\n- 1835 } else {\n- 1836 // Create an instance on demand (slow)\n- 1837 instance = Create(srcDesc, dstDesc,\n- 1838 duDesc, dvDesc,\n- 1839 deviceContext);\n- 1840 if (instance) {\n- 1841 bool r = instance->EvalPatchesFaceVarying(\n- 1842 srcBuffer, srcDesc,\n- 1843 dstBuffer, dstDesc,\n- 1844 duBuffer, duDesc,\n- 1845 dvBuffer, dvDesc,\n- 1846 numPatchCoords, patchCoords,\n- 1847 patchTable,\n- 1848 fvarChannel,\n- 1849 deviceContext);\n- 1850 delete instance;\n- 1851 return r;\n- 1852 }\n- 1853 return false;\n- 1854 }\n- 1855 }\n- 1856\n- 1898 template \n-1900 bool EvalPatchesFaceVarying(\n- 1901 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1902 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1903 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1904 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1905 int numPatchCoords,\n- 1906 PATCHCOORD_BUFFER *patchCoords,\n- 1907 PATCH_TABLE *patchTable,\n- 1908 int fvarChannel,\n- 1909 MTLContext* deviceContext) const {\n- 1910\n- 1911 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n- 1912 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n- 1913 duBuffer->BindMTLBuffer(deviceContext), duDesc,\n- 1914 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,\n- 1915 numPatchCoords,\n- 1916 patchCoords->BindMTLBuffer(deviceContext),\n- 1917 patchTable->GetFVarPatchArrays(fvarChannel),\n- 1918 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1919 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n- 1920 deviceContext\n- 1921 );\n- 1922 }\n- 1923\n- 1989 template \n-1991 static bool EvalPatchesFaceVarying(\n- 1992 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1993 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1994 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1995 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1996 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1997 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1998 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1999 int numPatchCoords,\n- 2000 PATCHCOORD_BUFFER *patchCoords,\n- 2001 PATCH_TABLE *patchTable,\n- 2002 int fvarChannel,\n- 2003 MTLComputeEvaluator const *instance,\n- 2004 MTLContext* deviceContext) {\n- 2005\n- 2006 if (instance) {\n- 2007 return instance->EvalPatchesFaceVarying(\n- 2008 srcBuffer, srcDesc,\n- 2009 dstBuffer, dstDesc,\n- 2010 duBuffer, duDesc,\n- 2011 dvBuffer, dvDesc,\n- 2012 duuBuffer, duuDesc,\n- 2013 duvBuffer, duvDesc,\n- 2014 dvvBuffer, dvvDesc,\n- 2015 numPatchCoords, patchCoords,\n- 2016 patchTable,\n- 2017 fvarChannel,\n- 2018 deviceContext);\n- 2019 } else {\n- 2020 // Create an instance on demand (slow)\n- 2021 instance = Create(srcDesc, dstDesc,\n- 2022 duDesc, dvDesc,\n- 2023 duuDesc, duvDesc, dvvDesc,\n- 2024 deviceContext);\n- 2025 if (instance) {\n- 2026 bool r = instance->EvalPatchesFaceVarying(\n- 2027 srcBuffer, srcDesc,\n- 2028 dstBuffer, dstDesc,\n- 2029 duBuffer, duDesc,\n- 2030 dvBuffer, dvDesc,\n- 2031 duuBuffer, duuDesc,\n- 2032 duvBuffer, duvDesc,\n- 2033 dvvBuffer, dvvDesc,\n- 2034 numPatchCoords, patchCoords,\n- 2035 patchTable,\n- 2036 fvarChannel,\n- 2037 deviceContext);\n- 2038 delete instance;\n- 2039 return r;\n- 2040 }\n- 2041 return false;\n- 2042 }\n- 2043 }\n- 2044\n- 2104 template \n-2106 bool EvalPatchesFaceVarying(\n- 2107 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 2108 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 2109 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 2110 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 2111 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 2112 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 2113 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 2114 int numPatchCoords,\n- 2115 PATCHCOORD_BUFFER *patchCoords,\n- 2116 PATCH_TABLE *patchTable,\n- 2117 int fvarChannel,\n- 2118 MTLContext* deviceContext) const {\n- 2119\n- 2120 return EvalPatches(srcBuffer->BindMTLBuffer(deviceContext), srcDesc,\n- 2121 dstBuffer->BindMTLBuffer(deviceContext), dstDesc,\n- 2122 duBuffer->BindMTLBuffer(deviceContext), duDesc,\n- 2123 dvBuffer->BindMTLBuffer(deviceContext), dvDesc,\n- 2124 duuBuffer->BindMTLBuffer(deviceContext), duuDesc,\n- 2125 duvBuffer->BindMTLBuffer(deviceContext), duvDesc,\n- 2126 dvvBuffer->BindMTLBuffer(deviceContext), dvvDesc,\n- 2127 numPatchCoords,\n- 2128 patchCoords->BindMTLBuffer(deviceContext),\n- 2129 patchTable->GetFVarPatchArrays(fvarChannel),\n- 2130 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 2131 patchTable->GetFVarPatchParamBuffer(fvarChannel),\n- 2132 fvarChannel,\n- 2133 deviceContext\n- 2134 );\n- 2135 }\n- 2136\n-2138 bool Compile(BufferDescriptor const &srcDesc,\n- 2139 BufferDescriptor const &dstDesc,\n- 2140 BufferDescriptor const &duDesc,\n- 2141 BufferDescriptor const &dvDesc,\n- 2142 BufferDescriptor const &duuDesc,\n- 2143 BufferDescriptor const &duvDesc,\n- 2144 BufferDescriptor const &dvvDesc,\n- 2145 MTLContext* context);\n- 2146\n-2148 static void Synchronize(MTLContext* context);\n- 2149\n- 2150 private:\n- 2151\n- 2152 id _computeLibrary;\n- 2153 id _evalStencils;\n- 2154 id _evalPatches;\n- 2155 id _parameterBuffer;\n- 2156\n- 2157 int _workGroupSize;\n- 2158};\n- 2159\n- 2160} //end namespace Osd\n- 2161\n- 2162} //end namespace OPENSUBDIV_VERSION\n- 2163using namespace OPENSUBDIV_VERSION;\n- 2164\n- 2165} //end namespace OpenSubdiv\n- 2166\n- 2167#endif // OPENSUBDIV3_OSD_MTL_COMPUTE_EVALUATOR_H\n+65 CPUMTLVertexBuffer(int numElements, int numVertices);\n+ 66\n+67 bool allocate(MTLContext* context);\n+ 68\n+ 69private:\n+ 70 int _numElements;\n+ 71 int _numVertices;\n+ 72 id _buffer;\n+ 73 bool _dirty;\n+ 74};\n+ 75\n+ 76} //end namespace Osd\n+ 77\n+ 78} //end namespace OPENSUBDIV_VERSION\n+ 79 using namespace OPENSUBDIV_VERSION;\n+ 80\n+ 81} //end namespace OpenSubdiv\n+ 82\n+ 83#endif // OPENSUBDIV3_OSD_MTL_VERTEX_BUFFER_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n-std::vector< PatchArray > PatchArrayVector\n-Definition: types.h:132\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n-Stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:273\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n-Limit stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:583\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n-BufferDescriptor is a struct which describes buffer elements in interleaved\n-data buffers....\n-Definition: bufferDescriptor.h:61\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLContext\n Definition: mtlCommon.h:40\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable\n-Definition: mtlComputeEvaluator.h:51\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::MTLStencilTable\n-MTLStencilTable(Far::LimitStencilTable const *stencilTable, MTLContext\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer\n+Definition: mtlVertexBuffer.h:39\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::BindVBO\n+id< MTLBuffer > BindVBO(MTLContext *context)\n+Definition: mtlVertexBuffer.h:58\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::CPUMTLVertexBuffer\n+CPUMTLVertexBuffer(int numElements, int numVertices)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::Create\n+static CPUMTLVertexBuffer * Create(int numElements, int numVertices, MTLContext\n *context)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetDuWeightsBuffer\n-id< MTLBuffer > GetDuWeightsBuffer() const\n-Definition: mtlComputeEvaluator.h:69\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::~MTLStencilTable\n-~MTLStencilTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetDvvWeightsBuffer\n-id< MTLBuffer > GetDvvWeightsBuffer() const\n-Definition: mtlComputeEvaluator.h:73\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::MTLStencilTable\n-MTLStencilTable(Far::StencilTable const *stencilTable, MTLContext *context)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetDuuWeightsBuffer\n-id< MTLBuffer > GetDuuWeightsBuffer() const\n-Definition: mtlComputeEvaluator.h:71\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetDuvWeightsBuffer\n-id< MTLBuffer > GetDuvWeightsBuffer() const\n-Definition: mtlComputeEvaluator.h:72\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetNumStencils\n-int GetNumStencils() const\n-Definition: mtlComputeEvaluator.h:75\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetIndicesBuffer\n-id< MTLBuffer > GetIndicesBuffer() const\n-Definition: mtlComputeEvaluator.h:67\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetSizesBuffer\n-id< MTLBuffer > GetSizesBuffer() const\n-Definition: mtlComputeEvaluator.h:65\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetOffsetsBuffer\n-id< MTLBuffer > GetOffsetsBuffer() const\n-Definition: mtlComputeEvaluator.h:66\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetWeightsBuffer\n-id< MTLBuffer > GetWeightsBuffer() const\n-Definition: mtlComputeEvaluator.h:68\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::Create\n-static MTLStencilTable * Create(STENCIL_TABLE *stencilTable, DEVICE_CONTEXT\n-context)\n-Definition: mtlComputeEvaluator.h:54\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetDvWeightsBuffer\n-id< MTLBuffer > GetDvWeightsBuffer() const\n-Definition: mtlComputeEvaluator.h:70\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator\n-Definition: mtlComputeEvaluator.h:92\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n-bool EvalStencils(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc,\n-id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer >\n-duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer,\n-BufferDescriptor const &dvDesc, id< MTLBuffer > duuBuffer, BufferDescriptor\n-const &duuDesc, id< MTLBuffer > duvBuffer, BufferDescriptor const &duvDesc, id<\n-MTLBuffer > dvvBuffer, BufferDescriptor const &dvvDesc, id< MTLBuffer >\n-sizesBuffer, id< MTLBuffer > offsetsBuffer, id< MTLBuffer > indicesBuffer, id<\n-MTLBuffer > weightsBuffer, id< MTLBuffer > duWeightsBuffer, id< MTLBuffer >\n-dvWeightsBuffer, id< MTLBuffer > duuWeightsBuffer, id< MTLBuffer >\n-duvWeightsBuffer, id< MTLBuffer > dvvWeightsBuffer, int start, int end,\n-MTLContext *context) const\n-Dispatch the MTL compute kernel on GPU asynchronously returns false if the\n-kernel hasn't been compile...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable, MTLComputeEvaluator *instance, MTLContext *context)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:786\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n-MTLComputeEvaluator const *instance, MTLContext *context)\n-Generic static stencil function. This function has a same signature as other\n-device kernels have so t...\n-Definition: mtlComputeEvaluator.h:312\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n-EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n-MTLComputeEvaluator const *instance, MTLContext *deviceContext)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:1991\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::MTLComputeEvaluator\n-MTLComputeEvaluator()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::~MTLComputeEvaluator\n-~MTLComputeEvaluator()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator\n-const *instance, MTLContext *context)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:705\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Instantiatable\n-bool Instantiatable\n-Definition: mtlComputeEvaluator.h:94\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n-bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, MTLContext *deviceContext) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:1606\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable,\n-MTLComputeEvaluator const *instance, MTLContext *context)\n-Generic static stencil function. This function has a same signature as other\n-device kernels have so t...\n-Definition: mtlComputeEvaluator.h:220\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n-EvalPatchesFaceVarying\n-bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLContext\n-*deviceContext) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:1900\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n-EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n-fvarChannel, MTLComputeEvaluator const *instance, MTLContext *deviceContext)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:1672\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator\n-const *instance, MTLContext *deviceContext)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:1496\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n-bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const\n-*stencilTable, MTLContext *context) const\n-Generic stencil function.\n-Definition: mtlComputeEvaluator.h:376\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n-bool EvalPatches(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc,\n-id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer >\n-duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer,\n-BufferDescriptor const &dvDesc, id< MTLBuffer > duuBuffer, BufferDescriptor\n-const &duuDesc, id< MTLBuffer > duvBuffer, BufferDescriptor const &duvDesc, id<\n-MTLBuffer > dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, id<\n-MTLBuffer > patchCoordsBuffer, const PatchArrayVector &patchArrays, id<\n-MTLBuffer > patchIndexBuffer, id< MTLBuffer > patchParamsBuffer, MTLContext\n-*context) const\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n-EvalPatchesFaceVarying\n-bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n-MTLContext *deviceContext) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:2106\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Synchronize\n-static void Synchronize(MTLContext *context)\n-Wait for the dispatched kernel to finish.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n-bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, STENCIL_TABLE const *stencilTable, MTLContext *context) const\n-Generic stencil function.\n-Definition: mtlComputeEvaluator.h:430\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator *instance,\n-MTLContext *context)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:889\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n-EvalPatchesFaceVarying\n-bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n-fvarChannel, MTLContext *deviceContext) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:1742\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n-MTLComputeEvaluator const *instance, MTLContext *deviceContext)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:1192\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n-bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, MTLContext\n-*context) const\n-Generic stencil function.\n-Definition: mtlComputeEvaluator.h:504\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n-bool EvalStencils(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc,\n-id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer >\n-duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer,\n-BufferDescriptor const &dvDesc, id< MTLBuffer > sizesBuffer, id< MTLBuffer >\n-offsetsBuffer, id< MTLBuffer > indicesBuffer, id< MTLBuffer > weightsBuffer,\n-id< MTLBuffer > duWeightsBuffer, id< MTLBuffer > dvWeightsBuffer, int start,\n-int end, MTLContext *context) const\n-Dispatch the MTL compute kernel on GPU asynchronously returns false if the\n-kernel hasn't been compile...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::\n-EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, MTLComputeEvaluator\n-const *instance, MTLContext *deviceContext)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:1813\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n-bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext *context)\n-const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:966\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n-const *stencilTable, MTLComputeEvaluator const *instance, MTLContext *context)\n-Generic static stencil function. This function has a same signature as other\n-device kernels have so t...\n-Definition: mtlComputeEvaluator.h:149\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, MTLComputeEvaluator const *instance,\n-MTLContext *deviceContext)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:1325\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n-bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, MTLContext *context) const\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: mtlComputeEvaluator.h:1104\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n-bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable, MTLContext *context) const\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: mtlComputeEvaluator.h:1025\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n-bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable, MTLContext *deviceContext) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:1407\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatchesVarying\n-bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, MTLContext\n-*deviceContext) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: mtlComputeEvaluator.h:1257\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Create\n-static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, MTLContext *context)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Create\n-static MTLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n-BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, MTLContext\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::GetNumElements\n+int GetNumElements() const\n+Definition: mtlVertexBuffer.h:45\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::UpdateData\n+void UpdateData(const float *src, int startVertex, int numVertices, MTLContext\n *context)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::Compile\n-bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc,\n-BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc,\n-BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc,\n-BufferDescriptor const &dvvDesc, MTLContext *context)\n-Configure compute pipline state. Returns false if it fails to create the\n-pipeline state.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::EvalPatches\n-bool EvalPatches(id< MTLBuffer > srcBuffer, BufferDescriptor const &srcDesc,\n-id< MTLBuffer > dstBuffer, BufferDescriptor const &dstDesc, id< MTLBuffer >\n-duBuffer, BufferDescriptor const &duDesc, id< MTLBuffer > dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, id< MTLBuffer >\n-patchCoordsBuffer, const PatchArrayVector &patchArrays, id< MTLBuffer >\n-patchIndexBuffer, id< MTLBuffer > patchParamsBuffer, MTLContext *context) const\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::GetNumVertices\n+int GetNumVertices() const\n+Definition: mtlVertexBuffer.h:50\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::BindMTLBuffer\n+id< MTLBuffer > BindMTLBuffer(MTLContext *context)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::BindCpuBuffer\n+float * BindCpuBuffer()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::allocate\n+bool allocate(MTLContext *context)\n * opensubdiv\n * osd\n- * mtlComputeEvaluator.h\n+ * mtlVertexBuffer.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00827.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00827.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/d3d11Mesh.h File Reference\n+OpenSubdiv: opensubdiv/osd/hlslPatchShaderSource.h File Reference\n \n \n \n \n \n \n \n@@ -81,43 +81,43 @@\n \n \n \n
\n
\n-Namespaces |\n-Typedefs
\n-
d3d11Mesh.h File Reference
\n+Classes |\n+Namespaces
\n+
hlslPatchShaderSource.h File Reference
\n \n
\n
#include "../version.h"
\n-#include "../osd/mesh.h"
\n-#include "../osd/d3d11PatchTable.h"
\n+#include <string>
\n+#include "../far/patchDescriptor.h"
\n
\n

Go to the source code of this file.

\n \n+\n+\n+\n+

\n+Classes

class  HLSLPatchShaderSource
 
\n \n \n \n \n \n \n \n-

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n-\n-\n-\n

\n-Typedefs

typedef MeshInterface< D3D11PatchTable > D3D11MeshInterface
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,27 +4,27 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-Namespaces | Typedefs\n-d3d11Mesh.h File Reference\n+Classes | Namespaces\n+hlslPatchShaderSource.h File Reference\n #include \"../version.h\"\n-#include \"../osd/mesh.h\"\n-#include \"../osd/d3d11PatchTable.h\"\n+#include \n+#include \"../far/patchDescriptor.h\"\n Go_to_the_source_code_of_this_file.\n+ Classes\n+class \u00a0HLSLPatchShaderSource\n+\u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n- Typedefs\n-typedef MeshInterface< D3D11PatchTable >\u00a0D3D11MeshInterface\n-\u00a0\n * opensubdiv\n * osd\n- * d3d11Mesh.h\n+ * hlslPatchShaderSource.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00827.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00827.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00827 = [\n- [\"D3D11MeshInterface\", \"a00827.html#ae5c8e80d05d6eaee105d8eecfa9ab609\", null]\n+ [\"HLSLPatchShaderSource\", \"a01257.html\", null]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00827_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00827_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/d3d11Mesh.h Source File\n+OpenSubdiv: opensubdiv/osd/hlslPatchShaderSource.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
d3d11Mesh.h
\n+
hlslPatchShaderSource.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2013 Pixar
\n+
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,45 +107,59 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_D3D11MESH_H
\n-
26#define OPENSUBDIV3_OSD_D3D11MESH_H
\n+
25#ifndef OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H
\n+
26#define OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H
\n
27
\n
28#include "../version.h"
\n-
29
\n-
30#include "../osd/mesh.h"
\n-
31#include "../osd/d3d11PatchTable.h"
\n-
32
\n-
33namespace OpenSubdiv {
\n-
34namespace OPENSUBDIV_VERSION {
\n-
35
\n-
36namespace Osd {
\n-
37
\n-
38typedef MeshInterface<D3D11PatchTable> D3D11MeshInterface;
\n-
39
\n+
29#include <string>
\n+
30#include "../far/patchDescriptor.h"
\n+
31
\n+
32namespace OpenSubdiv {
\n+
33namespace OPENSUBDIV_VERSION {
\n+
34
\n+
35namespace Osd {
\n+
36
\n+
37class HLSLPatchShaderSource {
\n+
38public:
\n+
39 static std::string GetCommonShaderSource();
\n
40
\n-
41} // end namespace Osd
\n+
41 static std::string GetPatchBasisShaderSource();
\n
42
\n-
43} // end namespace OPENSUBDIV_VERSION
\n-
44using namespace OPENSUBDIV_VERSION;
\n-
45
\n-
46} // end namespace OpenSubdiv
\n-
47
\n-
48#endif // OPENSUBDIV3_OSD_D3D11MESH_H
\n+
43 static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type);
\n+
44
\n+
45 static std::string GetHullShaderSource(Far::PatchDescriptor::Type type);
\n+
46
\n+
47 static std::string GetDomainShaderSource(Far::PatchDescriptor::Type type);
\n+
48};
\n+
49
\n+
50} // end namespace Osd
\n+
51
\n+
52} // end namespace OPENSUBDIV_VERSION
\n+
53using namespace OPENSUBDIV_VERSION;
\n+
54
\n+
55} // end namespace OpenSubdiv
\n+
56
\n+
57#endif // OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11MeshInterface
MeshInterface< D3D11PatchTable > D3D11MeshInterface
Definition: d3d11Mesh.h:38
\n-\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::GetVertexShaderSource
static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::GetDomainShaderSource
static std::string GetDomainShaderSource(Far::PatchDescriptor::Type type)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::GetHullShaderSource
static std::string GetHullShaderSource(Far::PatchDescriptor::Type type)
\n+\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-d3d11Mesh.h\n+hlslPatchShaderSource.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2013 Pixar\n+ 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,42 +30,65 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_D3D11MESH_H\n- 26#define OPENSUBDIV3_OSD_D3D11MESH_H\n+ 25#ifndef OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H\n+ 26#define OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H\n 27\n 28#include \"../version.h\"\n- 29\n- 30#include \"../osd/mesh.h\"\n- 31#include \"../osd/d3d11PatchTable.h\"\n- 32\n- 33namespace OpenSubdiv {\n- 34namespace OPENSUBDIV_VERSION {\n- 35\n- 36namespace Osd {\n- 37\n-38typedef MeshInterface D3D11MeshInterface;\n- 39\n+ 29#include \n+ 30#include \"../far/patchDescriptor.h\"\n+ 31\n+ 32namespace OpenSubdiv {\n+ 33namespace OPENSUBDIV_VERSION {\n+ 34\n+ 35namespace Osd {\n+ 36\n+37class HLSLPatchShaderSource {\n+ 38public:\n+39 static std::string GetCommonShaderSource();\n 40\n- 41} // end namespace Osd\n+41 static std::string GetPatchBasisShaderSource();\n 42\n- 43} // end namespace OPENSUBDIV_VERSION\n- 44using namespace OPENSUBDIV_VERSION;\n- 45\n- 46} // end namespace OpenSubdiv\n- 47\n- 48#endif // OPENSUBDIV3_OSD_D3D11MESH_H\n+43 static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type);\n+ 44\n+45 static std::string GetHullShaderSource(Far::PatchDescriptor::Type type);\n+ 46\n+47 static std::string GetDomainShaderSource(Far::PatchDescriptor::Type type);\n+ 48};\n+ 49\n+ 50} // end namespace Osd\n+ 51\n+ 52} // end namespace OPENSUBDIV_VERSION\n+ 53using namespace OPENSUBDIV_VERSION;\n+ 54\n+ 55} // end namespace OpenSubdiv\n+ 56\n+ 57#endif // OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11MeshInterface\n-MeshInterface< D3D11PatchTable > D3D11MeshInterface\n-Definition: d3d11Mesh.h:38\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface\n-Definition: mesh.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n+Type\n+Definition: patchDescriptor.h:48\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource\n+Definition: hlslPatchShaderSource.h:37\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::\n+GetVertexShaderSource\n+static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::\n+GetDomainShaderSource\n+static std::string GetDomainShaderSource(Far::PatchDescriptor::Type type)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::GetHullShaderSource\n+static std::string GetHullShaderSource(Far::PatchDescriptor::Type type)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::\n+GetPatchBasisShaderSource\n+static std::string GetPatchBasisShaderSource()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::\n+GetCommonShaderSource\n+static std::string GetCommonShaderSource()\n * opensubdiv\n * osd\n- * d3d11Mesh.h\n+ * hlslPatchShaderSource.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00830.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00830.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/d3d11VertexBuffer.h File Reference\n+OpenSubdiv: opensubdiv/osd/d3d11Mesh.h File Reference\n \n \n \n \n \n \n \n@@ -81,42 +81,43 @@\n \n \n \n
\n
\n-Classes |\n-Namespaces
\n-
d3d11VertexBuffer.h File Reference
\n+Namespaces |\n+Typedefs
\n+
d3d11Mesh.h File Reference
\n \n
\n
#include "../version.h"
\n+#include "../osd/mesh.h"
\n+#include "../osd/d3d11PatchTable.h"
\n
\n

Go to the source code of this file.

\n \n-\n-\n-\n-\n-

\n-Classes

class  D3D11VertexBuffer
 Concrete vertex buffer class for DirectX subdivision and DirectX drawing. More...
 
\n \n \n \n \n \n \n \n+

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n+\n+\n+\n

\n+Typedefs

typedef MeshInterface< D3D11PatchTable > D3D11MeshInterface
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,27 +4,27 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-Classes | Namespaces\n-d3d11VertexBuffer.h File Reference\n+Namespaces | Typedefs\n+d3d11Mesh.h File Reference\n #include \"../version.h\"\n+#include \"../osd/mesh.h\"\n+#include \"../osd/d3d11PatchTable.h\"\n Go_to_the_source_code_of_this_file.\n- Classes\n-class \u00a0D3D11VertexBuffer\n-\u00a0 Concrete vertex buffer class for DirectX subdivision and DirectX\n- drawing. More...\n-\u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n+ Typedefs\n+typedef MeshInterface< D3D11PatchTable >\u00a0D3D11MeshInterface\n+\u00a0\n * opensubdiv\n * osd\n- * d3d11VertexBuffer.h\n+ * d3d11Mesh.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00830.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00830.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00830 = [\n- [\"D3D11VertexBuffer\", \"a01205.html\", \"a01205\"]\n+ [\"D3D11MeshInterface\", \"a00830.html#ae5c8e80d05d6eaee105d8eecfa9ab609\", null]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00830_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00830_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/d3d11VertexBuffer.h Source File\n+OpenSubdiv: opensubdiv/osd/d3d11Mesh.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
d3d11VertexBuffer.h
\n+
d3d11Mesh.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,89 +107,45 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H
\n-
26#define OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H
\n+
25#ifndef OPENSUBDIV3_OSD_D3D11MESH_H
\n+
26#define OPENSUBDIV3_OSD_D3D11MESH_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30struct ID3D11Buffer;
\n-
31struct ID3D11Device;
\n-
32struct ID3D11DeviceContext;
\n-
33struct ID3D11UnorderedAccessView;
\n-
34
\n-
35namespace OpenSubdiv {
\n-
36namespace OPENSUBDIV_VERSION {
\n+
30#include "../osd/mesh.h"
\n+
31#include "../osd/d3d11PatchTable.h"
\n+
32
\n+
33namespace OpenSubdiv {
\n+
34namespace OPENSUBDIV_VERSION {
\n+
35
\n+
36namespace Osd {
\n
37
\n-
38namespace Osd {
\n+
38typedef MeshInterface<D3D11PatchTable> D3D11MeshInterface;
\n
39
\n-
46class D3D11VertexBuffer {
\n-
47public:
\n-
49 static D3D11VertexBuffer * Create(int numElements, int numVertices,
\n-
50 ID3D11DeviceContext *deviceContext);
\n-
51
\n-
53 virtual ~D3D11VertexBuffer();
\n-
54
\n-
57 void UpdateData(const float *src, int startVertex, int numVertices,
\n-
58 ID3D11DeviceContext *deviceContext);
\n-
59
\n-
61 int GetNumElements() const;
\n-
62
\n-
64 int GetNumVertices() const;
\n-
65
\n-
67 ID3D11Buffer *BindD3D11Buffer(ID3D11DeviceContext *deviceContext);
\n-
68
\n-
70 ID3D11Buffer *BindVBO(ID3D11DeviceContext *deviceContext) {
\n-
71 return BindD3D11Buffer(deviceContext);
\n-
72 }
\n-
73
\n-
75 ID3D11UnorderedAccessView *BindD3D11UAV(ID3D11DeviceContext *deviceContext);
\n-
76
\n-
77protected:
\n-
79 D3D11VertexBuffer(int numElements, int numVertices);
\n-
80
\n-
81 // Allocates D3D11 buffer
\n-
82 bool allocate(ID3D11Device *device);
\n-
83
\n-
84private:
\n-
85 int _numElements;
\n-
86 int _numVertices;
\n-
87 ID3D11Buffer *_buffer;
\n-
88 ID3D11Buffer *_uploadBuffer;
\n-
89 ID3D11UnorderedAccessView *_uav;
\n-
90};
\n-
91
\n-
92} // end namespace Osd
\n-
93
\n-
94} // end namespace OPENSUBDIV_VERSION
\n-
95using namespace OPENSUBDIV_VERSION;
\n-
96
\n-
97} // end namespace OpenSubdiv
\n-
98
\n-
99#endif // OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H
\n+
40
\n+
41} // end namespace Osd
\n+
42
\n+
43} // end namespace OPENSUBDIV_VERSION
\n+
44using namespace OPENSUBDIV_VERSION;
\n+
45
\n+
46} // end namespace OpenSubdiv
\n+
47
\n+
48#endif // OPENSUBDIV3_OSD_D3D11MESH_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer
Concrete vertex buffer class for DirectX subdivision and DirectX drawing.
Definition: d3d11VertexBuffer.h:46
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::GetNumElements
int GetNumElements() const
Returns how many elements defined in this vertex buffer.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::UpdateData
void UpdateData(const float *src, int startVertex, int numVertices, ID3D11DeviceContext *deviceContext)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::Create
static D3D11VertexBuffer * Create(int numElements, int numVertices, ID3D11DeviceContext *deviceContext)
Creator. Returns NULL if error.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::D3D11VertexBuffer
D3D11VertexBuffer(int numElements, int numVertices)
Constructor.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::GetNumVertices
int GetNumVertices() const
Returns how many vertices allocated in this vertex buffer.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindD3D11Buffer
ID3D11Buffer * BindD3D11Buffer(ID3D11DeviceContext *deviceContext)
Returns the D3D11 buffer object.
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindD3D11UAV
ID3D11UnorderedAccessView * BindD3D11UAV(ID3D11DeviceContext *deviceContext)
Returns the D3D11 UAV.
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindVBO
ID3D11Buffer * BindVBO(ID3D11DeviceContext *deviceContext)
Returns the D3D11 buffer object (for Osd::Mesh interface)
Definition: d3d11VertexBuffer.h:70
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11MeshInterface
MeshInterface< D3D11PatchTable > D3D11MeshInterface
Definition: d3d11Mesh.h:38
\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-d3d11VertexBuffer.h\n+d3d11Mesh.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,106 +30,42 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H\n- 26#define OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H\n+ 25#ifndef OPENSUBDIV3_OSD_D3D11MESH_H\n+ 26#define OPENSUBDIV3_OSD_D3D11MESH_H\n 27\n 28#include \"../version.h\"\n 29\n- 30struct ID3D11Buffer;\n- 31struct ID3D11Device;\n- 32struct ID3D11DeviceContext;\n- 33struct ID3D11UnorderedAccessView;\n- 34\n- 35namespace OpenSubdiv {\n- 36namespace OPENSUBDIV_VERSION {\n+ 30#include \"../osd/mesh.h\"\n+ 31#include \"../osd/d3d11PatchTable.h\"\n+ 32\n+ 33namespace OpenSubdiv {\n+ 34namespace OPENSUBDIV_VERSION {\n+ 35\n+ 36namespace Osd {\n 37\n- 38namespace Osd {\n+38typedef MeshInterface D3D11MeshInterface;\n 39\n-46class D3D11VertexBuffer {\n- 47public:\n-49 static D3D11VertexBuffer * Create(int numElements, int numVertices,\n- 50 ID3D11DeviceContext *deviceContext);\n- 51\n-53 virtual ~D3D11VertexBuffer();\n- 54\n-57 void UpdateData(const float *src, int startVertex, int numVertices,\n- 58 ID3D11DeviceContext *deviceContext);\n- 59\n-61 int GetNumElements() const;\n- 62\n-64 int GetNumVertices() const;\n- 65\n-67 ID3D11Buffer *BindD3D11Buffer(ID3D11DeviceContext *deviceContext);\n- 68\n-70 ID3D11Buffer *BindVBO(ID3D11DeviceContext *deviceContext) {\n- 71 return BindD3D11Buffer(deviceContext);\n- 72 }\n- 73\n-75 ID3D11UnorderedAccessView *BindD3D11UAV(ID3D11DeviceContext *deviceContext);\n- 76\n- 77protected:\n-79 D3D11VertexBuffer(int numElements, int numVertices);\n- 80\n- 81 // Allocates D3D11 buffer\n-82 bool allocate(ID3D11Device *device);\n- 83\n- 84private:\n- 85 int _numElements;\n- 86 int _numVertices;\n- 87 ID3D11Buffer *_buffer;\n- 88 ID3D11Buffer *_uploadBuffer;\n- 89 ID3D11UnorderedAccessView *_uav;\n- 90};\n- 91\n- 92} // end namespace Osd\n- 93\n- 94} // end namespace OPENSUBDIV_VERSION\n- 95using namespace OPENSUBDIV_VERSION;\n- 96\n- 97} // end namespace OpenSubdiv\n- 98\n- 99#endif // OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H\n+ 40\n+ 41} // end namespace Osd\n+ 42\n+ 43} // end namespace OPENSUBDIV_VERSION\n+ 44using namespace OPENSUBDIV_VERSION;\n+ 45\n+ 46} // end namespace OpenSubdiv\n+ 47\n+ 48#endif // OPENSUBDIV3_OSD_D3D11MESH_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer\n-Concrete vertex buffer class for DirectX subdivision and DirectX drawing.\n-Definition: d3d11VertexBuffer.h:46\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::GetNumElements\n-int GetNumElements() const\n-Returns how many elements defined in this vertex buffer.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::UpdateData\n-void UpdateData(const float *src, int startVertex, int numVertices,\n-ID3D11DeviceContext *deviceContext)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::Create\n-static D3D11VertexBuffer * Create(int numElements, int numVertices,\n-ID3D11DeviceContext *deviceContext)\n-Creator. Returns NULL if error.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::D3D11VertexBuffer\n-D3D11VertexBuffer(int numElements, int numVertices)\n-Constructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::GetNumVertices\n-int GetNumVertices() const\n-Returns how many vertices allocated in this vertex buffer.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindD3D11Buffer\n-ID3D11Buffer * BindD3D11Buffer(ID3D11DeviceContext *deviceContext)\n-Returns the D3D11 buffer object.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::~D3D11VertexBuffer\n-virtual ~D3D11VertexBuffer()\n-Destructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindD3D11UAV\n-ID3D11UnorderedAccessView * BindD3D11UAV(ID3D11DeviceContext *deviceContext)\n-Returns the D3D11 UAV.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::allocate\n-bool allocate(ID3D11Device *device)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindVBO\n-ID3D11Buffer * BindVBO(ID3D11DeviceContext *deviceContext)\n-Returns the D3D11 buffer object (for Osd::Mesh interface)\n-Definition: d3d11VertexBuffer.h:70\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11MeshInterface\n+MeshInterface< D3D11PatchTable > D3D11MeshInterface\n+Definition: d3d11Mesh.h:38\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface\n+Definition: mesh.h:68\n * opensubdiv\n * osd\n- * d3d11VertexBuffer.h\n+ * d3d11Mesh.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00833.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00833.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/hlslPatchShaderSource.h File Reference\n+OpenSubdiv: opensubdiv/osd/d3d11VertexBuffer.h File Reference\n \n \n \n \n \n \n \n@@ -83,26 +83,25 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
hlslPatchShaderSource.h File Reference
\n+
d3d11VertexBuffer.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include <string>
\n-#include "../far/patchDescriptor.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n+\n+\n \n

\n Classes

class  HLSLPatchShaderSource
class  D3D11VertexBuffer
 Concrete vertex buffer class for DirectX subdivision and DirectX drawing. More...
 
\n \n \n \n \n@@ -111,13 +110,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,26 +5,26 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-hlslPatchShaderSource.h File Reference\n+d3d11VertexBuffer.h File Reference\n #include \"../version.h\"\n-#include \n-#include \"../far/patchDescriptor.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0HLSLPatchShaderSource\n+class \u00a0D3D11VertexBuffer\n+\u00a0 Concrete vertex buffer class for DirectX subdivision and DirectX\n+ drawing. More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * hlslPatchShaderSource.h\n+ * d3d11VertexBuffer.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00833.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00833.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00833 = [\n- [\"HLSLPatchShaderSource\", \"a01257.html\", null]\n+ [\"D3D11VertexBuffer\", \"a01205.html\", \"a01205\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00833_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00833_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/hlslPatchShaderSource.h Source File\n+OpenSubdiv: opensubdiv/osd/d3d11VertexBuffer.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
hlslPatchShaderSource.h
\n+
d3d11VertexBuffer.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2015 Pixar
\n+
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,59 +107,89 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H
\n-
26#define OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H
\n+
25#ifndef OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H
\n+
26#define OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H
\n
27
\n
28#include "../version.h"
\n-
29#include <string>
\n-
30#include "../far/patchDescriptor.h"
\n-
31
\n-
32namespace OpenSubdiv {
\n-
33namespace OPENSUBDIV_VERSION {
\n+
29
\n+
30struct ID3D11Buffer;
\n+
31struct ID3D11Device;
\n+
32struct ID3D11DeviceContext;
\n+
33struct ID3D11UnorderedAccessView;
\n
34
\n-
35namespace Osd {
\n-
36
\n-
37class HLSLPatchShaderSource {
\n-
38public:
\n-
39 static std::string GetCommonShaderSource();
\n-
40
\n-
41 static std::string GetPatchBasisShaderSource();
\n-
42
\n-
43 static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type);
\n-
44
\n-
45 static std::string GetHullShaderSource(Far::PatchDescriptor::Type type);
\n-
46
\n-
47 static std::string GetDomainShaderSource(Far::PatchDescriptor::Type type);
\n-
48};
\n-
49
\n-
50} // end namespace Osd
\n+
35namespace OpenSubdiv {
\n+
36namespace OPENSUBDIV_VERSION {
\n+
37
\n+
38namespace Osd {
\n+
39
\n+
46class D3D11VertexBuffer {
\n+
47public:
\n+
49 static D3D11VertexBuffer * Create(int numElements, int numVertices,
\n+
50 ID3D11DeviceContext *deviceContext);
\n
51
\n-
52} // end namespace OPENSUBDIV_VERSION
\n-
53using namespace OPENSUBDIV_VERSION;
\n+
53 virtual ~D3D11VertexBuffer();
\n
54
\n-
55} // end namespace OpenSubdiv
\n-
56
\n-
57#endif // OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H
\n+
57 void UpdateData(const float *src, int startVertex, int numVertices,
\n+
58 ID3D11DeviceContext *deviceContext);
\n+
59
\n+
61 int GetNumElements() const;
\n+
62
\n+
64 int GetNumVertices() const;
\n+
65
\n+
67 ID3D11Buffer *BindD3D11Buffer(ID3D11DeviceContext *deviceContext);
\n+
68
\n+
70 ID3D11Buffer *BindVBO(ID3D11DeviceContext *deviceContext) {
\n+
71 return BindD3D11Buffer(deviceContext);
\n+
72 }
\n+
73
\n+
75 ID3D11UnorderedAccessView *BindD3D11UAV(ID3D11DeviceContext *deviceContext);
\n+
76
\n+
77protected:
\n+
79 D3D11VertexBuffer(int numElements, int numVertices);
\n+
80
\n+
81 // Allocates D3D11 buffer
\n+
82 bool allocate(ID3D11Device *device);
\n+
83
\n+
84private:
\n+
85 int _numElements;
\n+
86 int _numVertices;
\n+
87 ID3D11Buffer *_buffer;
\n+
88 ID3D11Buffer *_uploadBuffer;
\n+
89 ID3D11UnorderedAccessView *_uav;
\n+
90};
\n+
91
\n+
92} // end namespace Osd
\n+
93
\n+
94} // end namespace OPENSUBDIV_VERSION
\n+
95using namespace OPENSUBDIV_VERSION;
\n+
96
\n+
97} // end namespace OpenSubdiv
\n+
98
\n+
99#endif // OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H
\n \n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::GetVertexShaderSource
static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::GetDomainShaderSource
static std::string GetDomainShaderSource(Far::PatchDescriptor::Type type)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::GetHullShaderSource
static std::string GetHullShaderSource(Far::PatchDescriptor::Type type)
\n-\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer
Concrete vertex buffer class for DirectX subdivision and DirectX drawing.
Definition: d3d11VertexBuffer.h:46
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::GetNumElements
int GetNumElements() const
Returns how many elements defined in this vertex buffer.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::UpdateData
void UpdateData(const float *src, int startVertex, int numVertices, ID3D11DeviceContext *deviceContext)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::Create
static D3D11VertexBuffer * Create(int numElements, int numVertices, ID3D11DeviceContext *deviceContext)
Creator. Returns NULL if error.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::D3D11VertexBuffer
D3D11VertexBuffer(int numElements, int numVertices)
Constructor.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::GetNumVertices
int GetNumVertices() const
Returns how many vertices allocated in this vertex buffer.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindD3D11Buffer
ID3D11Buffer * BindD3D11Buffer(ID3D11DeviceContext *deviceContext)
Returns the D3D11 buffer object.
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindD3D11UAV
ID3D11UnorderedAccessView * BindD3D11UAV(ID3D11DeviceContext *deviceContext)
Returns the D3D11 UAV.
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindVBO
ID3D11Buffer * BindVBO(ID3D11DeviceContext *deviceContext)
Returns the D3D11 buffer object (for Osd::Mesh interface)
Definition: d3d11VertexBuffer.h:70
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-hlslPatchShaderSource.h\n+d3d11VertexBuffer.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2015 Pixar\n+ 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,65 +30,106 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H\n- 26#define OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H\n+ 25#ifndef OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H\n+ 26#define OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H\n 27\n 28#include \"../version.h\"\n- 29#include \n- 30#include \"../far/patchDescriptor.h\"\n- 31\n- 32namespace OpenSubdiv {\n- 33namespace OPENSUBDIV_VERSION {\n+ 29\n+ 30struct ID3D11Buffer;\n+ 31struct ID3D11Device;\n+ 32struct ID3D11DeviceContext;\n+ 33struct ID3D11UnorderedAccessView;\n 34\n- 35namespace Osd {\n- 36\n-37class HLSLPatchShaderSource {\n- 38public:\n-39 static std::string GetCommonShaderSource();\n- 40\n-41 static std::string GetPatchBasisShaderSource();\n- 42\n-43 static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type);\n- 44\n-45 static std::string GetHullShaderSource(Far::PatchDescriptor::Type type);\n- 46\n-47 static std::string GetDomainShaderSource(Far::PatchDescriptor::Type type);\n- 48};\n- 49\n- 50} // end namespace Osd\n+ 35namespace OpenSubdiv {\n+ 36namespace OPENSUBDIV_VERSION {\n+ 37\n+ 38namespace Osd {\n+ 39\n+46class D3D11VertexBuffer {\n+ 47public:\n+49 static D3D11VertexBuffer * Create(int numElements, int numVertices,\n+ 50 ID3D11DeviceContext *deviceContext);\n 51\n- 52} // end namespace OPENSUBDIV_VERSION\n- 53using namespace OPENSUBDIV_VERSION;\n+53 virtual ~D3D11VertexBuffer();\n 54\n- 55} // end namespace OpenSubdiv\n- 56\n- 57#endif // OPENSUBDIV3_OSD_HLSL_PATCH_SHADER_SOURCE_H\n+57 void UpdateData(const float *src, int startVertex, int numVertices,\n+ 58 ID3D11DeviceContext *deviceContext);\n+ 59\n+61 int GetNumElements() const;\n+ 62\n+64 int GetNumVertices() const;\n+ 65\n+67 ID3D11Buffer *BindD3D11Buffer(ID3D11DeviceContext *deviceContext);\n+ 68\n+70 ID3D11Buffer *BindVBO(ID3D11DeviceContext *deviceContext) {\n+ 71 return BindD3D11Buffer(deviceContext);\n+ 72 }\n+ 73\n+75 ID3D11UnorderedAccessView *BindD3D11UAV(ID3D11DeviceContext *deviceContext);\n+ 76\n+ 77protected:\n+79 D3D11VertexBuffer(int numElements, int numVertices);\n+ 80\n+ 81 // Allocates D3D11 buffer\n+82 bool allocate(ID3D11Device *device);\n+ 83\n+ 84private:\n+ 85 int _numElements;\n+ 86 int _numVertices;\n+ 87 ID3D11Buffer *_buffer;\n+ 88 ID3D11Buffer *_uploadBuffer;\n+ 89 ID3D11UnorderedAccessView *_uav;\n+ 90};\n+ 91\n+ 92} // end namespace Osd\n+ 93\n+ 94} // end namespace OPENSUBDIV_VERSION\n+ 95using namespace OPENSUBDIV_VERSION;\n+ 96\n+ 97} // end namespace OpenSubdiv\n+ 98\n+ 99#endif // OPENSUBDIV3_OSD_D3D11_VERTEX_BUFFER_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n-Type\n-Definition: patchDescriptor.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource\n-Definition: hlslPatchShaderSource.h:37\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::\n-GetVertexShaderSource\n-static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::\n-GetDomainShaderSource\n-static std::string GetDomainShaderSource(Far::PatchDescriptor::Type type)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::GetHullShaderSource\n-static std::string GetHullShaderSource(Far::PatchDescriptor::Type type)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::\n-GetPatchBasisShaderSource\n-static std::string GetPatchBasisShaderSource()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::HLSLPatchShaderSource::\n-GetCommonShaderSource\n-static std::string GetCommonShaderSource()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer\n+Concrete vertex buffer class for DirectX subdivision and DirectX drawing.\n+Definition: d3d11VertexBuffer.h:46\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::GetNumElements\n+int GetNumElements() const\n+Returns how many elements defined in this vertex buffer.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::UpdateData\n+void UpdateData(const float *src, int startVertex, int numVertices,\n+ID3D11DeviceContext *deviceContext)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::Create\n+static D3D11VertexBuffer * Create(int numElements, int numVertices,\n+ID3D11DeviceContext *deviceContext)\n+Creator. Returns NULL if error.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::D3D11VertexBuffer\n+D3D11VertexBuffer(int numElements, int numVertices)\n+Constructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::GetNumVertices\n+int GetNumVertices() const\n+Returns how many vertices allocated in this vertex buffer.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindD3D11Buffer\n+ID3D11Buffer * BindD3D11Buffer(ID3D11DeviceContext *deviceContext)\n+Returns the D3D11 buffer object.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::~D3D11VertexBuffer\n+virtual ~D3D11VertexBuffer()\n+Destructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindD3D11UAV\n+ID3D11UnorderedAccessView * BindD3D11UAV(ID3D11DeviceContext *deviceContext)\n+Returns the D3D11 UAV.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::allocate\n+bool allocate(ID3D11Device *device)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::BindVBO\n+ID3D11Buffer * BindVBO(ID3D11DeviceContext *deviceContext)\n+Returns the D3D11 buffer object (for Osd::Mesh interface)\n+Definition: d3d11VertexBuffer.h:70\n * opensubdiv\n * osd\n- * hlslPatchShaderSource.h\n+ * d3d11VertexBuffer.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00836.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00836.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/d3d11LegacyGregoryPatchTable.h File Reference\n+OpenSubdiv: opensubdiv/osd/d3d11PatchTable.h File Reference\n \n \n \n \n \n \n \n@@ -83,41 +83,45 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
d3d11LegacyGregoryPatchTable.h File Reference
\n+
d3d11PatchTable.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../far/patchTable.h"
\n+#include <vector>
\n+#include "../far/patchDescriptor.h"
\n #include "../osd/nonCopyable.h"
\n+#include "../osd/types.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n+\n \n

\n Classes

class  D3D11LegacyGregoryPatchTable
class  D3D11PatchTable
 
\n \n \n \n \n \n+\n+\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,26 +5,30 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-d3d11LegacyGregoryPatchTable.h File Reference\n+d3d11PatchTable.h File Reference\n #include \"../version.h\"\n-#include \"../far/patchTable.h\"\n+#include \n+#include \"../far/patchDescriptor.h\"\n #include \"../osd/nonCopyable.h\"\n+#include \"../osd/types.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0D3D11LegacyGregoryPatchTable\n+class \u00a0D3D11PatchTable\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n+\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * d3d11LegacyGregoryPatchTable.h\n+ * d3d11PatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00836.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00836.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00836 = [\n- [\"D3D11LegacyGregoryPatchTable\", \"a01197.html\", \"a01197\"]\n+ [\"D3D11PatchTable\", \"a01201.html\", \"a01201\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00836_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00836_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/d3d11LegacyGregoryPatchTable.h Source File\n+OpenSubdiv: opensubdiv/osd/d3d11PatchTable.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
d3d11LegacyGregoryPatchTable.h
\n+
d3d11PatchTable.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,110 +107,108 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H
\n-
26#define OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H
\n+
25#ifndef OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n+
26#define OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30#include "../far/patchTable.h"
\n-
31#include "../osd/nonCopyable.h"
\n-
32
\n-
33struct ID3D11Buffer;
\n-
34struct ID3D11ShaderResourceView;
\n-
35struct ID3D11Device;
\n-
36struct ID3D11DeviceContext;
\n-
37
\n-
38namespace OpenSubdiv {
\n-
39namespace OPENSUBDIV_VERSION {
\n-
40
\n-
41namespace Osd {
\n+
30#include <vector>
\n+
31#include "../far/patchDescriptor.h"
\n+
32#include "../osd/nonCopyable.h"
\n+
33#include "../osd/types.h"
\n+
34
\n+
35struct ID3D11Buffer;
\n+
36struct ID3D11ShaderResourceView;
\n+
37struct ID3D11Device;
\n+
38struct ID3D11DeviceContext;
\n+
39
\n+
40namespace OpenSubdiv {
\n+
41namespace OPENSUBDIV_VERSION {
\n
42
\n-
43class D3D11LegacyGregoryPatchTable
\n-
44 : private NonCopyable<D3D11LegacyGregoryPatchTable> {
\n-
45public:
\n-
46 ~D3D11LegacyGregoryPatchTable();
\n-
47
\n-
48 template<typename DEVICE_CONTEXT>
\n-
49 static D3D11LegacyGregoryPatchTable *Create(
\n-
50 Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context) {
\n-
51 return Create(farPatchTable, context->GetDeviceContext());
\n-
52 }
\n-
53
\n-
54 static D3D11LegacyGregoryPatchTable *Create(
\n-
55 Far::PatchTable const *farPatchTable,
\n-
56 ID3D11DeviceContext *deviceContext);
\n-
57
\n-
58 void UpdateVertexBuffer(ID3D11Buffer *vbo,
\n-
59 int numVertices, int numVertexElements,
\n-
60 ID3D11DeviceContext *pd3d11DeviceContext);
\n+
43namespace Far{
\n+
44 class PatchTable;
\n+
45};
\n+
46
\n+
47namespace Osd {
\n+
48
\n+
49class D3D11PatchTable : private NonCopyable<D3D11PatchTable> {
\n+
50public:
\n+
51 typedef ID3D11Buffer * VertexBufferBinding;
\n+
52
\n+
53 D3D11PatchTable();
\n+
54 ~D3D11PatchTable();
\n+
55
\n+
56 template<typename DEVICE_CONTEXT>
\n+
57 static D3D11PatchTable *Create(Far::PatchTable const *farPatchTable,
\n+
58 DEVICE_CONTEXT context) {
\n+
59 return Create(farPatchTable, context->GetDeviceContext());
\n+
60 }
\n
61
\n-
62 ID3D11ShaderResourceView* GetVertexSRV() const {
\n-
63 return _vertexSRV;
\n-
64 }
\n-
65
\n-
66 ID3D11ShaderResourceView* GetVertexValenceSRV() const {
\n-
67 return _vertexValenceSRV;
\n-
68 }
\n-
69
\n-
70 ID3D11ShaderResourceView* GetQuadOffsetsSRV() const {
\n-
71 return _quadOffsetsSRV;
\n+
62 static D3D11PatchTable *Create(Far::PatchTable const *farPatchTable,
\n+
63 ID3D11DeviceContext *deviceContext);
\n+
64
\n+
65 PatchArrayVector const &GetPatchArrays() const {
\n+
66 return _patchArrays;
\n+
67 }
\n+
68
\n+
70 ID3D11Buffer* GetPatchIndexBuffer() const {
\n+
71 return _indexBuffer;
\n
72 }
\n
73
\n-
74 int GetQuadOffsetsBase(Far::PatchDescriptor::Type type) {
\n-
75 if (type == Far::PatchDescriptor::GREGORY_BOUNDARY) {
\n-
76 return _quadOffsetsBase[1];
\n-
77 }
\n-
78 return _quadOffsetsBase[0];
\n-
79 }
\n-
80
\n-
81protected:
\n-
82 D3D11LegacyGregoryPatchTable();
\n+
75 ID3D11ShaderResourceView* GetPatchParamSRV() const {
\n+
76 return _patchParamBufferSRV;
\n+
77 }
\n+
78
\n+
79protected:
\n+
80 // allocate buffers from patchTable
\n+
81 bool allocate(Far::PatchTable const *farPatchTable,
\n+
82 ID3D11DeviceContext *deviceContext);
\n
83
\n-
84private:
\n-
85 ID3D11Buffer* _vertexValenceBuffer;
\n-
86 ID3D11Buffer* _quadOffsetsBuffer;
\n-
87 ID3D11ShaderResourceView* _vertexSRV;
\n-
88 ID3D11ShaderResourceView* _vertexValenceSRV;
\n-
89 ID3D11ShaderResourceView* _quadOffsetsSRV;
\n-
90 int _quadOffsetsBase[2]; // gregory, boundaryGregory
\n-
91};
\n-
92
\n+
84 PatchArrayVector _patchArrays;
\n+
85
\n+
86 ID3D11Buffer *_indexBuffer;
\n+
87 ID3D11Buffer *_patchParamBuffer;
\n+
88 ID3D11ShaderResourceView *_patchParamBufferSRV;
\n+
89};
\n+
90
\n+
91
\n+
92} // end namespace Osd
\n
93
\n-
94
\n-
95} // end namespace Osd
\n+
94} // end namespace OPENSUBDIV_VERSION
\n+
95using namespace OPENSUBDIV_VERSION;
\n
96
\n-
97} // end namespace OPENSUBDIV_VERSION
\n-
98using namespace OPENSUBDIV_VERSION;
\n-
99
\n-
100} // end namespace OpenSubdiv
\n-
101
\n-
102#endif // OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H
\n+
97} // end namespace OpenSubdiv
\n+
98
\n+
99#endif // OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n \n-\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::UpdateVertexBuffer
void UpdateVertexBuffer(ID3D11Buffer *vbo, int numVertices, int numVertexElements, ID3D11DeviceContext *pd3d11DeviceContext)
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::Create
static D3D11LegacyGregoryPatchTable * Create(Far::PatchTable const *farPatchTable, ID3D11DeviceContext *deviceContext)
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::Create
static D3D11LegacyGregoryPatchTable * Create(Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context)
Definition: d3d11LegacyGregoryPatchTable.h:49
\n-\n-\n-\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::Create
static D3D11PatchTable * Create(Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context)
Definition: d3d11PatchTable.h:57
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchArrays
PatchArrayVector const & GetPatchArrays() const
Definition: d3d11PatchTable.h:65
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::allocate
bool allocate(Far::PatchTable const *farPatchTable, ID3D11DeviceContext *deviceContext)
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchIndexBuffer
ID3D11Buffer * GetPatchIndexBuffer() const
Returns the index buffer containing the patch control vertices.
Definition: d3d11PatchTable.h:70
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::Create
static D3D11PatchTable * Create(Far::PatchTable const *farPatchTable, ID3D11DeviceContext *deviceContext)
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchParamSRV
ID3D11ShaderResourceView * GetPatchParamSRV() const
Returns the SRV containing the patch parameter.
Definition: d3d11PatchTable.h:75
\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-d3d11LegacyGregoryPatchTable.h\n+d3d11PatchTable.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,134 +30,134 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H\n- 26#define OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H\n+ 25#ifndef OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n+ 26#define OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n 27\n 28#include \"../version.h\"\n 29\n- 30#include \"../far/patchTable.h\"\n- 31#include \"../osd/nonCopyable.h\"\n- 32\n- 33struct ID3D11Buffer;\n- 34struct ID3D11ShaderResourceView;\n- 35struct ID3D11Device;\n- 36struct ID3D11DeviceContext;\n- 37\n- 38namespace OpenSubdiv {\n- 39namespace OPENSUBDIV_VERSION {\n- 40\n- 41namespace Osd {\n+ 30#include \n+ 31#include \"../far/patchDescriptor.h\"\n+ 32#include \"../osd/nonCopyable.h\"\n+ 33#include \"../osd/types.h\"\n+ 34\n+ 35struct ID3D11Buffer;\n+ 36struct ID3D11ShaderResourceView;\n+ 37struct ID3D11Device;\n+ 38struct ID3D11DeviceContext;\n+ 39\n+ 40namespace OpenSubdiv {\n+ 41namespace OPENSUBDIV_VERSION {\n 42\n-43class D3D11LegacyGregoryPatchTable\n- 44 : private NonCopyable {\n- 45public:\n-46 ~D3D11LegacyGregoryPatchTable();\n- 47\n- 48 template\n-49 static D3D11LegacyGregoryPatchTable *Create(\n- 50 Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context) {\n- 51 return Create(farPatchTable, context->GetDeviceContext());\n- 52 }\n- 53\n-54 static D3D11LegacyGregoryPatchTable *Create(\n- 55 Far::PatchTable const *farPatchTable,\n- 56 ID3D11DeviceContext *deviceContext);\n- 57\n-58 void UpdateVertexBuffer(ID3D11Buffer *vbo,\n- 59 int numVertices, int numVertexElements,\n- 60 ID3D11DeviceContext *pd3d11DeviceContext);\n+ 43namespace Far{\n+ 44 class PatchTable;\n+ 45};\n+ 46\n+ 47namespace Osd {\n+ 48\n+49class D3D11PatchTable : private NonCopyable {\n+ 50public:\n+51 typedef ID3D11Buffer * VertexBufferBinding;\n+ 52\n+53 D3D11PatchTable();\n+54 ~D3D11PatchTable();\n+ 55\n+ 56 template\n+57 static D3D11PatchTable *Create(Far::PatchTable const *farPatchTable,\n+ 58 DEVICE_CONTEXT context) {\n+ 59 return Create(farPatchTable, context->GetDeviceContext());\n+ 60 }\n 61\n-62 ID3D11ShaderResourceView* GetVertexSRV() const {\n- 63 return _vertexSRV;\n- 64 }\n- 65\n-66 ID3D11ShaderResourceView* GetVertexValenceSRV() const {\n- 67 return _vertexValenceSRV;\n- 68 }\n- 69\n-70 ID3D11ShaderResourceView* GetQuadOffsetsSRV() const {\n- 71 return _quadOffsetsSRV;\n+62 static D3D11PatchTable *Create(Far::PatchTable const *farPatchTable,\n+ 63 ID3D11DeviceContext *deviceContext);\n+ 64\n+65 PatchArrayVector const &GetPatchArrays() const {\n+ 66 return _patchArrays;\n+ 67 }\n+ 68\n+70 ID3D11Buffer* GetPatchIndexBuffer() const {\n+ 71 return _indexBuffer;\n 72 }\n 73\n-74 int GetQuadOffsetsBase(Far::PatchDescriptor::Type type) {\n- 75 if (type == Far::PatchDescriptor::GREGORY_BOUNDARY) {\n- 76 return _quadOffsetsBase[1];\n+75 ID3D11ShaderResourceView* GetPatchParamSRV() const {\n+ 76 return _patchParamBufferSRV;\n 77 }\n- 78 return _quadOffsetsBase[0];\n- 79 }\n- 80\n- 81protected:\n-82 D3D11LegacyGregoryPatchTable();\n+ 78\n+ 79protected:\n+ 80 // allocate buffers from patchTable\n+81 bool allocate(Far::PatchTable const *farPatchTable,\n+ 82 ID3D11DeviceContext *deviceContext);\n 83\n- 84private:\n- 85 ID3D11Buffer* _vertexValenceBuffer;\n- 86 ID3D11Buffer* _quadOffsetsBuffer;\n- 87 ID3D11ShaderResourceView* _vertexSRV;\n- 88 ID3D11ShaderResourceView* _vertexValenceSRV;\n- 89 ID3D11ShaderResourceView* _quadOffsetsSRV;\n- 90 int _quadOffsetsBase[2]; // gregory, boundaryGregory\n- 91};\n- 92\n+84 PatchArrayVector _patchArrays;\n+ 85\n+86 ID3D11Buffer *_indexBuffer;\n+87 ID3D11Buffer *_patchParamBuffer;\n+88 ID3D11ShaderResourceView *_patchParamBufferSRV;\n+ 89};\n+ 90\n+ 91\n+ 92} // end namespace Osd\n 93\n- 94\n- 95} // end namespace Osd\n+ 94} // end namespace OPENSUBDIV_VERSION\n+ 95using namespace OPENSUBDIV_VERSION;\n 96\n- 97} // end namespace OPENSUBDIV_VERSION\n- 98using namespace OPENSUBDIV_VERSION;\n- 99\n- 100} // end namespace OpenSubdiv\n- 101\n- 102#endif // OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H\n+ 97} // end namespace OpenSubdiv\n+ 98\n+ 99#endif // OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n-Type\n-Definition: patchDescriptor.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_BOUNDARY\n-@ GREGORY_BOUNDARY\n-Definition: patchDescriptor.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n+std::vector< PatchArray > PatchArrayVector\n+Definition: types.h:132\n OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n Container for arrays of parametric patches.\n Definition: patchTable.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable\n-Definition: d3d11LegacyGregoryPatchTable.h:44\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n-UpdateVertexBuffer\n-void UpdateVertexBuffer(ID3D11Buffer *vbo, int numVertices, int\n-numVertexElements, ID3D11DeviceContext *pd3d11DeviceContext)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n-GetQuadOffsetsBase\n-int GetQuadOffsetsBase(Far::PatchDescriptor::Type type)\n-Definition: d3d11LegacyGregoryPatchTable.h:74\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n-~D3D11LegacyGregoryPatchTable\n-~D3D11LegacyGregoryPatchTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::Create\n-static D3D11LegacyGregoryPatchTable * Create(Far::PatchTable const\n-*farPatchTable, ID3D11DeviceContext *deviceContext)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::GetVertexSRV\n-ID3D11ShaderResourceView * GetVertexSRV() const\n-Definition: d3d11LegacyGregoryPatchTable.h:62\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::Create\n-static D3D11LegacyGregoryPatchTable * Create(Far::PatchTable const\n-*farPatchTable, DEVICE_CONTEXT context)\n-Definition: d3d11LegacyGregoryPatchTable.h:49\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n-GetVertexValenceSRV\n-ID3D11ShaderResourceView * GetVertexValenceSRV() const\n-Definition: d3d11LegacyGregoryPatchTable.h:66\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n-GetQuadOffsetsSRV\n-ID3D11ShaderResourceView * GetQuadOffsetsSRV() const\n-Definition: d3d11LegacyGregoryPatchTable.h:70\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n-D3D11LegacyGregoryPatchTable\n-D3D11LegacyGregoryPatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable\n+Definition: d3d11PatchTable.h:49\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::Create\n+static D3D11PatchTable * Create(Far::PatchTable const *farPatchTable,\n+DEVICE_CONTEXT context)\n+Definition: d3d11PatchTable.h:57\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchArrays\n+PatchArrayVector const & GetPatchArrays() const\n+Definition: d3d11PatchTable.h:65\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::allocate\n+bool allocate(Far::PatchTable const *farPatchTable, ID3D11DeviceContext\n+*deviceContext)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::VertexBufferBinding\n+ID3D11Buffer * VertexBufferBinding\n+Definition: d3d11PatchTable.h:51\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::~D3D11PatchTable\n+~D3D11PatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchIndexBuffer\n+ID3D11Buffer * GetPatchIndexBuffer() const\n+Returns the index buffer containing the patch control vertices.\n+Definition: d3d11PatchTable.h:70\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::_patchArrays\n+PatchArrayVector _patchArrays\n+Definition: d3d11PatchTable.h:84\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::_indexBuffer\n+ID3D11Buffer * _indexBuffer\n+Definition: d3d11PatchTable.h:86\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::Create\n+static D3D11PatchTable * Create(Far::PatchTable const *farPatchTable,\n+ID3D11DeviceContext *deviceContext)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::_patchParamBufferSRV\n+ID3D11ShaderResourceView * _patchParamBufferSRV\n+Definition: d3d11PatchTable.h:88\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::_patchParamBuffer\n+ID3D11Buffer * _patchParamBuffer\n+Definition: d3d11PatchTable.h:87\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchParamSRV\n+ID3D11ShaderResourceView * GetPatchParamSRV() const\n+Returns the SRV containing the patch parameter.\n+Definition: d3d11PatchTable.h:75\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::D3D11PatchTable\n+D3D11PatchTable()\n * opensubdiv\n * osd\n- * d3d11LegacyGregoryPatchTable.h\n+ * d3d11PatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00839.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00839.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/d3d11ComputeEvaluator.h File Reference\n+OpenSubdiv: opensubdiv/osd/d3d11LegacyGregoryPatchTable.h File Reference\n \n \n \n \n \n \n \n@@ -83,45 +83,41 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
d3d11ComputeEvaluator.h File Reference
\n+
d3d11LegacyGregoryPatchTable.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../osd/bufferDescriptor.h"
\n+#include "../far/patchTable.h"
\n+#include "../osd/nonCopyable.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n-\n-\n+\n \n

\n Classes

class  D3D11StencilTable
 D3D11 stencil table. More...
 
class  D3D11ComputeEvaluator
class  D3D11LegacyGregoryPatchTable
 
\n \n \n \n \n \n-\n-\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,30 +5,26 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-d3d11ComputeEvaluator.h File Reference\n+d3d11LegacyGregoryPatchTable.h File Reference\n #include \"../version.h\"\n-#include \"../osd/bufferDescriptor.h\"\n+#include \"../far/patchTable.h\"\n+#include \"../osd/nonCopyable.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0D3D11StencilTable\n-\u00a0 D3D11 stencil table. More...\n-\u00a0\n-class \u00a0D3D11ComputeEvaluator\n+class \u00a0D3D11LegacyGregoryPatchTable\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n-\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * d3d11ComputeEvaluator.h\n+ * d3d11LegacyGregoryPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00839.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00839.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,3 @@\n var a00839 = [\n- [\"D3D11StencilTable\", \"a01189.html\", \"a01189\"],\n- [\"D3D11ComputeEvaluator\", \"a01193.html\", \"a01193\"]\n+ [\"D3D11LegacyGregoryPatchTable\", \"a01197.html\", \"a01197\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00839_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00839_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/d3d11ComputeEvaluator.h Source File\n+OpenSubdiv: opensubdiv/osd/d3d11LegacyGregoryPatchTable.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
d3d11ComputeEvaluator.h
\n+
d3d11LegacyGregoryPatchTable.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,213 +107,110 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H
\n-
26#define OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H
\n+
25#ifndef OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H
\n+
26#define OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30struct ID3D11DeviceContext;
\n-
31struct ID3D11Buffer;
\n-
32struct ID3D11ComputeShader;
\n-
33struct ID3D11ClassLinkage;
\n-
34struct ID3D11ClassInstance;
\n-
35struct ID3D11ShaderResourceView;
\n-
36struct ID3D11UnorderedAccessView;
\n+
30#include "../far/patchTable.h"
\n+
31#include "../osd/nonCopyable.h"
\n+
32
\n+
33struct ID3D11Buffer;
\n+
34struct ID3D11ShaderResourceView;
\n+
35struct ID3D11Device;
\n+
36struct ID3D11DeviceContext;
\n
37
\n-
38#include "../osd/bufferDescriptor.h"
\n-
39
\n-
40namespace OpenSubdiv {
\n-
41namespace OPENSUBDIV_VERSION {
\n+
38namespace OpenSubdiv {
\n+
39namespace OPENSUBDIV_VERSION {
\n+
40
\n+
41namespace Osd {
\n
42
\n-
43namespace Far {
\n-
44 class StencilTable;
\n-
45}
\n-
46
\n-
47namespace Osd {
\n-
48
\n-
56class D3D11StencilTable {
\n-
57public:
\n-
58 template <typename DEVICE_CONTEXT>
\n-
59 static D3D11StencilTable *Create(Far::StencilTable const *stencilTable,
\n-
60 DEVICE_CONTEXT context) {
\n-
61 return new D3D11StencilTable(stencilTable, context->GetDeviceContext());
\n-
62 }
\n-
63
\n-
64 static D3D11StencilTable *Create(Far::StencilTable const *stencilTable,
\n-
65 ID3D11DeviceContext *deviceContext) {
\n-
66 return new D3D11StencilTable(stencilTable, deviceContext);
\n-
67 }
\n-
68
\n-
69 D3D11StencilTable(Far::StencilTable const *stencilTable,
\n-
70 ID3D11DeviceContext *deviceContext);
\n-
71
\n-
72 ~D3D11StencilTable();
\n+
43class D3D11LegacyGregoryPatchTable
\n+
44 : private NonCopyable<D3D11LegacyGregoryPatchTable> {
\n+
45public:
\n+
46 ~D3D11LegacyGregoryPatchTable();
\n+
47
\n+
48 template<typename DEVICE_CONTEXT>
\n+
49 static D3D11LegacyGregoryPatchTable *Create(
\n+
50 Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context) {
\n+
51 return Create(farPatchTable, context->GetDeviceContext());
\n+
52 }
\n+
53
\n+
54 static D3D11LegacyGregoryPatchTable *Create(
\n+
55 Far::PatchTable const *farPatchTable,
\n+
56 ID3D11DeviceContext *deviceContext);
\n+
57
\n+
58 void UpdateVertexBuffer(ID3D11Buffer *vbo,
\n+
59 int numVertices, int numVertexElements,
\n+
60 ID3D11DeviceContext *pd3d11DeviceContext);
\n+
61
\n+
62 ID3D11ShaderResourceView* GetVertexSRV() const {
\n+
63 return _vertexSRV;
\n+
64 }
\n+
65
\n+
66 ID3D11ShaderResourceView* GetVertexValenceSRV() const {
\n+
67 return _vertexValenceSRV;
\n+
68 }
\n+
69
\n+
70 ID3D11ShaderResourceView* GetQuadOffsetsSRV() const {
\n+
71 return _quadOffsetsSRV;
\n+
72 }
\n
73
\n-
74 // interfaces needed for D3D11ComputeEvaluator
\n-
75 ID3D11ShaderResourceView *GetSizesSRV() const { return _sizes; }
\n-
76 ID3D11ShaderResourceView *GetOffsetsSRV() const { return _offsets; }
\n-
77 ID3D11ShaderResourceView *GetIndicesSRV() const { return _indices; }
\n-
78 ID3D11ShaderResourceView *GetWeightsSRV() const { return _weights; }
\n-
79 int GetNumStencils() const { return _numStencils; }
\n+
74 int GetQuadOffsetsBase(Far::PatchDescriptor::Type type) {
\n+
75 if (type == Far::PatchDescriptor::GREGORY_BOUNDARY) {
\n+
76 return _quadOffsetsBase[1];
\n+
77 }
\n+
78 return _quadOffsetsBase[0];
\n+
79 }
\n
80
\n-
81private:
\n-
82 ID3D11ShaderResourceView *_sizes;
\n-
83 ID3D11ShaderResourceView *_offsets;
\n-
84 ID3D11ShaderResourceView *_indices;
\n-
85 ID3D11ShaderResourceView *_weights;
\n-
86 ID3D11Buffer *_sizesBuffer;
\n-
87 ID3D11Buffer *_offsetsBuffer;
\n-
88 ID3D11Buffer *_indicesBuffer;
\n-
89 ID3D11Buffer *_weightsBuffer;
\n-
90
\n-
91 int _numStencils;
\n-
92};
\n+
81protected:
\n+
82 D3D11LegacyGregoryPatchTable();
\n+
83
\n+
84private:
\n+
85 ID3D11Buffer* _vertexValenceBuffer;
\n+
86 ID3D11Buffer* _quadOffsetsBuffer;
\n+
87 ID3D11ShaderResourceView* _vertexSRV;
\n+
88 ID3D11ShaderResourceView* _vertexValenceSRV;
\n+
89 ID3D11ShaderResourceView* _quadOffsetsSRV;
\n+
90 int _quadOffsetsBase[2]; // gregory, boundaryGregory
\n+
91};
\n+
92
\n
93
\n-
94// ---------------------------------------------------------------------------
\n-
95
\n-
96class D3D11ComputeEvaluator {
\n-
97public:
\n-
98 typedef bool Instantiatable;
\n-
99 static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n-
100 BufferDescriptor const &dstDesc,
\n-
101 BufferDescriptor const &duDesc,
\n-
102 BufferDescriptor const &dvDesc,
\n-
103 ID3D11DeviceContext *deviceContext);
\n-
104
\n-
105 static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n-
106 BufferDescriptor const &dstDesc,
\n-
107 BufferDescriptor const &duDesc,
\n-
108 BufferDescriptor const &dvDesc,
\n-
109 BufferDescriptor const &duuDesc,
\n-
110 BufferDescriptor const &duvDesc,
\n-
111 BufferDescriptor const &dvvDesc,
\n-
112 ID3D11DeviceContext *deviceContext);
\n-
113
\n-
115 D3D11ComputeEvaluator();
\n-
116
\n-
118 ~D3D11ComputeEvaluator();
\n-
119
\n-
147 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
148 static bool EvalStencils(
\n-
149 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
150 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
151 STENCIL_TABLE const *stencilTable,
\n-
152 D3D11ComputeEvaluator const *instance,
\n-
153 ID3D11DeviceContext * deviceContext) {
\n-
154 if (instance) {
\n-
155 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
156 dstBuffer, dstDesc,
\n-
157 stencilTable,
\n-
158 deviceContext);
\n-
159 } else {
\n-
160 // Create an instance on demand (slow)
\n-
161 (void)deviceContext; // unused
\n-
162 instance = Create(srcDesc, dstDesc,
\n-
163 BufferDescriptor(),
\n-
164 BufferDescriptor(),
\n-
165 deviceContext);
\n-
166 if (instance) {
\n-
167 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
168 dstBuffer, dstDesc,
\n-
169 stencilTable,
\n-
170 deviceContext);
\n-
171 delete instance;
\n-
172 return r;
\n-
173 }
\n-
174 return false;
\n-
175 }
\n-
176 }
\n-
177
\n-
180 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
181 bool EvalStencils(
\n-
182 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
183 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
184 STENCIL_TABLE const *stencilTable,
\n-
185 ID3D11DeviceContext *deviceContext) const {
\n-
186 return EvalStencils(srcBuffer->BindD3D11UAV(deviceContext), srcDesc,
\n-
187 dstBuffer->BindD3D11UAV(deviceContext), dstDesc,
\n-
188 stencilTable->GetSizesSRV(),
\n-
189 stencilTable->GetOffsetsSRV(),
\n-
190 stencilTable->GetIndicesSRV(),
\n-
191 stencilTable->GetWeightsSRV(),
\n-
192 /* start = */ 0,
\n-
193 /* end = */ stencilTable->GetNumStencils(),
\n-
194 deviceContext);
\n-
195 }
\n-
196
\n-
199 bool EvalStencils(ID3D11UnorderedAccessView *srcSRV,
\n-
200 BufferDescriptor const &srcDesc,
\n-
201 ID3D11UnorderedAccessView *dstUAV,
\n-
202 BufferDescriptor const &dstDesc,
\n-
203 ID3D11ShaderResourceView *sizesSRV,
\n-
204 ID3D11ShaderResourceView *offsetsSRV,
\n-
205 ID3D11ShaderResourceView *indicesSRV,
\n-
206 ID3D11ShaderResourceView *weightsSRV,
\n-
207 int start,
\n-
208 int end,
\n-
209 ID3D11DeviceContext *deviceContext) const;
\n-
210
\n-
212 bool Compile(BufferDescriptor const &srcDesc,
\n-
213 BufferDescriptor const &dstDesc,
\n-
214 ID3D11DeviceContext *deviceContext);
\n-
215
\n-
217 static void Synchronize(ID3D11DeviceContext *deviceContext);
\n-
218
\n-
219private:
\n-
220 ID3D11ComputeShader * _computeShader;
\n-
221 ID3D11ClassLinkage * _classLinkage;
\n-
222 ID3D11ClassInstance * _singleBufferKernel;
\n-
223 ID3D11ClassInstance * _separateBufferKernel;
\n-
224 ID3D11Buffer * _uniformArgs; // uniform parameters for kernels
\n-
225
\n-
226 int _workGroupSize;
\n-
227};
\n-
228
\n-
229} // end namespace Osd
\n-
230
\n-
231} // end namespace OPENSUBDIV_VERSION
\n-
232using namespace OPENSUBDIV_VERSION;
\n-
233
\n-
234} // end namespace OpenSubdiv
\n-
235
\n-
236
\n-
237#endif // OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H
\n+
94
\n+
95} // end namespace Osd
\n+
96
\n+
97} // end namespace OPENSUBDIV_VERSION
\n+
98using namespace OPENSUBDIV_VERSION;
\n+
99
\n+
100} // end namespace OpenSubdiv
\n+
101
\n+
102#endif // OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::Create
static D3D11StencilTable * Create(Far::StencilTable const *stencilTable, DEVICE_CONTEXT context)
Definition: d3d11ComputeEvaluator.h:59
\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::D3D11StencilTable
D3D11StencilTable(Far::StencilTable const *stencilTable, ID3D11DeviceContext *deviceContext)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::Create
static D3D11StencilTable * Create(Far::StencilTable const *stencilTable, ID3D11DeviceContext *deviceContext)
Definition: d3d11ComputeEvaluator.h:64
\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Create
static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, ID3D11DeviceContext *deviceContext)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, D3D11ComputeEvaluator const *instance, ID3D11DeviceContext *deviceContext)
Generic static compute function. This function has a same signature as other device kernels have so t...
Definition: d3d11ComputeEvaluator.h:148
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Create
static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, ID3D11DeviceContext *deviceContext)
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Synchronize
static void Synchronize(ID3D11DeviceContext *deviceContext)
Wait the dispatched kernel finishes.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, ID3D11DeviceContext *deviceContext) const
Definition: d3d11ComputeEvaluator.h:181
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Compile
bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, ID3D11DeviceContext *deviceContext)
Configure DX kernel. Returns false if it fails to compile the kernel.
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils
bool EvalStencils(ID3D11UnorderedAccessView *srcSRV, BufferDescriptor const &srcDesc, ID3D11UnorderedAccessView *dstUAV, BufferDescriptor const &dstDesc, ID3D11ShaderResourceView *sizesSRV, ID3D11ShaderResourceView *offsetsSRV, ID3D11ShaderResourceView *indicesSRV, ID3D11ShaderResourceView *weightsSRV, int start, int end, ID3D11DeviceContext *deviceContext) const
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::UpdateVertexBuffer
void UpdateVertexBuffer(ID3D11Buffer *vbo, int numVertices, int numVertexElements, ID3D11DeviceContext *pd3d11DeviceContext)
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::Create
static D3D11LegacyGregoryPatchTable * Create(Far::PatchTable const *farPatchTable, ID3D11DeviceContext *deviceContext)
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::Create
static D3D11LegacyGregoryPatchTable * Create(Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context)
Definition: d3d11LegacyGregoryPatchTable.h:49
\n+\n+\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-d3d11ComputeEvaluator.h\n+d3d11LegacyGregoryPatchTable.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,274 +30,134 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H\n- 26#define OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H\n+ 25#ifndef OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H\n+ 26#define OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H\n 27\n 28#include \"../version.h\"\n 29\n- 30struct ID3D11DeviceContext;\n- 31struct ID3D11Buffer;\n- 32struct ID3D11ComputeShader;\n- 33struct ID3D11ClassLinkage;\n- 34struct ID3D11ClassInstance;\n- 35struct ID3D11ShaderResourceView;\n- 36struct ID3D11UnorderedAccessView;\n+ 30#include \"../far/patchTable.h\"\n+ 31#include \"../osd/nonCopyable.h\"\n+ 32\n+ 33struct ID3D11Buffer;\n+ 34struct ID3D11ShaderResourceView;\n+ 35struct ID3D11Device;\n+ 36struct ID3D11DeviceContext;\n 37\n- 38#include \"../osd/bufferDescriptor.h\"\n- 39\n- 40namespace OpenSubdiv {\n- 41namespace OPENSUBDIV_VERSION {\n+ 38namespace OpenSubdiv {\n+ 39namespace OPENSUBDIV_VERSION {\n+ 40\n+ 41namespace Osd {\n 42\n- 43namespace Far {\n- 44 class StencilTable;\n- 45}\n- 46\n- 47namespace Osd {\n- 48\n-56class D3D11StencilTable {\n- 57public:\n- 58 template \n-59 static D3D11StencilTable *Create(Far::StencilTable const *stencilTable,\n- 60 DEVICE_CONTEXT context) {\n- 61 return new D3D11StencilTable(stencilTable, context->GetDeviceContext());\n- 62 }\n- 63\n-64 static D3D11StencilTable *Create(Far::StencilTable const *stencilTable,\n- 65 ID3D11DeviceContext *deviceContext) {\n- 66 return new D3D11StencilTable(stencilTable, deviceContext);\n- 67 }\n- 68\n-69 D3D11StencilTable(Far::StencilTable const *stencilTable,\n- 70 ID3D11DeviceContext *deviceContext);\n- 71\n-72 ~D3D11StencilTable();\n+43class D3D11LegacyGregoryPatchTable\n+ 44 : private NonCopyable {\n+ 45public:\n+46 ~D3D11LegacyGregoryPatchTable();\n+ 47\n+ 48 template\n+49 static D3D11LegacyGregoryPatchTable *Create(\n+ 50 Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context) {\n+ 51 return Create(farPatchTable, context->GetDeviceContext());\n+ 52 }\n+ 53\n+54 static D3D11LegacyGregoryPatchTable *Create(\n+ 55 Far::PatchTable const *farPatchTable,\n+ 56 ID3D11DeviceContext *deviceContext);\n+ 57\n+58 void UpdateVertexBuffer(ID3D11Buffer *vbo,\n+ 59 int numVertices, int numVertexElements,\n+ 60 ID3D11DeviceContext *pd3d11DeviceContext);\n+ 61\n+62 ID3D11ShaderResourceView* GetVertexSRV() const {\n+ 63 return _vertexSRV;\n+ 64 }\n+ 65\n+66 ID3D11ShaderResourceView* GetVertexValenceSRV() const {\n+ 67 return _vertexValenceSRV;\n+ 68 }\n+ 69\n+70 ID3D11ShaderResourceView* GetQuadOffsetsSRV() const {\n+ 71 return _quadOffsetsSRV;\n+ 72 }\n 73\n- 74 // interfaces needed for D3D11ComputeEvaluator\n-75 ID3D11ShaderResourceView *GetSizesSRV() const { return _sizes; }\n-76 ID3D11ShaderResourceView *GetOffsetsSRV() const { return _offsets; }\n-77 ID3D11ShaderResourceView *GetIndicesSRV() const { return _indices; }\n-78 ID3D11ShaderResourceView *GetWeightsSRV() const { return _weights; }\n-79 int GetNumStencils() const { return _numStencils; }\n+74 int GetQuadOffsetsBase(Far::PatchDescriptor::Type type) {\n+ 75 if (type == Far::PatchDescriptor::GREGORY_BOUNDARY) {\n+ 76 return _quadOffsetsBase[1];\n+ 77 }\n+ 78 return _quadOffsetsBase[0];\n+ 79 }\n 80\n- 81private:\n- 82 ID3D11ShaderResourceView *_sizes;\n- 83 ID3D11ShaderResourceView *_offsets;\n- 84 ID3D11ShaderResourceView *_indices;\n- 85 ID3D11ShaderResourceView *_weights;\n- 86 ID3D11Buffer *_sizesBuffer;\n- 87 ID3D11Buffer *_offsetsBuffer;\n- 88 ID3D11Buffer *_indicesBuffer;\n- 89 ID3D11Buffer *_weightsBuffer;\n- 90\n- 91 int _numStencils;\n- 92};\n+ 81protected:\n+82 D3D11LegacyGregoryPatchTable();\n+ 83\n+ 84private:\n+ 85 ID3D11Buffer* _vertexValenceBuffer;\n+ 86 ID3D11Buffer* _quadOffsetsBuffer;\n+ 87 ID3D11ShaderResourceView* _vertexSRV;\n+ 88 ID3D11ShaderResourceView* _vertexValenceSRV;\n+ 89 ID3D11ShaderResourceView* _quadOffsetsSRV;\n+ 90 int _quadOffsetsBase[2]; // gregory, boundaryGregory\n+ 91};\n+ 92\n 93\n- 94// -------------------------------------------------------------------------\n---\n- 95\n-96class D3D11ComputeEvaluator {\n- 97public:\n-98 typedef bool Instantiatable;\n-99 static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n- 100 BufferDescriptor const &dstDesc,\n- 101 BufferDescriptor const &duDesc,\n- 102 BufferDescriptor const &dvDesc,\n- 103 ID3D11DeviceContext *deviceContext);\n- 104\n-105 static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n- 106 BufferDescriptor const &dstDesc,\n- 107 BufferDescriptor const &duDesc,\n- 108 BufferDescriptor const &dvDesc,\n- 109 BufferDescriptor const &duuDesc,\n- 110 BufferDescriptor const &duvDesc,\n- 111 BufferDescriptor const &dvvDesc,\n- 112 ID3D11DeviceContext *deviceContext);\n- 113\n-115 D3D11ComputeEvaluator();\n- 116\n-118 ~D3D11ComputeEvaluator();\n- 119\n- 147 template \n-148 static bool EvalStencils(\n- 149 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 150 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 151 STENCIL_TABLE const *stencilTable,\n- 152 D3D11ComputeEvaluator const *instance,\n- 153 ID3D11DeviceContext * deviceContext) {\n- 154 if (instance) {\n- 155 return instance->EvalStencils(srcBuffer, srcDesc,\n- 156 dstBuffer, dstDesc,\n- 157 stencilTable,\n- 158 deviceContext);\n- 159 } else {\n- 160 // Create an instance on demand (slow)\n- 161 (void)deviceContext; // unused\n- 162 instance = Create(srcDesc, dstDesc,\n- 163 BufferDescriptor(),\n- 164 BufferDescriptor(),\n- 165 deviceContext);\n- 166 if (instance) {\n- 167 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 168 dstBuffer, dstDesc,\n- 169 stencilTable,\n- 170 deviceContext);\n- 171 delete instance;\n- 172 return r;\n- 173 }\n- 174 return false;\n- 175 }\n- 176 }\n- 177\n- 180 template \n-181 bool EvalStencils(\n- 182 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 183 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 184 STENCIL_TABLE const *stencilTable,\n- 185 ID3D11DeviceContext *deviceContext) const {\n- 186 return EvalStencils(srcBuffer->BindD3D11UAV(deviceContext), srcDesc,\n- 187 dstBuffer->BindD3D11UAV(deviceContext), dstDesc,\n- 188 stencilTable->GetSizesSRV(),\n- 189 stencilTable->GetOffsetsSRV(),\n- 190 stencilTable->GetIndicesSRV(),\n- 191 stencilTable->GetWeightsSRV(),\n- 192 /* start = */ 0,\n- 193 /* end = */ stencilTable->GetNumStencils(),\n- 194 deviceContext);\n- 195 }\n- 196\n-199 bool EvalStencils(ID3D11UnorderedAccessView *srcSRV,\n- 200 BufferDescriptor const &srcDesc,\n- 201 ID3D11UnorderedAccessView *dstUAV,\n- 202 BufferDescriptor const &dstDesc,\n- 203 ID3D11ShaderResourceView *sizesSRV,\n- 204 ID3D11ShaderResourceView *offsetsSRV,\n- 205 ID3D11ShaderResourceView *indicesSRV,\n- 206 ID3D11ShaderResourceView *weightsSRV,\n- 207 int start,\n- 208 int end,\n- 209 ID3D11DeviceContext *deviceContext) const;\n- 210\n-212 bool Compile(BufferDescriptor const &srcDesc,\n- 213 BufferDescriptor const &dstDesc,\n- 214 ID3D11DeviceContext *deviceContext);\n- 215\n-217 static void Synchronize(ID3D11DeviceContext *deviceContext);\n- 218\n- 219private:\n- 220 ID3D11ComputeShader * _computeShader;\n- 221 ID3D11ClassLinkage * _classLinkage;\n- 222 ID3D11ClassInstance * _singleBufferKernel;\n- 223 ID3D11ClassInstance * _separateBufferKernel;\n- 224 ID3D11Buffer * _uniformArgs; // uniform parameters for kernels\n- 225\n- 226 int _workGroupSize;\n- 227};\n- 228\n- 229} // end namespace Osd\n- 230\n- 231} // end namespace OPENSUBDIV_VERSION\n- 232using namespace OPENSUBDIV_VERSION;\n- 233\n- 234} // end namespace OpenSubdiv\n- 235\n- 236\n- 237#endif // OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H\n+ 94\n+ 95} // end namespace Osd\n+ 96\n+ 97} // end namespace OPENSUBDIV_VERSION\n+ 98using namespace OPENSUBDIV_VERSION;\n+ 99\n+ 100} // end namespace OpenSubdiv\n+ 101\n+ 102#endif // OPENSUBDIV3_OSD_D3D11_LEGACY_GREGORY_PATCH_TABLE_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n-Stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:273\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n-BufferDescriptor is a struct which describes buffer elements in interleaved\n-data buffers....\n-Definition: bufferDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable\n-D3D11 stencil table.\n-Definition: d3d11ComputeEvaluator.h:56\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::Create\n-static D3D11StencilTable * Create(Far::StencilTable const *stencilTable,\n-DEVICE_CONTEXT context)\n-Definition: d3d11ComputeEvaluator.h:59\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::GetOffsetsSRV\n-ID3D11ShaderResourceView * GetOffsetsSRV() const\n-Definition: d3d11ComputeEvaluator.h:76\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::GetNumStencils\n-int GetNumStencils() const\n-Definition: d3d11ComputeEvaluator.h:79\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::GetIndicesSRV\n-ID3D11ShaderResourceView * GetIndicesSRV() const\n-Definition: d3d11ComputeEvaluator.h:77\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::~D3D11StencilTable\n-~D3D11StencilTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::D3D11StencilTable\n-D3D11StencilTable(Far::StencilTable const *stencilTable, ID3D11DeviceContext\n-*deviceContext)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::Create\n-static D3D11StencilTable * Create(Far::StencilTable const *stencilTable,\n-ID3D11DeviceContext *deviceContext)\n-Definition: d3d11ComputeEvaluator.h:64\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::GetSizesSRV\n-ID3D11ShaderResourceView * GetSizesSRV() const\n-Definition: d3d11ComputeEvaluator.h:75\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::GetWeightsSRV\n-ID3D11ShaderResourceView * GetWeightsSRV() const\n-Definition: d3d11ComputeEvaluator.h:78\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator\n-Definition: d3d11ComputeEvaluator.h:96\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Create\n-static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, ID3D11DeviceContext *deviceContext)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n-const *stencilTable, D3D11ComputeEvaluator const *instance, ID3D11DeviceContext\n-*deviceContext)\n-Generic static compute function. This function has a same signature as other\n-device kernels have so t...\n-Definition: d3d11ComputeEvaluator.h:148\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Create\n-static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n-BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc,\n-ID3D11DeviceContext *deviceContext)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Instantiatable\n-bool Instantiatable\n-Definition: d3d11ComputeEvaluator.h:98\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Synchronize\n-static void Synchronize(ID3D11DeviceContext *deviceContext)\n-Wait the dispatched kernel finishes.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils\n-bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const\n-*stencilTable, ID3D11DeviceContext *deviceContext) const\n-Definition: d3d11ComputeEvaluator.h:181\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Compile\n-bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc,\n-ID3D11DeviceContext *deviceContext)\n-Configure DX kernel. Returns false if it fails to compile the kernel.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::\n-D3D11ComputeEvaluator\n-D3D11ComputeEvaluator()\n-Constructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::\n-~D3D11ComputeEvaluator\n-~D3D11ComputeEvaluator()\n-Destructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils\n-bool EvalStencils(ID3D11UnorderedAccessView *srcSRV, BufferDescriptor const\n-&srcDesc, ID3D11UnorderedAccessView *dstUAV, BufferDescriptor const &dstDesc,\n-ID3D11ShaderResourceView *sizesSRV, ID3D11ShaderResourceView *offsetsSRV,\n-ID3D11ShaderResourceView *indicesSRV, ID3D11ShaderResourceView *weightsSRV, int\n-start, int end, ID3D11DeviceContext *deviceContext) const\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n+Type\n+Definition: patchDescriptor.h:48\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_BOUNDARY\n+@ GREGORY_BOUNDARY\n+Definition: patchDescriptor.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n+Container for arrays of parametric patches.\n+Definition: patchTable.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable\n+Definition: d3d11LegacyGregoryPatchTable.h:44\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n+UpdateVertexBuffer\n+void UpdateVertexBuffer(ID3D11Buffer *vbo, int numVertices, int\n+numVertexElements, ID3D11DeviceContext *pd3d11DeviceContext)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n+GetQuadOffsetsBase\n+int GetQuadOffsetsBase(Far::PatchDescriptor::Type type)\n+Definition: d3d11LegacyGregoryPatchTable.h:74\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n+~D3D11LegacyGregoryPatchTable\n+~D3D11LegacyGregoryPatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::Create\n+static D3D11LegacyGregoryPatchTable * Create(Far::PatchTable const\n+*farPatchTable, ID3D11DeviceContext *deviceContext)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::GetVertexSRV\n+ID3D11ShaderResourceView * GetVertexSRV() const\n+Definition: d3d11LegacyGregoryPatchTable.h:62\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::Create\n+static D3D11LegacyGregoryPatchTable * Create(Far::PatchTable const\n+*farPatchTable, DEVICE_CONTEXT context)\n+Definition: d3d11LegacyGregoryPatchTable.h:49\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n+GetVertexValenceSRV\n+ID3D11ShaderResourceView * GetVertexValenceSRV() const\n+Definition: d3d11LegacyGregoryPatchTable.h:66\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n+GetQuadOffsetsSRV\n+ID3D11ShaderResourceView * GetQuadOffsetsSRV() const\n+Definition: d3d11LegacyGregoryPatchTable.h:70\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::\n+D3D11LegacyGregoryPatchTable\n+D3D11LegacyGregoryPatchTable()\n * opensubdiv\n * osd\n- * d3d11ComputeEvaluator.h\n+ * d3d11LegacyGregoryPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00842.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00842.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cpuD3D11VertexBuffer.h File Reference\n+OpenSubdiv: opensubdiv/osd/d3d11ComputeEvaluator.h File Reference\n \n \n \n \n \n \n \n@@ -83,41 +83,45 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
cpuD3D11VertexBuffer.h File Reference
\n+
d3d11ComputeEvaluator.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include <cstddef>
\n+#include "../osd/bufferDescriptor.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n+\n+\n+\n+\n \n

\n Classes

class  CpuD3D11VertexBuffer
 Concrete vertex buffer class for Cpu subdivision and DirectX drawing. More...
class  D3D11StencilTable
 D3D11 stencil table. More...
 
class  D3D11ComputeEvaluator
 
\n \n \n \n \n \n+\n+\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,27 +5,30 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-cpuD3D11VertexBuffer.h File Reference\n+d3d11ComputeEvaluator.h File Reference\n #include \"../version.h\"\n-#include \n+#include \"../osd/bufferDescriptor.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0CpuD3D11VertexBuffer\n-\u00a0 Concrete vertex buffer class for Cpu subdivision and DirectX drawing.\n- More...\n+class \u00a0D3D11StencilTable\n+\u00a0 D3D11 stencil table. More...\n+\u00a0\n+class \u00a0D3D11ComputeEvaluator\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n+\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * cpuD3D11VertexBuffer.h\n+ * d3d11ComputeEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00842.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00842.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,4 @@\n var a00842 = [\n- [\"CpuD3D11VertexBuffer\", \"a01153.html\", \"a01153\"]\n+ [\"D3D11StencilTable\", \"a01189.html\", \"a01189\"],\n+ [\"D3D11ComputeEvaluator\", \"a01193.html\", \"a01193\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00842_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00842_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cpuD3D11VertexBuffer.h Source File\n+OpenSubdiv: opensubdiv/osd/d3d11ComputeEvaluator.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
cpuD3D11VertexBuffer.h
\n+
d3d11ComputeEvaluator.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2013 Pixar
\n+
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,87 +107,213 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H
\n-
26#define OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H
\n+
25#ifndef OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H
\n+
26#define OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H
\n
27
\n
28#include "../version.h"
\n-
29#include <cstddef>
\n-
30
\n+
29
\n+
30struct ID3D11DeviceContext;
\n
31struct ID3D11Buffer;
\n-
32struct ID3D11Device;
\n-
33struct ID3D11DeviceContext;
\n-
34
\n-
35namespace OpenSubdiv {
\n-
36namespace OPENSUBDIV_VERSION {
\n+
32struct ID3D11ComputeShader;
\n+
33struct ID3D11ClassLinkage;
\n+
34struct ID3D11ClassInstance;
\n+
35struct ID3D11ShaderResourceView;
\n+
36struct ID3D11UnorderedAccessView;
\n
37
\n-
38namespace Osd {
\n+
38#include "../osd/bufferDescriptor.h"
\n
39
\n-
48class CpuD3D11VertexBuffer {
\n-
49public:
\n-
51 static CpuD3D11VertexBuffer * Create(int numElements, int numVertices,
\n-
52 ID3D11DeviceContext *deviceContext);
\n-
53
\n-
55 virtual ~CpuD3D11VertexBuffer();
\n-
56
\n-
59 void UpdateData(const float *src, int startVertex, int numVertices,
\n-
60 void *deviceContext = NULL);
\n-
61
\n-
63 int GetNumElements() const;
\n-
64
\n-
66 int GetNumVertices() const;
\n-
67
\n-
69 float * BindCpuBuffer();
\n-
70
\n-
72 ID3D11Buffer *BindD3D11Buffer(ID3D11DeviceContext *deviceContext);
\n+
40namespace OpenSubdiv {
\n+
41namespace OPENSUBDIV_VERSION {
\n+
42
\n+
43namespace Far {
\n+
44 class StencilTable;
\n+
45}
\n+
46
\n+
47namespace Osd {
\n+
48
\n+
56class D3D11StencilTable {
\n+
57public:
\n+
58 template <typename DEVICE_CONTEXT>
\n+
59 static D3D11StencilTable *Create(Far::StencilTable const *stencilTable,
\n+
60 DEVICE_CONTEXT context) {
\n+
61 return new D3D11StencilTable(stencilTable, context->GetDeviceContext());
\n+
62 }
\n+
63
\n+
64 static D3D11StencilTable *Create(Far::StencilTable const *stencilTable,
\n+
65 ID3D11DeviceContext *deviceContext) {
\n+
66 return new D3D11StencilTable(stencilTable, deviceContext);
\n+
67 }
\n+
68
\n+
69 D3D11StencilTable(Far::StencilTable const *stencilTable,
\n+
70 ID3D11DeviceContext *deviceContext);
\n+
71
\n+
72 ~D3D11StencilTable();
\n
73
\n-
75 ID3D11Buffer *BindVBO(ID3D11DeviceContext *deviceContext) {
\n-
76 return BindD3D11Buffer(deviceContext);
\n-
77 }
\n-
78
\n-
79protected:
\n-
81 CpuD3D11VertexBuffer(int numElements, int numVertices);
\n-
82
\n-
83 bool allocate(ID3D11Device *device);
\n-
84
\n-
85private:
\n-
86 int _numElements;
\n-
87 int _numVertices;
\n-
88 ID3D11Buffer *_d3d11Buffer;
\n-
89 float *_cpuBuffer;
\n-
90};
\n-
91
\n-
92} // end namespace Osd
\n+
74 // interfaces needed for D3D11ComputeEvaluator
\n+
75 ID3D11ShaderResourceView *GetSizesSRV() const { return _sizes; }
\n+
76 ID3D11ShaderResourceView *GetOffsetsSRV() const { return _offsets; }
\n+
77 ID3D11ShaderResourceView *GetIndicesSRV() const { return _indices; }
\n+
78 ID3D11ShaderResourceView *GetWeightsSRV() const { return _weights; }
\n+
79 int GetNumStencils() const { return _numStencils; }
\n+
80
\n+
81private:
\n+
82 ID3D11ShaderResourceView *_sizes;
\n+
83 ID3D11ShaderResourceView *_offsets;
\n+
84 ID3D11ShaderResourceView *_indices;
\n+
85 ID3D11ShaderResourceView *_weights;
\n+
86 ID3D11Buffer *_sizesBuffer;
\n+
87 ID3D11Buffer *_offsetsBuffer;
\n+
88 ID3D11Buffer *_indicesBuffer;
\n+
89 ID3D11Buffer *_weightsBuffer;
\n+
90
\n+
91 int _numStencils;
\n+
92};
\n
93
\n-
94} // end namespace OPENSUBDIV_VERSION
\n-
95using namespace OPENSUBDIV_VERSION;
\n-
96
\n-
97} // end namespace OpenSubdiv
\n-
98
\n-
99#endif // OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H
\n+
94// ---------------------------------------------------------------------------
\n+
95
\n+
96class D3D11ComputeEvaluator {
\n+
97public:
\n+
98 typedef bool Instantiatable;
\n+
99 static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n+
100 BufferDescriptor const &dstDesc,
\n+
101 BufferDescriptor const &duDesc,
\n+
102 BufferDescriptor const &dvDesc,
\n+
103 ID3D11DeviceContext *deviceContext);
\n+
104
\n+
105 static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n+
106 BufferDescriptor const &dstDesc,
\n+
107 BufferDescriptor const &duDesc,
\n+
108 BufferDescriptor const &dvDesc,
\n+
109 BufferDescriptor const &duuDesc,
\n+
110 BufferDescriptor const &duvDesc,
\n+
111 BufferDescriptor const &dvvDesc,
\n+
112 ID3D11DeviceContext *deviceContext);
\n+
113
\n+
115 D3D11ComputeEvaluator();
\n+
116
\n+
118 ~D3D11ComputeEvaluator();
\n+
119
\n+
147 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
148 static bool EvalStencils(
\n+
149 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
150 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
151 STENCIL_TABLE const *stencilTable,
\n+
152 D3D11ComputeEvaluator const *instance,
\n+
153 ID3D11DeviceContext * deviceContext) {
\n+
154 if (instance) {
\n+
155 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
156 dstBuffer, dstDesc,
\n+
157 stencilTable,
\n+
158 deviceContext);
\n+
159 } else {
\n+
160 // Create an instance on demand (slow)
\n+
161 (void)deviceContext; // unused
\n+
162 instance = Create(srcDesc, dstDesc,
\n+
163 BufferDescriptor(),
\n+
164 BufferDescriptor(),
\n+
165 deviceContext);
\n+
166 if (instance) {
\n+
167 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
168 dstBuffer, dstDesc,
\n+
169 stencilTable,
\n+
170 deviceContext);
\n+
171 delete instance;
\n+
172 return r;
\n+
173 }
\n+
174 return false;
\n+
175 }
\n+
176 }
\n+
177
\n+
180 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
181 bool EvalStencils(
\n+
182 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
183 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
184 STENCIL_TABLE const *stencilTable,
\n+
185 ID3D11DeviceContext *deviceContext) const {
\n+
186 return EvalStencils(srcBuffer->BindD3D11UAV(deviceContext), srcDesc,
\n+
187 dstBuffer->BindD3D11UAV(deviceContext), dstDesc,
\n+
188 stencilTable->GetSizesSRV(),
\n+
189 stencilTable->GetOffsetsSRV(),
\n+
190 stencilTable->GetIndicesSRV(),
\n+
191 stencilTable->GetWeightsSRV(),
\n+
192 /* start = */ 0,
\n+
193 /* end = */ stencilTable->GetNumStencils(),
\n+
194 deviceContext);
\n+
195 }
\n+
196
\n+
199 bool EvalStencils(ID3D11UnorderedAccessView *srcSRV,
\n+
200 BufferDescriptor const &srcDesc,
\n+
201 ID3D11UnorderedAccessView *dstUAV,
\n+
202 BufferDescriptor const &dstDesc,
\n+
203 ID3D11ShaderResourceView *sizesSRV,
\n+
204 ID3D11ShaderResourceView *offsetsSRV,
\n+
205 ID3D11ShaderResourceView *indicesSRV,
\n+
206 ID3D11ShaderResourceView *weightsSRV,
\n+
207 int start,
\n+
208 int end,
\n+
209 ID3D11DeviceContext *deviceContext) const;
\n+
210
\n+
212 bool Compile(BufferDescriptor const &srcDesc,
\n+
213 BufferDescriptor const &dstDesc,
\n+
214 ID3D11DeviceContext *deviceContext);
\n+
215
\n+
217 static void Synchronize(ID3D11DeviceContext *deviceContext);
\n+
218
\n+
219private:
\n+
220 ID3D11ComputeShader * _computeShader;
\n+
221 ID3D11ClassLinkage * _classLinkage;
\n+
222 ID3D11ClassInstance * _singleBufferKernel;
\n+
223 ID3D11ClassInstance * _separateBufferKernel;
\n+
224 ID3D11Buffer * _uniformArgs; // uniform parameters for kernels
\n+
225
\n+
226 int _workGroupSize;
\n+
227};
\n+
228
\n+
229} // end namespace Osd
\n+
230
\n+
231} // end namespace OPENSUBDIV_VERSION
\n+
232using namespace OPENSUBDIV_VERSION;
\n+
233
\n+
234} // end namespace OpenSubdiv
\n+
235
\n+
236
\n+
237#endif // OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer
Concrete vertex buffer class for Cpu subdivision and DirectX drawing.
Definition: cpuD3D11VertexBuffer.h:48
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::Create
static CpuD3D11VertexBuffer * Create(int numElements, int numVertices, ID3D11DeviceContext *deviceContext)
Creator. Returns NULL if error.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::CpuD3D11VertexBuffer
CpuD3D11VertexBuffer(int numElements, int numVertices)
Constructor.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::GetNumElements
int GetNumElements() const
Returns how many elements defined in this vertex buffer.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::UpdateData
void UpdateData(const float *src, int startVertex, int numVertices, void *deviceContext=NULL)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::GetNumVertices
int GetNumVertices() const
Returns how many vertices allocated in this vertex buffer.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindD3D11Buffer
ID3D11Buffer * BindD3D11Buffer(ID3D11DeviceContext *deviceContext)
Returns the D3D11 buffer object.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindCpuBuffer
float * BindCpuBuffer()
Returns the address of CPU buffer.
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindVBO
ID3D11Buffer * BindVBO(ID3D11DeviceContext *deviceContext)
Returns the D3D11 buffer object (for Osd::Mesh interface)
Definition: cpuD3D11VertexBuffer.h:75
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::Create
static D3D11StencilTable * Create(Far::StencilTable const *stencilTable, DEVICE_CONTEXT context)
Definition: d3d11ComputeEvaluator.h:59
\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::D3D11StencilTable
D3D11StencilTable(Far::StencilTable const *stencilTable, ID3D11DeviceContext *deviceContext)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::Create
static D3D11StencilTable * Create(Far::StencilTable const *stencilTable, ID3D11DeviceContext *deviceContext)
Definition: d3d11ComputeEvaluator.h:64
\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Create
static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, ID3D11DeviceContext *deviceContext)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, D3D11ComputeEvaluator const *instance, ID3D11DeviceContext *deviceContext)
Generic static compute function. This function has a same signature as other device kernels have so t...
Definition: d3d11ComputeEvaluator.h:148
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Create
static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, ID3D11DeviceContext *deviceContext)
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Synchronize
static void Synchronize(ID3D11DeviceContext *deviceContext)
Wait the dispatched kernel finishes.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, ID3D11DeviceContext *deviceContext) const
Definition: d3d11ComputeEvaluator.h:181
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Compile
bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, ID3D11DeviceContext *deviceContext)
Configure DX kernel. Returns false if it fails to compile the kernel.
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils
bool EvalStencils(ID3D11UnorderedAccessView *srcSRV, BufferDescriptor const &srcDesc, ID3D11UnorderedAccessView *dstUAV, BufferDescriptor const &dstDesc, ID3D11ShaderResourceView *sizesSRV, ID3D11ShaderResourceView *offsetsSRV, ID3D11ShaderResourceView *indicesSRV, ID3D11ShaderResourceView *weightsSRV, int start, int end, ID3D11DeviceContext *deviceContext) const
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-cpuD3D11VertexBuffer.h\n+d3d11ComputeEvaluator.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2013 Pixar\n+ 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,105 +30,274 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H\n- 26#define OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H\n+ 25#ifndef OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H\n+ 26#define OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H\n 27\n 28#include \"../version.h\"\n- 29#include \n- 30\n+ 29\n+ 30struct ID3D11DeviceContext;\n 31struct ID3D11Buffer;\n- 32struct ID3D11Device;\n- 33struct ID3D11DeviceContext;\n- 34\n- 35namespace OpenSubdiv {\n- 36namespace OPENSUBDIV_VERSION {\n+ 32struct ID3D11ComputeShader;\n+ 33struct ID3D11ClassLinkage;\n+ 34struct ID3D11ClassInstance;\n+ 35struct ID3D11ShaderResourceView;\n+ 36struct ID3D11UnorderedAccessView;\n 37\n- 38namespace Osd {\n+ 38#include \"../osd/bufferDescriptor.h\"\n 39\n-48class CpuD3D11VertexBuffer {\n- 49public:\n-51 static CpuD3D11VertexBuffer * Create(int numElements, int numVertices,\n- 52 ID3D11DeviceContext *deviceContext);\n- 53\n-55 virtual ~CpuD3D11VertexBuffer();\n- 56\n-59 void UpdateData(const float *src, int startVertex, int numVertices,\n- 60 void *deviceContext = NULL);\n- 61\n-63 int GetNumElements() const;\n- 64\n-66 int GetNumVertices() const;\n- 67\n-69 float * BindCpuBuffer();\n- 70\n-72 ID3D11Buffer *BindD3D11Buffer(ID3D11DeviceContext *deviceContext);\n+ 40namespace OpenSubdiv {\n+ 41namespace OPENSUBDIV_VERSION {\n+ 42\n+ 43namespace Far {\n+ 44 class StencilTable;\n+ 45}\n+ 46\n+ 47namespace Osd {\n+ 48\n+56class D3D11StencilTable {\n+ 57public:\n+ 58 template \n+59 static D3D11StencilTable *Create(Far::StencilTable const *stencilTable,\n+ 60 DEVICE_CONTEXT context) {\n+ 61 return new D3D11StencilTable(stencilTable, context->GetDeviceContext());\n+ 62 }\n+ 63\n+64 static D3D11StencilTable *Create(Far::StencilTable const *stencilTable,\n+ 65 ID3D11DeviceContext *deviceContext) {\n+ 66 return new D3D11StencilTable(stencilTable, deviceContext);\n+ 67 }\n+ 68\n+69 D3D11StencilTable(Far::StencilTable const *stencilTable,\n+ 70 ID3D11DeviceContext *deviceContext);\n+ 71\n+72 ~D3D11StencilTable();\n 73\n-75 ID3D11Buffer *BindVBO(ID3D11DeviceContext *deviceContext) {\n- 76 return BindD3D11Buffer(deviceContext);\n- 77 }\n- 78\n- 79protected:\n-81 CpuD3D11VertexBuffer(int numElements, int numVertices);\n- 82\n-83 bool allocate(ID3D11Device *device);\n- 84\n- 85private:\n- 86 int _numElements;\n- 87 int _numVertices;\n- 88 ID3D11Buffer *_d3d11Buffer;\n- 89 float *_cpuBuffer;\n- 90};\n- 91\n- 92} // end namespace Osd\n+ 74 // interfaces needed for D3D11ComputeEvaluator\n+75 ID3D11ShaderResourceView *GetSizesSRV() const { return _sizes; }\n+76 ID3D11ShaderResourceView *GetOffsetsSRV() const { return _offsets; }\n+77 ID3D11ShaderResourceView *GetIndicesSRV() const { return _indices; }\n+78 ID3D11ShaderResourceView *GetWeightsSRV() const { return _weights; }\n+79 int GetNumStencils() const { return _numStencils; }\n+ 80\n+ 81private:\n+ 82 ID3D11ShaderResourceView *_sizes;\n+ 83 ID3D11ShaderResourceView *_offsets;\n+ 84 ID3D11ShaderResourceView *_indices;\n+ 85 ID3D11ShaderResourceView *_weights;\n+ 86 ID3D11Buffer *_sizesBuffer;\n+ 87 ID3D11Buffer *_offsetsBuffer;\n+ 88 ID3D11Buffer *_indicesBuffer;\n+ 89 ID3D11Buffer *_weightsBuffer;\n+ 90\n+ 91 int _numStencils;\n+ 92};\n 93\n- 94} // end namespace OPENSUBDIV_VERSION\n- 95using namespace OPENSUBDIV_VERSION;\n- 96\n- 97} // end namespace OpenSubdiv\n- 98\n- 99#endif // OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H\n+ 94// -------------------------------------------------------------------------\n+--\n+ 95\n+96class D3D11ComputeEvaluator {\n+ 97public:\n+98 typedef bool Instantiatable;\n+99 static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n+ 100 BufferDescriptor const &dstDesc,\n+ 101 BufferDescriptor const &duDesc,\n+ 102 BufferDescriptor const &dvDesc,\n+ 103 ID3D11DeviceContext *deviceContext);\n+ 104\n+105 static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n+ 106 BufferDescriptor const &dstDesc,\n+ 107 BufferDescriptor const &duDesc,\n+ 108 BufferDescriptor const &dvDesc,\n+ 109 BufferDescriptor const &duuDesc,\n+ 110 BufferDescriptor const &duvDesc,\n+ 111 BufferDescriptor const &dvvDesc,\n+ 112 ID3D11DeviceContext *deviceContext);\n+ 113\n+115 D3D11ComputeEvaluator();\n+ 116\n+118 ~D3D11ComputeEvaluator();\n+ 119\n+ 147 template \n+148 static bool EvalStencils(\n+ 149 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 150 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 151 STENCIL_TABLE const *stencilTable,\n+ 152 D3D11ComputeEvaluator const *instance,\n+ 153 ID3D11DeviceContext * deviceContext) {\n+ 154 if (instance) {\n+ 155 return instance->EvalStencils(srcBuffer, srcDesc,\n+ 156 dstBuffer, dstDesc,\n+ 157 stencilTable,\n+ 158 deviceContext);\n+ 159 } else {\n+ 160 // Create an instance on demand (slow)\n+ 161 (void)deviceContext; // unused\n+ 162 instance = Create(srcDesc, dstDesc,\n+ 163 BufferDescriptor(),\n+ 164 BufferDescriptor(),\n+ 165 deviceContext);\n+ 166 if (instance) {\n+ 167 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 168 dstBuffer, dstDesc,\n+ 169 stencilTable,\n+ 170 deviceContext);\n+ 171 delete instance;\n+ 172 return r;\n+ 173 }\n+ 174 return false;\n+ 175 }\n+ 176 }\n+ 177\n+ 180 template \n+181 bool EvalStencils(\n+ 182 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 183 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 184 STENCIL_TABLE const *stencilTable,\n+ 185 ID3D11DeviceContext *deviceContext) const {\n+ 186 return EvalStencils(srcBuffer->BindD3D11UAV(deviceContext), srcDesc,\n+ 187 dstBuffer->BindD3D11UAV(deviceContext), dstDesc,\n+ 188 stencilTable->GetSizesSRV(),\n+ 189 stencilTable->GetOffsetsSRV(),\n+ 190 stencilTable->GetIndicesSRV(),\n+ 191 stencilTable->GetWeightsSRV(),\n+ 192 /* start = */ 0,\n+ 193 /* end = */ stencilTable->GetNumStencils(),\n+ 194 deviceContext);\n+ 195 }\n+ 196\n+199 bool EvalStencils(ID3D11UnorderedAccessView *srcSRV,\n+ 200 BufferDescriptor const &srcDesc,\n+ 201 ID3D11UnorderedAccessView *dstUAV,\n+ 202 BufferDescriptor const &dstDesc,\n+ 203 ID3D11ShaderResourceView *sizesSRV,\n+ 204 ID3D11ShaderResourceView *offsetsSRV,\n+ 205 ID3D11ShaderResourceView *indicesSRV,\n+ 206 ID3D11ShaderResourceView *weightsSRV,\n+ 207 int start,\n+ 208 int end,\n+ 209 ID3D11DeviceContext *deviceContext) const;\n+ 210\n+212 bool Compile(BufferDescriptor const &srcDesc,\n+ 213 BufferDescriptor const &dstDesc,\n+ 214 ID3D11DeviceContext *deviceContext);\n+ 215\n+217 static void Synchronize(ID3D11DeviceContext *deviceContext);\n+ 218\n+ 219private:\n+ 220 ID3D11ComputeShader * _computeShader;\n+ 221 ID3D11ClassLinkage * _classLinkage;\n+ 222 ID3D11ClassInstance * _singleBufferKernel;\n+ 223 ID3D11ClassInstance * _separateBufferKernel;\n+ 224 ID3D11Buffer * _uniformArgs; // uniform parameters for kernels\n+ 225\n+ 226 int _workGroupSize;\n+ 227};\n+ 228\n+ 229} // end namespace Osd\n+ 230\n+ 231} // end namespace OPENSUBDIV_VERSION\n+ 232using namespace OPENSUBDIV_VERSION;\n+ 233\n+ 234} // end namespace OpenSubdiv\n+ 235\n+ 236\n+ 237#endif // OPENSUBDIV3_OSD_D3D11_COMPUTE_EVALUATOR_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer\n-Concrete vertex buffer class for Cpu subdivision and DirectX drawing.\n-Definition: cpuD3D11VertexBuffer.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::Create\n-static CpuD3D11VertexBuffer * Create(int numElements, int numVertices,\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n+Stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:273\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n+BufferDescriptor is a struct which describes buffer elements in interleaved\n+data buffers....\n+Definition: bufferDescriptor.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable\n+D3D11 stencil table.\n+Definition: d3d11ComputeEvaluator.h:56\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::Create\n+static D3D11StencilTable * Create(Far::StencilTable const *stencilTable,\n+DEVICE_CONTEXT context)\n+Definition: d3d11ComputeEvaluator.h:59\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::GetOffsetsSRV\n+ID3D11ShaderResourceView * GetOffsetsSRV() const\n+Definition: d3d11ComputeEvaluator.h:76\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::GetNumStencils\n+int GetNumStencils() const\n+Definition: d3d11ComputeEvaluator.h:79\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::GetIndicesSRV\n+ID3D11ShaderResourceView * GetIndicesSRV() const\n+Definition: d3d11ComputeEvaluator.h:77\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::~D3D11StencilTable\n+~D3D11StencilTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::D3D11StencilTable\n+D3D11StencilTable(Far::StencilTable const *stencilTable, ID3D11DeviceContext\n+*deviceContext)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::Create\n+static D3D11StencilTable * Create(Far::StencilTable const *stencilTable,\n ID3D11DeviceContext *deviceContext)\n-Creator. Returns NULL if error.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::CpuD3D11VertexBuffer\n-CpuD3D11VertexBuffer(int numElements, int numVertices)\n+Definition: d3d11ComputeEvaluator.h:64\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::GetSizesSRV\n+ID3D11ShaderResourceView * GetSizesSRV() const\n+Definition: d3d11ComputeEvaluator.h:75\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::GetWeightsSRV\n+ID3D11ShaderResourceView * GetWeightsSRV() const\n+Definition: d3d11ComputeEvaluator.h:78\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator\n+Definition: d3d11ComputeEvaluator.h:96\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Create\n+static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, ID3D11DeviceContext *deviceContext)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n+const *stencilTable, D3D11ComputeEvaluator const *instance, ID3D11DeviceContext\n+*deviceContext)\n+Generic static compute function. This function has a same signature as other\n+device kernels have so t...\n+Definition: d3d11ComputeEvaluator.h:148\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Create\n+static D3D11ComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n+BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc,\n+ID3D11DeviceContext *deviceContext)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Instantiatable\n+bool Instantiatable\n+Definition: d3d11ComputeEvaluator.h:98\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Synchronize\n+static void Synchronize(ID3D11DeviceContext *deviceContext)\n+Wait the dispatched kernel finishes.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils\n+bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const\n+*stencilTable, ID3D11DeviceContext *deviceContext) const\n+Definition: d3d11ComputeEvaluator.h:181\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::Compile\n+bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc,\n+ID3D11DeviceContext *deviceContext)\n+Configure DX kernel. Returns false if it fails to compile the kernel.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::\n+D3D11ComputeEvaluator\n+D3D11ComputeEvaluator()\n Constructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::GetNumElements\n-int GetNumElements() const\n-Returns how many elements defined in this vertex buffer.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::UpdateData\n-void UpdateData(const float *src, int startVertex, int numVertices, void\n-*deviceContext=NULL)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::GetNumVertices\n-int GetNumVertices() const\n-Returns how many vertices allocated in this vertex buffer.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindD3D11Buffer\n-ID3D11Buffer * BindD3D11Buffer(ID3D11DeviceContext *deviceContext)\n-Returns the D3D11 buffer object.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindCpuBuffer\n-float * BindCpuBuffer()\n-Returns the address of CPU buffer.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::allocate\n-bool allocate(ID3D11Device *device)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::\n-~CpuD3D11VertexBuffer\n-virtual ~CpuD3D11VertexBuffer()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::\n+~D3D11ComputeEvaluator\n+~D3D11ComputeEvaluator()\n Destructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindVBO\n-ID3D11Buffer * BindVBO(ID3D11DeviceContext *deviceContext)\n-Returns the D3D11 buffer object (for Osd::Mesh interface)\n-Definition: cpuD3D11VertexBuffer.h:75\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::EvalStencils\n+bool EvalStencils(ID3D11UnorderedAccessView *srcSRV, BufferDescriptor const\n+&srcDesc, ID3D11UnorderedAccessView *dstUAV, BufferDescriptor const &dstDesc,\n+ID3D11ShaderResourceView *sizesSRV, ID3D11ShaderResourceView *offsetsSRV,\n+ID3D11ShaderResourceView *indicesSRV, ID3D11ShaderResourceView *weightsSRV, int\n+start, int end, ID3D11DeviceContext *deviceContext) const\n * opensubdiv\n * osd\n- * cpuD3D11VertexBuffer.h\n+ * d3d11ComputeEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00845.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00845.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/d3d11PatchTable.h File Reference\n+OpenSubdiv: opensubdiv/osd/glXFBEvaluator.h File Reference\n \n \n \n \n \n \n \n@@ -83,28 +83,30 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
d3d11PatchTable.h File Reference
\n+
glXFBEvaluator.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include <vector>
\n-#include "../far/patchDescriptor.h"
\n-#include "../osd/nonCopyable.h"
\n+#include "../osd/opengl.h"
\n #include "../osd/types.h"
\n+#include "../osd/bufferDescriptor.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n+\n+\n+\n+\n \n

\n Classes

class  D3D11PatchTable
class  GLStencilTableTBO
 GL TextureBuffer stencil table. More...
 
class  GLXFBEvaluator
 
\n \n \n \n \n@@ -115,13 +117,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,30 +5,32 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-d3d11PatchTable.h File Reference\n+glXFBEvaluator.h File Reference\n #include \"../version.h\"\n-#include \n-#include \"../far/patchDescriptor.h\"\n-#include \"../osd/nonCopyable.h\"\n+#include \"../osd/opengl.h\"\n #include \"../osd/types.h\"\n+#include \"../osd/bufferDescriptor.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0D3D11PatchTable\n+class \u00a0GLStencilTableTBO\n+\u00a0 GL TextureBuffer stencil table. More...\n+\u00a0\n+class \u00a0GLXFBEvaluator\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * d3d11PatchTable.h\n+ * glXFBEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00845.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00845.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,4 @@\n var a00845 = [\n- [\"D3D11PatchTable\", \"a01201.html\", \"a01201\"]\n+ [\"GLStencilTableTBO\", \"a01241.html\", \"a01241\"],\n+ [\"GLXFBEvaluator\", \"a01245.html\", \"a01245\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00845_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00845_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/d3d11PatchTable.h Source File\n+OpenSubdiv: opensubdiv/osd/glXFBEvaluator.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
d3d11PatchTable.h
\n+
glXFBEvaluator.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,108 +107,1137 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n-
26#define OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n+
25#ifndef OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H
\n+
26#define OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30#include <vector>
\n-
31#include "../far/patchDescriptor.h"
\n-
32#include "../osd/nonCopyable.h"
\n-
33#include "../osd/types.h"
\n-
34
\n-
35struct ID3D11Buffer;
\n-
36struct ID3D11ShaderResourceView;
\n-
37struct ID3D11Device;
\n-
38struct ID3D11DeviceContext;
\n-
39
\n-
40namespace OpenSubdiv {
\n-
41namespace OPENSUBDIV_VERSION {
\n+
30#include "../osd/opengl.h"
\n+
31#include "../osd/types.h"
\n+
32#include "../osd/bufferDescriptor.h"
\n+
33
\n+
34namespace OpenSubdiv {
\n+
35namespace OPENSUBDIV_VERSION {
\n+
36
\n+
37namespace Far {
\n+
38 class PatchTable;
\n+
39 class StencilTable;
\n+
40 class LimitStencilTable;
\n+
41}
\n
42
\n-
43namespace Far{
\n-
44 class PatchTable;
\n-
45};
\n-
46
\n-
47namespace Osd {
\n-
48
\n-
49class D3D11PatchTable : private NonCopyable<D3D11PatchTable> {
\n-
50public:
\n-
51 typedef ID3D11Buffer * VertexBufferBinding;
\n-
52
\n-
53 D3D11PatchTable();
\n-
54 ~D3D11PatchTable();
\n-
55
\n-
56 template<typename DEVICE_CONTEXT>
\n-
57 static D3D11PatchTable *Create(Far::PatchTable const *farPatchTable,
\n-
58 DEVICE_CONTEXT context) {
\n-
59 return Create(farPatchTable, context->GetDeviceContext());
\n-
60 }
\n-
61
\n-
62 static D3D11PatchTable *Create(Far::PatchTable const *farPatchTable,
\n-
63 ID3D11DeviceContext *deviceContext);
\n-
64
\n-
65 PatchArrayVector const &GetPatchArrays() const {
\n-
66 return _patchArrays;
\n-
67 }
\n-
68
\n-
70 ID3D11Buffer* GetPatchIndexBuffer() const {
\n-
71 return _indexBuffer;
\n-
72 }
\n-
73
\n-
75 ID3D11ShaderResourceView* GetPatchParamSRV() const {
\n-
76 return _patchParamBufferSRV;
\n-
77 }
\n-
78
\n-
79protected:
\n-
80 // allocate buffers from patchTable
\n-
81 bool allocate(Far::PatchTable const *farPatchTable,
\n-
82 ID3D11DeviceContext *deviceContext);
\n-
83
\n-
84 PatchArrayVector _patchArrays;
\n-
85
\n-
86 ID3D11Buffer *_indexBuffer;
\n-
87 ID3D11Buffer *_patchParamBuffer;
\n-
88 ID3D11ShaderResourceView *_patchParamBufferSRV;
\n-
89};
\n-
90
\n-
91
\n-
92} // end namespace Osd
\n-
93
\n-
94} // end namespace OPENSUBDIV_VERSION
\n-
95using namespace OPENSUBDIV_VERSION;
\n-
96
\n-
97} // end namespace OpenSubdiv
\n-
98
\n-
99#endif // OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n+
43namespace Osd {
\n+
44
\n+
52class GLStencilTableTBO {
\n+
53public:
\n+
54 static GLStencilTableTBO *Create(
\n+
55 Far::StencilTable const *stencilTable, void *deviceContext = NULL) {
\n+
56 (void)deviceContext; // unused
\n+
57 return new GLStencilTableTBO(stencilTable);
\n+
58 }
\n+
59
\n+
60 static GLStencilTableTBO *Create(
\n+
61 Far::LimitStencilTable const *limitStencilTable,
\n+
62 void *deviceContext = NULL) {
\n+
63 (void)deviceContext; // unused
\n+
64 return new GLStencilTableTBO(limitStencilTable);
\n+
65 }
\n+
66
\n+
67 explicit GLStencilTableTBO(Far::StencilTable const *stencilTable);
\n+
68 explicit GLStencilTableTBO(Far::LimitStencilTable const *limitStencilTable);
\n+
69 ~GLStencilTableTBO();
\n+
70
\n+
71 // interfaces needed for GLSLTransformFeedbackKernel
\n+
72 GLuint GetSizesTexture() const { return _sizes; }
\n+
73 GLuint GetOffsetsTexture() const { return _offsets; }
\n+
74 GLuint GetIndicesTexture() const { return _indices; }
\n+
75 GLuint GetWeightsTexture() const { return _weights; }
\n+
76 GLuint GetDuWeightsTexture() const { return _duWeights; }
\n+
77 GLuint GetDvWeightsTexture() const { return _dvWeights; }
\n+
78 GLuint GetDuuWeightsTexture() const { return _duuWeights; }
\n+
79 GLuint GetDuvWeightsTexture() const { return _duvWeights; }
\n+
80 GLuint GetDvvWeightsTexture() const { return _dvvWeights; }
\n+
81 int GetNumStencils() const { return _numStencils; }
\n+
82
\n+
83private:
\n+
84 GLuint _sizes;
\n+
85 GLuint _offsets;
\n+
86 GLuint _indices;
\n+
87 GLuint _weights;
\n+
88 GLuint _duWeights;
\n+
89 GLuint _dvWeights;
\n+
90 GLuint _duuWeights;
\n+
91 GLuint _duvWeights;
\n+
92 GLuint _dvvWeights;
\n+
93 int _numStencils;
\n+
94};
\n+
95
\n+
96// ---------------------------------------------------------------------------
\n+
97
\n+
98class GLXFBEvaluator {
\n+
99public:
\n+
100 typedef bool Instantiatable;
\n+
101
\n+
103 template <typename DEVICE_CONTEXT>
\n+
104 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,
\n+
105 BufferDescriptor const &dstDesc,
\n+
106 BufferDescriptor const &duDesc,
\n+
107 BufferDescriptor const &dvDesc,
\n+
108 DEVICE_CONTEXT deviceContext) {
\n+
109 bool interleavedDerivativeBuffers = deviceContext
\n+
110 ? deviceContext->AreInterleavedDerivativeBuffers()
\n+
111 : false;
\n+
112 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n+
113 interleavedDerivativeBuffers);
\n+
114 }
\n+
115
\n+
117 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,
\n+
118 BufferDescriptor const &dstDesc,
\n+
119 BufferDescriptor const &duDesc,
\n+
120 BufferDescriptor const &dvDesc,
\n+
121 void * deviceContext) {
\n+
122 (void)deviceContext; // not used
\n+
123 return Create(srcDesc, dstDesc, duDesc, dvDesc);
\n+
124 }
\n+
125
\n+
126 static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,
\n+
127 BufferDescriptor const &dstDesc,
\n+
128 BufferDescriptor const &duDesc,
\n+
129 BufferDescriptor const &dvDesc,
\n+
130 bool interleavedDerivativeBuffers = false) {
\n+
131 GLXFBEvaluator *instance = new GLXFBEvaluator(interleavedDerivativeBuffers);
\n+
132 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc))
\n+
133 return instance;
\n+
134 delete instance;
\n+
135 return NULL;
\n+
136 }
\n+
137
\n+
139 template <typename DEVICE_CONTEXT>
\n+
140 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,
\n+
141 BufferDescriptor const &dstDesc,
\n+
142 BufferDescriptor const &duDesc,
\n+
143 BufferDescriptor const &dvDesc,
\n+
144 BufferDescriptor const &duuDesc,
\n+
145 BufferDescriptor const &duvDesc,
\n+
146 BufferDescriptor const &dvvDesc,
\n+
147 DEVICE_CONTEXT deviceContext) {
\n+
148 bool interleavedDerivativeBuffers = deviceContext
\n+
149 ? deviceContext->AreInterleavedDerivativeBuffers()
\n+
150 : false;
\n+
151 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n+
152 duuDesc, duvDesc, dvvDesc,
\n+
153 interleavedDerivativeBuffers);
\n+
154 }
\n+
155
\n+
157 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,
\n+
158 BufferDescriptor const &dstDesc,
\n+
159 BufferDescriptor const &duDesc,
\n+
160 BufferDescriptor const &dvDesc,
\n+
161 BufferDescriptor const &duuDesc,
\n+
162 BufferDescriptor const &duvDesc,
\n+
163 BufferDescriptor const &dvvDesc,
\n+
164 void * deviceContext) {
\n+
165 (void)deviceContext; // not used
\n+
166 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n+
167 duuDesc, duvDesc, dvvDesc);
\n+
168 }
\n+
169
\n+
170 static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,
\n+
171 BufferDescriptor const &dstDesc,
\n+
172 BufferDescriptor const &duDesc,
\n+
173 BufferDescriptor const &dvDesc,
\n+
174 BufferDescriptor const &duuDesc,
\n+
175 BufferDescriptor const &duvDesc,
\n+
176 BufferDescriptor const &dvvDesc,
\n+
177 bool interleavedDerivativeBuffers = false) {
\n+
178 GLXFBEvaluator *instance = new GLXFBEvaluator(interleavedDerivativeBuffers);
\n+
179 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc,
\n+
180 duuDesc, duvDesc, dvvDesc))
\n+
181 return instance;
\n+
182 delete instance;
\n+
183 return NULL;
\n+
184 }
\n+
185
\n+
199 GLXFBEvaluator(bool interleavedDerivativeBuffers = false);
\n+
200
\n+
202 ~GLXFBEvaluator();
\n+
203
\n+
209
\n+
237 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
238 static bool EvalStencils(
\n+
239 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
240 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
241 STENCIL_TABLE const *stencilTable,
\n+
242 GLXFBEvaluator const *instance,
\n+
243 void * deviceContext = NULL) {
\n+
244
\n+
245 if (instance) {
\n+
246 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
247 dstBuffer, dstDesc,
\n+
248 stencilTable);
\n+
249 } else {
\n+
250 // Create an instance on demand (slow)
\n+
251 (void)deviceContext; // unused
\n+
252 instance = Create(srcDesc, dstDesc,
\n+
253 BufferDescriptor(),
\n+
254 BufferDescriptor());
\n+
255 if (instance) {
\n+
256 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
257 dstBuffer, dstDesc,
\n+
258 stencilTable);
\n+
259 delete instance;
\n+
260 return r;
\n+
261 }
\n+
262 return false;
\n+
263 }
\n+
264 }
\n+
265
\n+
305 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
306 static bool EvalStencils(
\n+
307 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
308 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
309 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
310 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
311 STENCIL_TABLE const *stencilTable,
\n+
312 GLXFBEvaluator const *instance,
\n+
313 void * deviceContext = NULL) {
\n+
314
\n+
315 if (instance) {
\n+
316 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
317 dstBuffer, dstDesc,
\n+
318 duBuffer, duDesc,
\n+
319 dvBuffer, dvDesc,
\n+
320 stencilTable);
\n+
321 } else {
\n+
322 // Create an instance on demand (slow)
\n+
323 (void)deviceContext; // unused
\n+
324 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);
\n+
325 if (instance) {
\n+
326 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
327 dstBuffer, dstDesc,
\n+
328 duBuffer, duDesc,
\n+
329 dvBuffer, dvDesc,
\n+
330 stencilTable);
\n+
331 delete instance;
\n+
332 return r;
\n+
333 }
\n+
334 return false;
\n+
335 }
\n+
336 }
\n+
337
\n+
395 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
396 static bool EvalStencils(
\n+
397 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
398 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
399 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
400 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
401 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
402 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
403 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
404 STENCIL_TABLE const *stencilTable,
\n+
405 GLXFBEvaluator const *instance,
\n+
406 void * deviceContext = NULL) {
\n+
407
\n+
408 if (instance) {
\n+
409 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
410 dstBuffer, dstDesc,
\n+
411 duBuffer, duDesc,
\n+
412 dvBuffer, dvDesc,
\n+
413 duuBuffer, duuDesc,
\n+
414 duvBuffer, duvDesc,
\n+
415 dvvBuffer, dvvDesc,
\n+
416 stencilTable);
\n+
417 } else {
\n+
418 // Create an instance on demand (slow)
\n+
419 (void)deviceContext; // unused
\n+
420 instance = Create(srcDesc, dstDesc,
\n+
421 duDesc, dvDesc,
\n+
422 duuDesc, duvDesc, dvvDesc);
\n+
423 if (instance) {
\n+
424 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
425 dstBuffer, dstDesc,
\n+
426 duBuffer, duDesc,
\n+
427 dvBuffer, dvDesc,
\n+
428 duuBuffer, duuDesc,
\n+
429 duvBuffer, duvDesc,
\n+
430 dvvBuffer, dvvDesc,
\n+
431 stencilTable);
\n+
432 delete instance;
\n+
433 return r;
\n+
434 }
\n+
435 return false;
\n+
436 }
\n+
437 }
\n+
438
\n+
456 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
457 bool EvalStencils(
\n+
458 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
459 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
460 STENCIL_TABLE const *stencilTable) const {
\n+
461
\n+
462 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n+
463 dstBuffer->BindVBO(), dstDesc,
\n+
464 0, BufferDescriptor(),
\n+
465 0, BufferDescriptor(),
\n+
466 stencilTable->GetSizesTexture(),
\n+
467 stencilTable->GetOffsetsTexture(),
\n+
468 stencilTable->GetIndicesTexture(),
\n+
469 stencilTable->GetWeightsTexture(),
\n+
470 0,
\n+
471 0,
\n+
472 /* start = */ 0,
\n+
473 /* end = */ stencilTable->GetNumStencils());
\n+
474 }
\n+
475
\n+
505 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
506 bool EvalStencils(
\n+
507 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
508 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
509 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
510 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
511 STENCIL_TABLE const *stencilTable) const {
\n+
512
\n+
513 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n+
514 dstBuffer->BindVBO(), dstDesc,
\n+
515 duBuffer->BindVBO(), duDesc,
\n+
516 dvBuffer->BindVBO(), dvDesc,
\n+
517 stencilTable->GetSizesTexture(),
\n+
518 stencilTable->GetOffsetsTexture(),
\n+
519 stencilTable->GetIndicesTexture(),
\n+
520 stencilTable->GetWeightsTexture(),
\n+
521 stencilTable->GetDuWeightsTexture(),
\n+
522 stencilTable->GetDvWeightsTexture(),
\n+
523 /* start = */ 0,
\n+
524 /* end = */ stencilTable->GetNumStencils());
\n+
525 }
\n+
526
\n+
574 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
575 bool EvalStencils(
\n+
576 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
577 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
578 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
579 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
580 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
581 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
582 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
583 STENCIL_TABLE const *stencilTable) const {
\n+
584
\n+
585 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n+
586 dstBuffer->BindVBO(), dstDesc,
\n+
587 duBuffer->BindVBO(), duDesc,
\n+
588 dvBuffer->BindVBO(), dvDesc,
\n+
589 duuBuffer->BindVBO(), duuDesc,
\n+
590 duvBuffer->BindVBO(), duvDesc,
\n+
591 dvvBuffer->BindVBO(), dvvDesc,
\n+
592 stencilTable->GetSizesTexture(),
\n+
593 stencilTable->GetOffsetsTexture(),
\n+
594 stencilTable->GetIndicesTexture(),
\n+
595 stencilTable->GetWeightsTexture(),
\n+
596 stencilTable->GetDuWeightsTexture(),
\n+
597 stencilTable->GetDvWeightsTexture(),
\n+
598 stencilTable->GetDuuWeightsTexture(),
\n+
599 stencilTable->GetDuvWeightsTexture(),
\n+
600 stencilTable->GetDvvWeightsTexture(),
\n+
601 /* start = */ 0,
\n+
602 /* end = */ stencilTable->GetNumStencils());
\n+
603 }
\n+
604
\n+
640 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n+
641 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n+
642 GLuint duBuffer, BufferDescriptor const &duDesc,
\n+
643 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n+
644 GLuint sizesBuffer,
\n+
645 GLuint offsetsBuffer,
\n+
646 GLuint indicesBuffer,
\n+
647 GLuint weightsBuffer,
\n+
648 GLuint duWeightsBuffer,
\n+
649 GLuint dvWeightsBuffer,
\n+
650 int start,
\n+
651 int end) const;
\n+
652
\n+
706 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n+
707 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n+
708 GLuint duBuffer, BufferDescriptor const &duDesc,
\n+
709 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n+
710 GLuint duuBuffer, BufferDescriptor const &duuDesc,
\n+
711 GLuint duvBuffer, BufferDescriptor const &duvDesc,
\n+
712 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
713 GLuint sizesBuffer,
\n+
714 GLuint offsetsBuffer,
\n+
715 GLuint indicesBuffer,
\n+
716 GLuint weightsBuffer,
\n+
717 GLuint duWeightsBuffer,
\n+
718 GLuint dvWeightsBuffer,
\n+
719 GLuint duuWeightsBuffer,
\n+
720 GLuint duvWeightsBuffer,
\n+
721 GLuint dvvWeightsBuffer,
\n+
722 int start,
\n+
723 int end) const;
\n+
724
\n+
730
\n+
763 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
764 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
765 static bool EvalPatches(
\n+
766 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
767 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
768 int numPatchCoords,
\n+
769 PATCHCOORD_BUFFER *patchCoords,
\n+
770 PATCH_TABLE *patchTable,
\n+
771 GLXFBEvaluator const *instance,
\n+
772 void * deviceContext = NULL) {
\n+
773
\n+
774 if (instance) {
\n+
775 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
776 dstBuffer, dstDesc,
\n+
777 numPatchCoords, patchCoords,
\n+
778 patchTable);
\n+
779 } else {
\n+
780 // Create an instance on demand (slow)
\n+
781 (void)deviceContext; // unused
\n+
782 instance = Create(srcDesc, dstDesc,
\n+
783 BufferDescriptor(),
\n+
784 BufferDescriptor());
\n+
785 if (instance) {
\n+
786 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
787 dstBuffer, dstDesc,
\n+
788 numPatchCoords, patchCoords,
\n+
789 patchTable);
\n+
790 delete instance;
\n+
791 return r;
\n+
792 }
\n+
793 return false;
\n+
794 }
\n+
795 }
\n+
796
\n+
841 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
842 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
843 static bool EvalPatches(
\n+
844 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
845 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
846 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
847 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
848 int numPatchCoords,
\n+
849 PATCHCOORD_BUFFER *patchCoords,
\n+
850 PATCH_TABLE *patchTable,
\n+
851 GLXFBEvaluator const *instance,
\n+
852 void * deviceContext = NULL) {
\n+
853
\n+
854 if (instance) {
\n+
855 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
856 dstBuffer, dstDesc,
\n+
857 duBuffer, duDesc,
\n+
858 dvBuffer, dvDesc,
\n+
859 numPatchCoords, patchCoords,
\n+
860 patchTable);
\n+
861 } else {
\n+
862 // Create an instance on demand (slow)
\n+
863 (void)deviceContext; // unused
\n+
864 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);
\n+
865 if (instance) {
\n+
866 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
867 dstBuffer, dstDesc,
\n+
868 duBuffer, duDesc,
\n+
869 dvBuffer, dvDesc,
\n+
870 numPatchCoords, patchCoords,
\n+
871 patchTable);
\n+
872 delete instance;
\n+
873 return r;
\n+
874 }
\n+
875 return false;
\n+
876 }
\n+
877 }
\n+
878
\n+
941 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
942 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
943 static bool EvalPatches(
\n+
944 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
945 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
946 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
947 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
948 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
949 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
950 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
951 int numPatchCoords,
\n+
952 PATCHCOORD_BUFFER *patchCoords,
\n+
953 PATCH_TABLE *patchTable,
\n+
954 GLXFBEvaluator const *instance,
\n+
955 void * deviceContext = NULL) {
\n+
956
\n+
957 if (instance) {
\n+
958 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
959 dstBuffer, dstDesc,
\n+
960 duBuffer, duDesc,
\n+
961 dvBuffer, dvDesc,
\n+
962 duuBuffer, duuDesc,
\n+
963 duvBuffer, duvDesc,
\n+
964 dvvBuffer, dvvDesc,
\n+
965 numPatchCoords, patchCoords,
\n+
966 patchTable);
\n+
967 } else {
\n+
968 // Create an instance on demand (slow)
\n+
969 (void)deviceContext; // unused
\n+
970 instance = Create(srcDesc, dstDesc,
\n+
971 duDesc, dvDesc,
\n+
972 duuDesc, duvDesc, dvvDesc);
\n+
973 if (instance) {
\n+
974 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
975 dstBuffer, dstDesc,
\n+
976 duBuffer, duDesc,
\n+
977 dvBuffer, dvDesc,
\n+
978 duuBuffer, duuDesc,
\n+
979 duvBuffer, duvDesc,
\n+
980 dvvBuffer, dvvDesc,
\n+
981 numPatchCoords, patchCoords,
\n+
982 patchTable);
\n+
983 delete instance;
\n+
984 return r;
\n+
985 }
\n+
986 return false;
\n+
987 }
\n+
988 }
\n+
989
\n+
1014 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1015 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1016 bool EvalPatches(
\n+
1017 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1018 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1019 int numPatchCoords,
\n+
1020 PATCHCOORD_BUFFER *patchCoords,
\n+
1021 PATCH_TABLE *patchTable) const {
\n+
1022
\n+
1023 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1024 dstBuffer->BindVBO(), dstDesc,
\n+
1025 0, BufferDescriptor(),
\n+
1026 0, BufferDescriptor(),
\n+
1027 numPatchCoords,
\n+
1028 patchCoords->BindVBO(),
\n+
1029 patchTable->GetPatchArrays(),
\n+
1030 patchTable->GetPatchIndexTextureBuffer(),
\n+
1031 patchTable->GetPatchParamTextureBuffer());
\n+
1032 }
\n+
1033
\n+
1068 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1069 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1070 bool EvalPatches(
\n+
1071 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1072 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1073 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1074 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1075 int numPatchCoords,
\n+
1076 PATCHCOORD_BUFFER *patchCoords,
\n+
1077 PATCH_TABLE *patchTable) const {
\n+
1078
\n+
1079 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1080 dstBuffer->BindVBO(), dstDesc,
\n+
1081 duBuffer->BindVBO(), duDesc,
\n+
1082 dvBuffer->BindVBO(), dvDesc,
\n+
1083 numPatchCoords,
\n+
1084 patchCoords->BindVBO(),
\n+
1085 patchTable->GetPatchArrays(),
\n+
1086 patchTable->GetPatchIndexTextureBuffer(),
\n+
1087 patchTable->GetPatchParamTextureBuffer());
\n+
1088 }
\n+
1089
\n+
1142 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1143 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1144 bool EvalPatches(
\n+
1145 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1146 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1147 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1148 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1149 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1150 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1151 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1152 int numPatchCoords,
\n+
1153 PATCHCOORD_BUFFER *patchCoords,
\n+
1154 PATCH_TABLE *patchTable) const {
\n+
1155
\n+
1156 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1157 dstBuffer->BindVBO(), dstDesc,
\n+
1158 duBuffer->BindVBO(), duDesc,
\n+
1159 dvBuffer->BindVBO(), dvDesc,
\n+
1160 duuBuffer->BindVBO(), duuDesc,
\n+
1161 duvBuffer->BindVBO(), duvDesc,
\n+
1162 dvvBuffer->BindVBO(), dvvDesc,
\n+
1163 numPatchCoords,
\n+
1164 patchCoords->BindVBO(),
\n+
1165 patchTable->GetPatchArrays(),
\n+
1166 patchTable->GetPatchIndexTextureBuffer(),
\n+
1167 patchTable->GetPatchParamTextureBuffer());
\n+
1168 }
\n+
1169
\n+
1170 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n+
1171 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n+
1172 GLuint duBuffer, BufferDescriptor const &duDesc,
\n+
1173 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n+
1174 int numPatchCoords,
\n+
1175 GLuint patchCoordsBuffer,
\n+
1176 const PatchArrayVector &patchArrays,
\n+
1177 GLuint patchIndexBuffer,
\n+
1178 GLuint patchParamsBuffer) const;
\n+
1179
\n+
1180 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n+
1181 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n+
1182 GLuint duBuffer, BufferDescriptor const &duDesc,
\n+
1183 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n+
1184 GLuint duuBuffer, BufferDescriptor const &duuDesc,
\n+
1185 GLuint duvBuffer, BufferDescriptor const &duvDesc,
\n+
1186 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1187 int numPatchCoords,
\n+
1188 GLuint patchCoordsBuffer,
\n+
1189 const PatchArrayVector &patchArrays,
\n+
1190 GLuint patchIndexBuffer,
\n+
1191 GLuint patchParamsBuffer) const;
\n+
1192
\n+
1225 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1226 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1227 static bool EvalPatchesVarying(
\n+
1228 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1229 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1230 int numPatchCoords,
\n+
1231 PATCHCOORD_BUFFER *patchCoords,
\n+
1232 PATCH_TABLE *patchTable,
\n+
1233 GLXFBEvaluator const *instance,
\n+
1234 void * deviceContext = NULL) {
\n+
1235
\n+
1236 if (instance) {
\n+
1237 return instance->EvalPatchesVarying(
\n+
1238 srcBuffer, srcDesc,
\n+
1239 dstBuffer, dstDesc,
\n+
1240 numPatchCoords, patchCoords,
\n+
1241 patchTable);
\n+
1242 } else {
\n+
1243 // Create an instance on demand (slow)
\n+
1244 (void)deviceContext; // unused
\n+
1245 instance = Create(srcDesc, dstDesc,
\n+
1246 BufferDescriptor(),
\n+
1247 BufferDescriptor());
\n+
1248 if (instance) {
\n+
1249 bool r = instance->EvalPatchesVarying(
\n+
1250 srcBuffer, srcDesc,
\n+
1251 dstBuffer, dstDesc,
\n+
1252 numPatchCoords, patchCoords,
\n+
1253 patchTable);
\n+
1254 delete instance;
\n+
1255 return r;
\n+
1256 }
\n+
1257 return false;
\n+
1258 }
\n+
1259 }
\n+
1260
\n+
1285 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1286 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1287 bool EvalPatchesVarying(
\n+
1288 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1289 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1290 int numPatchCoords,
\n+
1291 PATCHCOORD_BUFFER *patchCoords,
\n+
1292 PATCH_TABLE *patchTable) const {
\n+
1293
\n+
1294 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1295 dstBuffer->BindVBO(), dstDesc,
\n+
1296 0, BufferDescriptor(),
\n+
1297 0, BufferDescriptor(),
\n+
1298 numPatchCoords,
\n+
1299 patchCoords->BindVBO(),
\n+
1300 patchTable->GetVaryingPatchArrays(),
\n+
1301 patchTable->GetVaryingPatchIndexTextureBuffer(),
\n+
1302 patchTable->GetPatchParamTextureBuffer());
\n+
1303 }
\n+
1304
\n+
1349 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1350 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1351 static bool EvalPatchesVarying(
\n+
1352 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1353 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1354 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1355 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1356 int numPatchCoords,
\n+
1357 PATCHCOORD_BUFFER *patchCoords,
\n+
1358 PATCH_TABLE *patchTable,
\n+
1359 GLXFBEvaluator const *instance,
\n+
1360 void * deviceContext = NULL) {
\n+
1361
\n+
1362 if (instance) {
\n+
1363 return instance->EvalPatchesVarying(
\n+
1364 srcBuffer, srcDesc,
\n+
1365 dstBuffer, dstDesc,
\n+
1366 duBuffer, duDesc,
\n+
1367 dvBuffer, dvDesc,
\n+
1368 numPatchCoords, patchCoords,
\n+
1369 patchTable);
\n+
1370 } else {
\n+
1371 // Create an instance on demand (slow)
\n+
1372 (void)deviceContext; // unused
\n+
1373 instance = Create(srcDesc, dstDesc,
\n+
1374 duDesc, dvDesc);
\n+
1375 if (instance) {
\n+
1376 bool r = instance->EvalPatchesVarying(
\n+
1377 srcBuffer, srcDesc,
\n+
1378 dstBuffer, dstDesc,
\n+
1379 duBuffer, duDesc,
\n+
1380 dvBuffer, dvDesc,
\n+
1381 numPatchCoords, patchCoords,
\n+
1382 patchTable);
\n+
1383 delete instance;
\n+
1384 return r;
\n+
1385 }
\n+
1386 return false;
\n+
1387 }
\n+
1388 }
\n+
1389
\n+
1426 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1427 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1428 bool EvalPatchesVarying(
\n+
1429 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1430 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1431 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1432 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1433 int numPatchCoords,
\n+
1434 PATCHCOORD_BUFFER *patchCoords,
\n+
1435 PATCH_TABLE *patchTable) const {
\n+
1436
\n+
1437 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1438 dstBuffer->BindVBO(), dstDesc,
\n+
1439 duBuffer->BindVBO(), duDesc,
\n+
1440 dvBuffer->BindVBO(), dvDesc,
\n+
1441 numPatchCoords,
\n+
1442 patchCoords->BindVBO(),
\n+
1443 patchTable->GetVaryingPatchArrays(),
\n+
1444 patchTable->GetVaryingPatchIndexTextureBuffer(),
\n+
1445 patchTable->GetPatchParamTextureBuffer());
\n+
1446 }
\n+
1447
\n+
1510 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1511 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1512 static bool EvalPatchesVarying(
\n+
1513 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1514 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1515 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1516 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1517 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1518 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1519 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1520 int numPatchCoords,
\n+
1521 PATCHCOORD_BUFFER *patchCoords,
\n+
1522 PATCH_TABLE *patchTable,
\n+
1523 GLXFBEvaluator const *instance,
\n+
1524 void * deviceContext = NULL) {
\n+
1525
\n+
1526 if (instance) {
\n+
1527 return instance->EvalPatchesVarying(
\n+
1528 srcBuffer, srcDesc,
\n+
1529 dstBuffer, dstDesc,
\n+
1530 duBuffer, duDesc,
\n+
1531 dvBuffer, dvDesc,
\n+
1532 duuBuffer, duuDesc,
\n+
1533 duvBuffer, duvDesc,
\n+
1534 dvvBuffer, dvvDesc,
\n+
1535 numPatchCoords, patchCoords,
\n+
1536 patchTable);
\n+
1537 } else {
\n+
1538 // Create an instance on demand (slow)
\n+
1539 (void)deviceContext; // unused
\n+
1540 instance = Create(srcDesc, dstDesc,
\n+
1541 duDesc, dvDesc,
\n+
1542 duuDesc, duvDesc, dvvDesc);
\n+
1543 if (instance) {
\n+
1544 bool r = instance->EvalPatchesVarying(
\n+
1545 srcBuffer, srcDesc,
\n+
1546 dstBuffer, dstDesc,
\n+
1547 duBuffer, duDesc,
\n+
1548 dvBuffer, dvDesc,
\n+
1549 duuBuffer, duuDesc,
\n+
1550 duvBuffer, duvDesc,
\n+
1551 dvvBuffer, dvvDesc,
\n+
1552 numPatchCoords, patchCoords,
\n+
1553 patchTable);
\n+
1554 delete instance;
\n+
1555 return r;
\n+
1556 }
\n+
1557 return false;
\n+
1558 }
\n+
1559 }
\n+
1560
\n+
1615 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1616 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1617 bool EvalPatchesVarying(
\n+
1618 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1619 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1620 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1621 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1622 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1623 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1624 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1625 int numPatchCoords,
\n+
1626 PATCHCOORD_BUFFER *patchCoords,
\n+
1627 PATCH_TABLE *patchTable) const {
\n+
1628
\n+
1629 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1630 dstBuffer->BindVBO(), dstDesc,
\n+
1631 duBuffer->BindVBO(), duDesc,
\n+
1632 dvBuffer->BindVBO(), dvDesc,
\n+
1633 duuBuffer->BindVBO(), duuDesc,
\n+
1634 duvBuffer->BindVBO(), duvDesc,
\n+
1635 dvvBuffer->BindVBO(), dvvDesc,
\n+
1636 numPatchCoords,
\n+
1637 patchCoords->BindVBO(),
\n+
1638 patchTable->GetVaryingPatchArrays(),
\n+
1639 patchTable->GetVaryingPatchIndexTextureBuffer(),
\n+
1640 patchTable->GetPatchParamTextureBuffer());
\n+
1641 }
\n+
1642
\n+
1677 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1678 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1679 static bool EvalPatchesFaceVarying(
\n+
1680 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1681 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1682 int numPatchCoords,
\n+
1683 PATCHCOORD_BUFFER *patchCoords,
\n+
1684 PATCH_TABLE *patchTable,
\n+
1685 int fvarChannel,
\n+
1686 GLXFBEvaluator const *instance,
\n+
1687 void * deviceContext = NULL) {
\n+
1688
\n+
1689 if (instance) {
\n+
1690 return instance->EvalPatchesFaceVarying(
\n+
1691 srcBuffer, srcDesc,
\n+
1692 dstBuffer, dstDesc,
\n+
1693 numPatchCoords, patchCoords,
\n+
1694 patchTable, fvarChannel);
\n+
1695 } else {
\n+
1696 // Create an instance on demand (slow)
\n+
1697 (void)deviceContext; // unused
\n+
1698 instance = Create(srcDesc, dstDesc,
\n+
1699 BufferDescriptor(),
\n+
1700 BufferDescriptor());
\n+
1701 if (instance) {
\n+
1702 bool r = instance->EvalPatchesFaceVarying(
\n+
1703 srcBuffer, srcDesc,
\n+
1704 dstBuffer, dstDesc,
\n+
1705 numPatchCoords, patchCoords,
\n+
1706 patchTable, fvarChannel);
\n+
1707 delete instance;
\n+
1708 return r;
\n+
1709 }
\n+
1710 return false;
\n+
1711 }
\n+
1712 }
\n+
1713
\n+
1740 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1741 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1742 bool EvalPatchesFaceVarying(
\n+
1743 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1744 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1745 int numPatchCoords,
\n+
1746 PATCHCOORD_BUFFER *patchCoords,
\n+
1747 PATCH_TABLE *patchTable,
\n+
1748 int fvarChannel = 0) const {
\n+
1749
\n+
1750 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1751 dstBuffer->BindVBO(), dstDesc,
\n+
1752 0, BufferDescriptor(),
\n+
1753 0, BufferDescriptor(),
\n+
1754 numPatchCoords,
\n+
1755 patchCoords->BindVBO(),
\n+
1756 patchTable->GetFVarPatchArrays(fvarChannel),
\n+
1757 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),
\n+
1758 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));
\n+
1759 }
\n+
1760
\n+
1807 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1808 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1809 static bool EvalPatchesFaceVarying(
\n+
1810 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1811 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1812 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1813 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1814 int numPatchCoords,
\n+
1815 PATCHCOORD_BUFFER *patchCoords,
\n+
1816 PATCH_TABLE *patchTable,
\n+
1817 int fvarChannel,
\n+
1818 GLXFBEvaluator const *instance,
\n+
1819 void * deviceContext = NULL) {
\n+
1820
\n+
1821 if (instance) {
\n+
1822 return instance->EvalPatchesFaceVarying(
\n+
1823 srcBuffer, srcDesc,
\n+
1824 dstBuffer, dstDesc,
\n+
1825 duBuffer, duDesc,
\n+
1826 dvBuffer, dvDesc,
\n+
1827 numPatchCoords, patchCoords,
\n+
1828 patchTable, fvarChannel);
\n+
1829 } else {
\n+
1830 // Create an instance on demand (slow)
\n+
1831 (void)deviceContext; // unused
\n+
1832 instance = Create(srcDesc, dstDesc,
\n+
1833 duDesc, dvDesc);
\n+
1834 if (instance) {
\n+
1835 bool r = instance->EvalPatchesFaceVarying(
\n+
1836 srcBuffer, srcDesc,
\n+
1837 dstBuffer, dstDesc,
\n+
1838 duBuffer, duDesc,
\n+
1839 dvBuffer, dvDesc,
\n+
1840 numPatchCoords, patchCoords,
\n+
1841 patchTable, fvarChannel);
\n+
1842 delete instance;
\n+
1843 return r;
\n+
1844 }
\n+
1845 return false;
\n+
1846 }
\n+
1847 }
\n+
1848
\n+
1887 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1888 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1889 bool EvalPatchesFaceVarying(
\n+
1890 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1891 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1892 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1893 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1894 int numPatchCoords,
\n+
1895 PATCHCOORD_BUFFER *patchCoords,
\n+
1896 PATCH_TABLE *patchTable,
\n+
1897 int fvarChannel = 0) const {
\n+
1898
\n+
1899 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1900 dstBuffer->BindVBO(), dstDesc,
\n+
1901 duBuffer->BindVBO(), duDesc,
\n+
1902 dvBuffer->BindVBO(), dvDesc,
\n+
1903 numPatchCoords,
\n+
1904 patchCoords->BindVBO(),
\n+
1905 patchTable->GetFVarPatchArrays(fvarChannel),
\n+
1906 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),
\n+
1907 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));
\n+
1908 }
\n+
1909
\n+
1974 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1975 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1976 static bool EvalPatchesFaceVarying(
\n+
1977 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1978 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1979 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1980 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1981 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1982 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1983 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1984 int numPatchCoords,
\n+
1985 PATCHCOORD_BUFFER *patchCoords,
\n+
1986 PATCH_TABLE *patchTable,
\n+
1987 int fvarChannel,
\n+
1988 GLXFBEvaluator const *instance,
\n+
1989 void * deviceContext = NULL) {
\n+
1990
\n+
1991 if (instance) {
\n+
1992 return instance->EvalPatchesFaceVarying(
\n+
1993 srcBuffer, srcDesc,
\n+
1994 dstBuffer, dstDesc,
\n+
1995 duBuffer, duDesc,
\n+
1996 dvBuffer, dvDesc,
\n+
1997 duuBuffer, duuDesc,
\n+
1998 duvBuffer, duvDesc,
\n+
1999 dvvBuffer, dvvDesc,
\n+
2000 numPatchCoords, patchCoords,
\n+
2001 patchTable, fvarChannel);
\n+
2002 } else {
\n+
2003 // Create an instance on demand (slow)
\n+
2004 (void)deviceContext; // unused
\n+
2005 instance = Create(srcDesc, dstDesc,
\n+
2006 duDesc, dvDesc,
\n+
2007 duuDesc, duvDesc, dvvDesc);
\n+
2008 if (instance) {
\n+
2009 bool r = instance->EvalPatchesFaceVarying(
\n+
2010 srcBuffer, srcDesc,
\n+
2011 dstBuffer, dstDesc,
\n+
2012 duBuffer, duDesc,
\n+
2013 dvBuffer, dvDesc,
\n+
2014 duuBuffer, duuDesc,
\n+
2015 duvBuffer, duvDesc,
\n+
2016 dvvBuffer, dvvDesc,
\n+
2017 numPatchCoords, patchCoords,
\n+
2018 patchTable, fvarChannel);
\n+
2019 delete instance;
\n+
2020 return r;
\n+
2021 }
\n+
2022 return false;
\n+
2023 }
\n+
2024 }
\n+
2025
\n+
2082 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
2083 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
2084 bool EvalPatchesFaceVarying(
\n+
2085 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
2086 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
2087 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
2088 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
2089 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
2090 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
2091 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
2092 int numPatchCoords,
\n+
2093 PATCHCOORD_BUFFER *patchCoords,
\n+
2094 PATCH_TABLE *patchTable,
\n+
2095 int fvarChannel = 0) const {
\n+
2096
\n+
2097 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
2098 dstBuffer->BindVBO(), dstDesc,
\n+
2099 duBuffer->BindVBO(), duDesc,
\n+
2100 dvBuffer->BindVBO(), dvDesc,
\n+
2101 duuBuffer->BindVBO(), duuDesc,
\n+
2102 duvBuffer->BindVBO(), duvDesc,
\n+
2103 dvvBuffer->BindVBO(), dvvDesc,
\n+
2104 numPatchCoords,
\n+
2105 patchCoords->BindVBO(),
\n+
2106 patchTable->GetFVarPatchArrays(fvarChannel),
\n+
2107 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),
\n+
2108 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));
\n+
2109 }
\n+
2110
\n+
2116
\n+
2119 bool Compile(BufferDescriptor const &srcDesc,
\n+
2120 BufferDescriptor const &dstDesc,
\n+
2121 BufferDescriptor const &duDesc = BufferDescriptor(),
\n+
2122 BufferDescriptor const &dvDesc = BufferDescriptor(),
\n+
2123 BufferDescriptor const &duuDesc = BufferDescriptor(),
\n+
2124 BufferDescriptor const &duvDesc = BufferDescriptor(),
\n+
2125 BufferDescriptor const &dvvDesc = BufferDescriptor());
\n+
2126
\n+
2128 static void Synchronize(void *kernel);
\n+
2129
\n+
2130private:
\n+
2131 GLuint _srcBufferTexture;
\n+
2132 GLuint _patchArraysUBO;
\n+
2133 bool _interleavedDerivativeBuffers;
\n+
2134
\n+
2135 struct _StencilKernel {
\n+
2136 _StencilKernel();
\n+
2137 ~_StencilKernel();
\n+
2138 bool Compile(BufferDescriptor const &srcDesc,
\n+
2139 BufferDescriptor const &dstDesc,
\n+
2140 BufferDescriptor const &duDesc,
\n+
2141 BufferDescriptor const &dvDesc,
\n+
2142 BufferDescriptor const &duuDesc,
\n+
2143 BufferDescriptor const &duvDesc,
\n+
2144 BufferDescriptor const &dvvDesc,
\n+
2145 bool interleavedDerivativeBuffers);
\n+
2146 GLuint program;
\n+
2147 GLint uniformSrcBufferTexture;
\n+
2148 GLint uniformSrcOffset; // src buffer offset (in elements)
\n+
2149
\n+
2150 GLint uniformSizesTexture;
\n+
2151 GLint uniformOffsetsTexture;
\n+
2152 GLint uniformIndicesTexture;
\n+
2153 GLint uniformWeightsTexture;
\n+
2154 GLint uniformDuWeightsTexture;
\n+
2155 GLint uniformDvWeightsTexture;
\n+
2156 GLint uniformDuuWeightsTexture;
\n+
2157 GLint uniformDuvWeightsTexture;
\n+
2158 GLint uniformDvvWeightsTexture;
\n+
2159 GLint uniformStart; // range
\n+
2160 GLint uniformEnd;
\n+
2161 } _stencilKernel;
\n+
2162
\n+
2163 struct _PatchKernel {
\n+
2164 _PatchKernel();
\n+
2165 ~_PatchKernel();
\n+
2166 bool Compile(BufferDescriptor const &srcDesc,
\n+
2167 BufferDescriptor const &dstDesc,
\n+
2168 BufferDescriptor const &duDesc,
\n+
2169 BufferDescriptor const &dvDesc,
\n+
2170 BufferDescriptor const &duuDesc,
\n+
2171 BufferDescriptor const &duvDesc,
\n+
2172 BufferDescriptor const &dvvDesc,
\n+
2173 bool interleavedDerivativeBuffers);
\n+
2174 GLuint program;
\n+
2175 GLint uniformSrcBufferTexture;
\n+
2176 GLint uniformSrcOffset; // src buffer offset (in elements)
\n+
2177
\n+
2178 GLint uniformPatchArraysUBOBinding;
\n+
2179 GLint uniformPatchParamTexture;
\n+
2180 GLint uniformPatchIndexTexture;
\n+
2181 } _patchKernel;
\n+
2182
\n+
2183};
\n+
2184
\n+
2185} // end namespace Osd
\n+
2186
\n+
2187} // end namespace OPENSUBDIV_VERSION
\n+
2188using namespace OPENSUBDIV_VERSION;
\n+
2189
\n+
2190} // end namespace OpenSubdiv
\n+
2191
\n+
2192
\n+
2193#endif // OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::Create
static D3D11PatchTable * Create(Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context)
Definition: d3d11PatchTable.h:57
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchArrays
PatchArrayVector const & GetPatchArrays() const
Definition: d3d11PatchTable.h:65
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::allocate
bool allocate(Far::PatchTable const *farPatchTable, ID3D11DeviceContext *deviceContext)
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchIndexBuffer
ID3D11Buffer * GetPatchIndexBuffer() const
Returns the index buffer containing the patch control vertices.
Definition: d3d11PatchTable.h:70
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::Create
static D3D11PatchTable * Create(Far::PatchTable const *farPatchTable, ID3D11DeviceContext *deviceContext)
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchParamSRV
ID3D11ShaderResourceView * GetPatchParamSRV() const
Returns the SRV containing the patch parameter.
Definition: d3d11PatchTable.h:75
\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::Create
static GLStencilTableTBO * Create(Far::StencilTable const *stencilTable, void *deviceContext=NULL)
Definition: glXFBEvaluator.h:54
\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GLStencilTableTBO
GLStencilTableTBO(Far::StencilTable const *stencilTable)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GLStencilTableTBO
GLStencilTableTBO(Far::LimitStencilTable const *limitStencilTable)
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::Create
static GLStencilTableTBO * Create(Far::LimitStencilTable const *limitStencilTable, void *deviceContext=NULL)
Definition: glXFBEvaluator.h:60
\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1287
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, bool interleavedDerivativeBuffers=false)
Definition: glXFBEvaluator.h:126
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glXFBEvaluator.h:506
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::GLXFBEvaluator
GLXFBEvaluator(bool interleavedDerivativeBuffers=false)
Constructor.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glXFBEvaluator.h:575
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays, GLuint patchIndexBuffer, GLuint patchParamsBuffer) const
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glXFBEvaluator.h:396
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: glXFBEvaluator.h:1144
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, bool interleavedDerivativeBuffers=false)
Definition: glXFBEvaluator.h:170
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, void *deviceContext)
Specialization to allow creation without a device context.
Definition: glXFBEvaluator.h:117
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glXFBEvaluator.h:306
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1351
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, GLuint sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer, GLuint duWeightsBuffer, GLuint dvWeightsBuffer, GLuint duuWeightsBuffer, GLuint duvWeightsBuffer, GLuint dvvWeightsBuffer, int start, int end) const
Dispatch the GLSL XFB kernel on on GPU asynchronously returns false if the kernel hasn't been compile...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:843
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, DEVICE_CONTEXT deviceContext)
Generic creator template.
Definition: glXFBEvaluator.h:140
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, DEVICE_CONTEXT deviceContext)
Generic creator template.
Definition: glXFBEvaluator.h:104
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Synchronize
static void Synchronize(void *kernel)
Wait the dispatched kernel finishes.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1227
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1617
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1889
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:765
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1809
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1016
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:943
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: glXFBEvaluator.h:1070
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glXFBEvaluator.h:457
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays, GLuint patchIndexBuffer, GLuint patchParamsBuffer) const
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, void *deviceContext)
Specialization to allow creation without a device context.
Definition: glXFBEvaluator.h:157
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:2084
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1512
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer, GLuint duWeightsBuffer, GLuint dvWeightsBuffer, int start, int end) const
Dispatch the GLSL XFB kernel on on GPU asynchronously returns false if the kernel hasn't been compile...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1679
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glXFBEvaluator.h:238
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::~GLXFBEvaluator
~GLXFBEvaluator()
Destructor. note that the GL context must be made current.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Compile
bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc=BufferDescriptor(), BufferDescriptor const &dvDesc=BufferDescriptor(), BufferDescriptor const &duuDesc=BufferDescriptor(), BufferDescriptor const &duvDesc=BufferDescriptor(), BufferDescriptor const &dvvDesc=BufferDescriptor())
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1428
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1742
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1976
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-d3d11PatchTable.h\n+glXFBEvaluator.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,134 +30,1454 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n- 26#define OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n+ 25#ifndef OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H\n+ 26#define OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H\n 27\n 28#include \"../version.h\"\n 29\n- 30#include \n- 31#include \"../far/patchDescriptor.h\"\n- 32#include \"../osd/nonCopyable.h\"\n- 33#include \"../osd/types.h\"\n- 34\n- 35struct ID3D11Buffer;\n- 36struct ID3D11ShaderResourceView;\n- 37struct ID3D11Device;\n- 38struct ID3D11DeviceContext;\n- 39\n- 40namespace OpenSubdiv {\n- 41namespace OPENSUBDIV_VERSION {\n+ 30#include \"../osd/opengl.h\"\n+ 31#include \"../osd/types.h\"\n+ 32#include \"../osd/bufferDescriptor.h\"\n+ 33\n+ 34namespace OpenSubdiv {\n+ 35namespace OPENSUBDIV_VERSION {\n+ 36\n+ 37namespace Far {\n+ 38 class PatchTable;\n+ 39 class StencilTable;\n+ 40 class LimitStencilTable;\n+ 41}\n 42\n- 43namespace Far{\n- 44 class PatchTable;\n- 45};\n- 46\n- 47namespace Osd {\n- 48\n-49class D3D11PatchTable : private NonCopyable {\n- 50public:\n-51 typedef ID3D11Buffer * VertexBufferBinding;\n- 52\n-53 D3D11PatchTable();\n-54 ~D3D11PatchTable();\n- 55\n- 56 template\n-57 static D3D11PatchTable *Create(Far::PatchTable const *farPatchTable,\n- 58 DEVICE_CONTEXT context) {\n- 59 return Create(farPatchTable, context->GetDeviceContext());\n- 60 }\n- 61\n-62 static D3D11PatchTable *Create(Far::PatchTable const *farPatchTable,\n- 63 ID3D11DeviceContext *deviceContext);\n- 64\n-65 PatchArrayVector const &GetPatchArrays() const {\n- 66 return _patchArrays;\n- 67 }\n- 68\n-70 ID3D11Buffer* GetPatchIndexBuffer() const {\n- 71 return _indexBuffer;\n- 72 }\n- 73\n-75 ID3D11ShaderResourceView* GetPatchParamSRV() const {\n- 76 return _patchParamBufferSRV;\n- 77 }\n- 78\n- 79protected:\n- 80 // allocate buffers from patchTable\n-81 bool allocate(Far::PatchTable const *farPatchTable,\n- 82 ID3D11DeviceContext *deviceContext);\n- 83\n-84 PatchArrayVector _patchArrays;\n- 85\n-86 ID3D11Buffer *_indexBuffer;\n-87 ID3D11Buffer *_patchParamBuffer;\n-88 ID3D11ShaderResourceView *_patchParamBufferSRV;\n- 89};\n- 90\n- 91\n- 92} // end namespace Osd\n- 93\n- 94} // end namespace OPENSUBDIV_VERSION\n- 95using namespace OPENSUBDIV_VERSION;\n- 96\n- 97} // end namespace OpenSubdiv\n- 98\n- 99#endif // OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n+ 43namespace Osd {\n+ 44\n+52class GLStencilTableTBO {\n+ 53public:\n+54 static GLStencilTableTBO *Create(\n+ 55 Far::StencilTable const *stencilTable, void *deviceContext = NULL) {\n+ 56 (void)deviceContext; // unused\n+ 57 return new GLStencilTableTBO(stencilTable);\n+ 58 }\n+ 59\n+60 static GLStencilTableTBO *Create(\n+ 61 Far::LimitStencilTable const *limitStencilTable,\n+ 62 void *deviceContext = NULL) {\n+ 63 (void)deviceContext; // unused\n+ 64 return new GLStencilTableTBO(limitStencilTable);\n+ 65 }\n+ 66\n+67 explicit GLStencilTableTBO(Far::StencilTable const *stencilTable);\n+68 explicit GLStencilTableTBO(Far::LimitStencilTable const *limitStencilTable);\n+69 ~GLStencilTableTBO();\n+ 70\n+ 71 // interfaces needed for GLSLTransformFeedbackKernel\n+72 GLuint GetSizesTexture() const { return _sizes; }\n+73 GLuint GetOffsetsTexture() const { return _offsets; }\n+74 GLuint GetIndicesTexture() const { return _indices; }\n+75 GLuint GetWeightsTexture() const { return _weights; }\n+76 GLuint GetDuWeightsTexture() const { return _duWeights; }\n+77 GLuint GetDvWeightsTexture() const { return _dvWeights; }\n+78 GLuint GetDuuWeightsTexture() const { return _duuWeights; }\n+79 GLuint GetDuvWeightsTexture() const { return _duvWeights; }\n+80 GLuint GetDvvWeightsTexture() const { return _dvvWeights; }\n+81 int GetNumStencils() const { return _numStencils; }\n+ 82\n+ 83private:\n+ 84 GLuint _sizes;\n+ 85 GLuint _offsets;\n+ 86 GLuint _indices;\n+ 87 GLuint _weights;\n+ 88 GLuint _duWeights;\n+ 89 GLuint _dvWeights;\n+ 90 GLuint _duuWeights;\n+ 91 GLuint _duvWeights;\n+ 92 GLuint _dvvWeights;\n+ 93 int _numStencils;\n+ 94};\n+ 95\n+ 96// -------------------------------------------------------------------------\n+--\n+ 97\n+98class GLXFBEvaluator {\n+ 99public:\n+100 typedef bool Instantiatable;\n+ 101\n+ 103 template \n+104 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,\n+ 105 BufferDescriptor const &dstDesc,\n+ 106 BufferDescriptor const &duDesc,\n+ 107 BufferDescriptor const &dvDesc,\n+ 108 DEVICE_CONTEXT deviceContext) {\n+ 109 bool interleavedDerivativeBuffers = deviceContext\n+ 110 ? deviceContext->AreInterleavedDerivativeBuffers()\n+ 111 : false;\n+ 112 return Create(srcDesc, dstDesc, duDesc, dvDesc,\n+ 113 interleavedDerivativeBuffers);\n+ 114 }\n+ 115\n+117 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,\n+ 118 BufferDescriptor const &dstDesc,\n+ 119 BufferDescriptor const &duDesc,\n+ 120 BufferDescriptor const &dvDesc,\n+ 121 void * deviceContext) {\n+ 122 (void)deviceContext; // not used\n+ 123 return Create(srcDesc, dstDesc, duDesc, dvDesc);\n+ 124 }\n+ 125\n+126 static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n+ 127 BufferDescriptor const &dstDesc,\n+ 128 BufferDescriptor const &duDesc,\n+ 129 BufferDescriptor const &dvDesc,\n+ 130 bool interleavedDerivativeBuffers = false) {\n+ 131 GLXFBEvaluator *instance = new GLXFBEvaluator\n+(interleavedDerivativeBuffers);\n+ 132 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc))\n+ 133 return instance;\n+ 134 delete instance;\n+ 135 return NULL;\n+ 136 }\n+ 137\n+ 139 template \n+140 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,\n+ 141 BufferDescriptor const &dstDesc,\n+ 142 BufferDescriptor const &duDesc,\n+ 143 BufferDescriptor const &dvDesc,\n+ 144 BufferDescriptor const &duuDesc,\n+ 145 BufferDescriptor const &duvDesc,\n+ 146 BufferDescriptor const &dvvDesc,\n+ 147 DEVICE_CONTEXT deviceContext) {\n+ 148 bool interleavedDerivativeBuffers = deviceContext\n+ 149 ? deviceContext->AreInterleavedDerivativeBuffers()\n+ 150 : false;\n+ 151 return Create(srcDesc, dstDesc, duDesc, dvDesc,\n+ 152 duuDesc, duvDesc, dvvDesc,\n+ 153 interleavedDerivativeBuffers);\n+ 154 }\n+ 155\n+157 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,\n+ 158 BufferDescriptor const &dstDesc,\n+ 159 BufferDescriptor const &duDesc,\n+ 160 BufferDescriptor const &dvDesc,\n+ 161 BufferDescriptor const &duuDesc,\n+ 162 BufferDescriptor const &duvDesc,\n+ 163 BufferDescriptor const &dvvDesc,\n+ 164 void * deviceContext) {\n+ 165 (void)deviceContext; // not used\n+ 166 return Create(srcDesc, dstDesc, duDesc, dvDesc,\n+ 167 duuDesc, duvDesc, dvvDesc);\n+ 168 }\n+ 169\n+170 static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n+ 171 BufferDescriptor const &dstDesc,\n+ 172 BufferDescriptor const &duDesc,\n+ 173 BufferDescriptor const &dvDesc,\n+ 174 BufferDescriptor const &duuDesc,\n+ 175 BufferDescriptor const &duvDesc,\n+ 176 BufferDescriptor const &dvvDesc,\n+ 177 bool interleavedDerivativeBuffers = false) {\n+ 178 GLXFBEvaluator *instance = new GLXFBEvaluator\n+(interleavedDerivativeBuffers);\n+ 179 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc,\n+ 180 duuDesc, duvDesc, dvvDesc))\n+ 181 return instance;\n+ 182 delete instance;\n+ 183 return NULL;\n+ 184 }\n+ 185\n+199 GLXFBEvaluator(bool interleavedDerivativeBuffers = false);\n+ 200\n+202 ~GLXFBEvaluator();\n+ 203\n+ 209\n+ 237 template \n+238 static bool EvalStencils(\n+ 239 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 240 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 241 STENCIL_TABLE const *stencilTable,\n+ 242 GLXFBEvaluator const *instance,\n+ 243 void * deviceContext = NULL) {\n+ 244\n+ 245 if (instance) {\n+ 246 return instance->EvalStencils(srcBuffer, srcDesc,\n+ 247 dstBuffer, dstDesc,\n+ 248 stencilTable);\n+ 249 } else {\n+ 250 // Create an instance on demand (slow)\n+ 251 (void)deviceContext; // unused\n+ 252 instance = Create(srcDesc, dstDesc,\n+ 253 BufferDescriptor(),\n+ 254 BufferDescriptor());\n+ 255 if (instance) {\n+ 256 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 257 dstBuffer, dstDesc,\n+ 258 stencilTable);\n+ 259 delete instance;\n+ 260 return r;\n+ 261 }\n+ 262 return false;\n+ 263 }\n+ 264 }\n+ 265\n+ 305 template \n+306 static bool EvalStencils(\n+ 307 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 308 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 309 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 310 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 311 STENCIL_TABLE const *stencilTable,\n+ 312 GLXFBEvaluator const *instance,\n+ 313 void * deviceContext = NULL) {\n+ 314\n+ 315 if (instance) {\n+ 316 return instance->EvalStencils(srcBuffer, srcDesc,\n+ 317 dstBuffer, dstDesc,\n+ 318 duBuffer, duDesc,\n+ 319 dvBuffer, dvDesc,\n+ 320 stencilTable);\n+ 321 } else {\n+ 322 // Create an instance on demand (slow)\n+ 323 (void)deviceContext; // unused\n+ 324 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);\n+ 325 if (instance) {\n+ 326 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 327 dstBuffer, dstDesc,\n+ 328 duBuffer, duDesc,\n+ 329 dvBuffer, dvDesc,\n+ 330 stencilTable);\n+ 331 delete instance;\n+ 332 return r;\n+ 333 }\n+ 334 return false;\n+ 335 }\n+ 336 }\n+ 337\n+ 395 template \n+396 static bool EvalStencils(\n+ 397 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 398 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 399 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 400 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 401 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 402 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 403 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 404 STENCIL_TABLE const *stencilTable,\n+ 405 GLXFBEvaluator const *instance,\n+ 406 void * deviceContext = NULL) {\n+ 407\n+ 408 if (instance) {\n+ 409 return instance->EvalStencils(srcBuffer, srcDesc,\n+ 410 dstBuffer, dstDesc,\n+ 411 duBuffer, duDesc,\n+ 412 dvBuffer, dvDesc,\n+ 413 duuBuffer, duuDesc,\n+ 414 duvBuffer, duvDesc,\n+ 415 dvvBuffer, dvvDesc,\n+ 416 stencilTable);\n+ 417 } else {\n+ 418 // Create an instance on demand (slow)\n+ 419 (void)deviceContext; // unused\n+ 420 instance = Create(srcDesc, dstDesc,\n+ 421 duDesc, dvDesc,\n+ 422 duuDesc, duvDesc, dvvDesc);\n+ 423 if (instance) {\n+ 424 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 425 dstBuffer, dstDesc,\n+ 426 duBuffer, duDesc,\n+ 427 dvBuffer, dvDesc,\n+ 428 duuBuffer, duuDesc,\n+ 429 duvBuffer, duvDesc,\n+ 430 dvvBuffer, dvvDesc,\n+ 431 stencilTable);\n+ 432 delete instance;\n+ 433 return r;\n+ 434 }\n+ 435 return false;\n+ 436 }\n+ 437 }\n+ 438\n+ 456 template \n+457 bool EvalStencils(\n+ 458 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 459 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 460 STENCIL_TABLE const *stencilTable) const {\n+ 461\n+ 462 return EvalStencils(srcBuffer->BindVBO(), srcDesc,\n+ 463 dstBuffer->BindVBO(), dstDesc,\n+ 464 0, BufferDescriptor(),\n+ 465 0, BufferDescriptor(),\n+ 466 stencilTable->GetSizesTexture(),\n+ 467 stencilTable->GetOffsetsTexture(),\n+ 468 stencilTable->GetIndicesTexture(),\n+ 469 stencilTable->GetWeightsTexture(),\n+ 470 0,\n+ 471 0,\n+ 472 /* start = */ 0,\n+ 473 /* end = */ stencilTable->GetNumStencils());\n+ 474 }\n+ 475\n+ 505 template \n+506 bool EvalStencils(\n+ 507 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 508 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 509 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 510 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 511 STENCIL_TABLE const *stencilTable) const {\n+ 512\n+ 513 return EvalStencils(srcBuffer->BindVBO(), srcDesc,\n+ 514 dstBuffer->BindVBO(), dstDesc,\n+ 515 duBuffer->BindVBO(), duDesc,\n+ 516 dvBuffer->BindVBO(), dvDesc,\n+ 517 stencilTable->GetSizesTexture(),\n+ 518 stencilTable->GetOffsetsTexture(),\n+ 519 stencilTable->GetIndicesTexture(),\n+ 520 stencilTable->GetWeightsTexture(),\n+ 521 stencilTable->GetDuWeightsTexture(),\n+ 522 stencilTable->GetDvWeightsTexture(),\n+ 523 /* start = */ 0,\n+ 524 /* end = */ stencilTable->GetNumStencils());\n+ 525 }\n+ 526\n+ 574 template \n+575 bool EvalStencils(\n+ 576 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 577 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 578 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 579 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 580 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 581 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 582 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 583 STENCIL_TABLE const *stencilTable) const {\n+ 584\n+ 585 return EvalStencils(srcBuffer->BindVBO(), srcDesc,\n+ 586 dstBuffer->BindVBO(), dstDesc,\n+ 587 duBuffer->BindVBO(), duDesc,\n+ 588 dvBuffer->BindVBO(), dvDesc,\n+ 589 duuBuffer->BindVBO(), duuDesc,\n+ 590 duvBuffer->BindVBO(), duvDesc,\n+ 591 dvvBuffer->BindVBO(), dvvDesc,\n+ 592 stencilTable->GetSizesTexture(),\n+ 593 stencilTable->GetOffsetsTexture(),\n+ 594 stencilTable->GetIndicesTexture(),\n+ 595 stencilTable->GetWeightsTexture(),\n+ 596 stencilTable->GetDuWeightsTexture(),\n+ 597 stencilTable->GetDvWeightsTexture(),\n+ 598 stencilTable->GetDuuWeightsTexture(),\n+ 599 stencilTable->GetDuvWeightsTexture(),\n+ 600 stencilTable->GetDvvWeightsTexture(),\n+ 601 /* start = */ 0,\n+ 602 /* end = */ stencilTable->GetNumStencils());\n+ 603 }\n+ 604\n+640 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n+ 641 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n+ 642 GLuint duBuffer, BufferDescriptor const &duDesc,\n+ 643 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n+ 644 GLuint sizesBuffer,\n+ 645 GLuint offsetsBuffer,\n+ 646 GLuint indicesBuffer,\n+ 647 GLuint weightsBuffer,\n+ 648 GLuint duWeightsBuffer,\n+ 649 GLuint dvWeightsBuffer,\n+ 650 int start,\n+ 651 int end) const;\n+ 652\n+706 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n+ 707 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n+ 708 GLuint duBuffer, BufferDescriptor const &duDesc,\n+ 709 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n+ 710 GLuint duuBuffer, BufferDescriptor const &duuDesc,\n+ 711 GLuint duvBuffer, BufferDescriptor const &duvDesc,\n+ 712 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 713 GLuint sizesBuffer,\n+ 714 GLuint offsetsBuffer,\n+ 715 GLuint indicesBuffer,\n+ 716 GLuint weightsBuffer,\n+ 717 GLuint duWeightsBuffer,\n+ 718 GLuint dvWeightsBuffer,\n+ 719 GLuint duuWeightsBuffer,\n+ 720 GLuint duvWeightsBuffer,\n+ 721 GLuint dvvWeightsBuffer,\n+ 722 int start,\n+ 723 int end) const;\n+ 724\n+ 730\n+ 763 template \n+765 static bool EvalPatches(\n+ 766 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 767 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 768 int numPatchCoords,\n+ 769 PATCHCOORD_BUFFER *patchCoords,\n+ 770 PATCH_TABLE *patchTable,\n+ 771 GLXFBEvaluator const *instance,\n+ 772 void * deviceContext = NULL) {\n+ 773\n+ 774 if (instance) {\n+ 775 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 776 dstBuffer, dstDesc,\n+ 777 numPatchCoords, patchCoords,\n+ 778 patchTable);\n+ 779 } else {\n+ 780 // Create an instance on demand (slow)\n+ 781 (void)deviceContext; // unused\n+ 782 instance = Create(srcDesc, dstDesc,\n+ 783 BufferDescriptor(),\n+ 784 BufferDescriptor());\n+ 785 if (instance) {\n+ 786 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 787 dstBuffer, dstDesc,\n+ 788 numPatchCoords, patchCoords,\n+ 789 patchTable);\n+ 790 delete instance;\n+ 791 return r;\n+ 792 }\n+ 793 return false;\n+ 794 }\n+ 795 }\n+ 796\n+ 841 template \n+843 static bool EvalPatches(\n+ 844 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 845 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 846 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 847 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 848 int numPatchCoords,\n+ 849 PATCHCOORD_BUFFER *patchCoords,\n+ 850 PATCH_TABLE *patchTable,\n+ 851 GLXFBEvaluator const *instance,\n+ 852 void * deviceContext = NULL) {\n+ 853\n+ 854 if (instance) {\n+ 855 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 856 dstBuffer, dstDesc,\n+ 857 duBuffer, duDesc,\n+ 858 dvBuffer, dvDesc,\n+ 859 numPatchCoords, patchCoords,\n+ 860 patchTable);\n+ 861 } else {\n+ 862 // Create an instance on demand (slow)\n+ 863 (void)deviceContext; // unused\n+ 864 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);\n+ 865 if (instance) {\n+ 866 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 867 dstBuffer, dstDesc,\n+ 868 duBuffer, duDesc,\n+ 869 dvBuffer, dvDesc,\n+ 870 numPatchCoords, patchCoords,\n+ 871 patchTable);\n+ 872 delete instance;\n+ 873 return r;\n+ 874 }\n+ 875 return false;\n+ 876 }\n+ 877 }\n+ 878\n+ 941 template \n+943 static bool EvalPatches(\n+ 944 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 945 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 946 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 947 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 948 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 949 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 950 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 951 int numPatchCoords,\n+ 952 PATCHCOORD_BUFFER *patchCoords,\n+ 953 PATCH_TABLE *patchTable,\n+ 954 GLXFBEvaluator const *instance,\n+ 955 void * deviceContext = NULL) {\n+ 956\n+ 957 if (instance) {\n+ 958 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 959 dstBuffer, dstDesc,\n+ 960 duBuffer, duDesc,\n+ 961 dvBuffer, dvDesc,\n+ 962 duuBuffer, duuDesc,\n+ 963 duvBuffer, duvDesc,\n+ 964 dvvBuffer, dvvDesc,\n+ 965 numPatchCoords, patchCoords,\n+ 966 patchTable);\n+ 967 } else {\n+ 968 // Create an instance on demand (slow)\n+ 969 (void)deviceContext; // unused\n+ 970 instance = Create(srcDesc, dstDesc,\n+ 971 duDesc, dvDesc,\n+ 972 duuDesc, duvDesc, dvvDesc);\n+ 973 if (instance) {\n+ 974 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 975 dstBuffer, dstDesc,\n+ 976 duBuffer, duDesc,\n+ 977 dvBuffer, dvDesc,\n+ 978 duuBuffer, duuDesc,\n+ 979 duvBuffer, duvDesc,\n+ 980 dvvBuffer, dvvDesc,\n+ 981 numPatchCoords, patchCoords,\n+ 982 patchTable);\n+ 983 delete instance;\n+ 984 return r;\n+ 985 }\n+ 986 return false;\n+ 987 }\n+ 988 }\n+ 989\n+ 1014 template \n+1016 bool EvalPatches(\n+ 1017 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1018 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1019 int numPatchCoords,\n+ 1020 PATCHCOORD_BUFFER *patchCoords,\n+ 1021 PATCH_TABLE *patchTable) const {\n+ 1022\n+ 1023 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1024 dstBuffer->BindVBO(), dstDesc,\n+ 1025 0, BufferDescriptor(),\n+ 1026 0, BufferDescriptor(),\n+ 1027 numPatchCoords,\n+ 1028 patchCoords->BindVBO(),\n+ 1029 patchTable->GetPatchArrays(),\n+ 1030 patchTable->GetPatchIndexTextureBuffer(),\n+ 1031 patchTable->GetPatchParamTextureBuffer());\n+ 1032 }\n+ 1033\n+ 1068 template \n+1070 bool EvalPatches(\n+ 1071 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1072 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1073 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1074 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1075 int numPatchCoords,\n+ 1076 PATCHCOORD_BUFFER *patchCoords,\n+ 1077 PATCH_TABLE *patchTable) const {\n+ 1078\n+ 1079 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1080 dstBuffer->BindVBO(), dstDesc,\n+ 1081 duBuffer->BindVBO(), duDesc,\n+ 1082 dvBuffer->BindVBO(), dvDesc,\n+ 1083 numPatchCoords,\n+ 1084 patchCoords->BindVBO(),\n+ 1085 patchTable->GetPatchArrays(),\n+ 1086 patchTable->GetPatchIndexTextureBuffer(),\n+ 1087 patchTable->GetPatchParamTextureBuffer());\n+ 1088 }\n+ 1089\n+ 1142 template \n+1144 bool EvalPatches(\n+ 1145 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1146 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1147 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1148 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1149 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1150 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1151 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1152 int numPatchCoords,\n+ 1153 PATCHCOORD_BUFFER *patchCoords,\n+ 1154 PATCH_TABLE *patchTable) const {\n+ 1155\n+ 1156 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1157 dstBuffer->BindVBO(), dstDesc,\n+ 1158 duBuffer->BindVBO(), duDesc,\n+ 1159 dvBuffer->BindVBO(), dvDesc,\n+ 1160 duuBuffer->BindVBO(), duuDesc,\n+ 1161 duvBuffer->BindVBO(), duvDesc,\n+ 1162 dvvBuffer->BindVBO(), dvvDesc,\n+ 1163 numPatchCoords,\n+ 1164 patchCoords->BindVBO(),\n+ 1165 patchTable->GetPatchArrays(),\n+ 1166 patchTable->GetPatchIndexTextureBuffer(),\n+ 1167 patchTable->GetPatchParamTextureBuffer());\n+ 1168 }\n+ 1169\n+1170 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n+ 1171 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n+ 1172 GLuint duBuffer, BufferDescriptor const &duDesc,\n+ 1173 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n+ 1174 int numPatchCoords,\n+ 1175 GLuint patchCoordsBuffer,\n+ 1176 const PatchArrayVector &patchArrays,\n+ 1177 GLuint patchIndexBuffer,\n+ 1178 GLuint patchParamsBuffer) const;\n+ 1179\n+1180 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n+ 1181 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n+ 1182 GLuint duBuffer, BufferDescriptor const &duDesc,\n+ 1183 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n+ 1184 GLuint duuBuffer, BufferDescriptor const &duuDesc,\n+ 1185 GLuint duvBuffer, BufferDescriptor const &duvDesc,\n+ 1186 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1187 int numPatchCoords,\n+ 1188 GLuint patchCoordsBuffer,\n+ 1189 const PatchArrayVector &patchArrays,\n+ 1190 GLuint patchIndexBuffer,\n+ 1191 GLuint patchParamsBuffer) const;\n+ 1192\n+ 1225 template \n+1227 static bool EvalPatchesVarying(\n+ 1228 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1229 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1230 int numPatchCoords,\n+ 1231 PATCHCOORD_BUFFER *patchCoords,\n+ 1232 PATCH_TABLE *patchTable,\n+ 1233 GLXFBEvaluator const *instance,\n+ 1234 void * deviceContext = NULL) {\n+ 1235\n+ 1236 if (instance) {\n+ 1237 return instance->EvalPatchesVarying(\n+ 1238 srcBuffer, srcDesc,\n+ 1239 dstBuffer, dstDesc,\n+ 1240 numPatchCoords, patchCoords,\n+ 1241 patchTable);\n+ 1242 } else {\n+ 1243 // Create an instance on demand (slow)\n+ 1244 (void)deviceContext; // unused\n+ 1245 instance = Create(srcDesc, dstDesc,\n+ 1246 BufferDescriptor(),\n+ 1247 BufferDescriptor());\n+ 1248 if (instance) {\n+ 1249 bool r = instance->EvalPatchesVarying(\n+ 1250 srcBuffer, srcDesc,\n+ 1251 dstBuffer, dstDesc,\n+ 1252 numPatchCoords, patchCoords,\n+ 1253 patchTable);\n+ 1254 delete instance;\n+ 1255 return r;\n+ 1256 }\n+ 1257 return false;\n+ 1258 }\n+ 1259 }\n+ 1260\n+ 1285 template \n+1287 bool EvalPatchesVarying(\n+ 1288 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1289 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1290 int numPatchCoords,\n+ 1291 PATCHCOORD_BUFFER *patchCoords,\n+ 1292 PATCH_TABLE *patchTable) const {\n+ 1293\n+ 1294 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1295 dstBuffer->BindVBO(), dstDesc,\n+ 1296 0, BufferDescriptor(),\n+ 1297 0, BufferDescriptor(),\n+ 1298 numPatchCoords,\n+ 1299 patchCoords->BindVBO(),\n+ 1300 patchTable->GetVaryingPatchArrays(),\n+ 1301 patchTable->GetVaryingPatchIndexTextureBuffer(),\n+ 1302 patchTable->GetPatchParamTextureBuffer());\n+ 1303 }\n+ 1304\n+ 1349 template \n+1351 static bool EvalPatchesVarying(\n+ 1352 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1353 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1354 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1355 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1356 int numPatchCoords,\n+ 1357 PATCHCOORD_BUFFER *patchCoords,\n+ 1358 PATCH_TABLE *patchTable,\n+ 1359 GLXFBEvaluator const *instance,\n+ 1360 void * deviceContext = NULL) {\n+ 1361\n+ 1362 if (instance) {\n+ 1363 return instance->EvalPatchesVarying(\n+ 1364 srcBuffer, srcDesc,\n+ 1365 dstBuffer, dstDesc,\n+ 1366 duBuffer, duDesc,\n+ 1367 dvBuffer, dvDesc,\n+ 1368 numPatchCoords, patchCoords,\n+ 1369 patchTable);\n+ 1370 } else {\n+ 1371 // Create an instance on demand (slow)\n+ 1372 (void)deviceContext; // unused\n+ 1373 instance = Create(srcDesc, dstDesc,\n+ 1374 duDesc, dvDesc);\n+ 1375 if (instance) {\n+ 1376 bool r = instance->EvalPatchesVarying(\n+ 1377 srcBuffer, srcDesc,\n+ 1378 dstBuffer, dstDesc,\n+ 1379 duBuffer, duDesc,\n+ 1380 dvBuffer, dvDesc,\n+ 1381 numPatchCoords, patchCoords,\n+ 1382 patchTable);\n+ 1383 delete instance;\n+ 1384 return r;\n+ 1385 }\n+ 1386 return false;\n+ 1387 }\n+ 1388 }\n+ 1389\n+ 1426 template \n+1428 bool EvalPatchesVarying(\n+ 1429 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1430 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1431 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1432 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1433 int numPatchCoords,\n+ 1434 PATCHCOORD_BUFFER *patchCoords,\n+ 1435 PATCH_TABLE *patchTable) const {\n+ 1436\n+ 1437 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1438 dstBuffer->BindVBO(), dstDesc,\n+ 1439 duBuffer->BindVBO(), duDesc,\n+ 1440 dvBuffer->BindVBO(), dvDesc,\n+ 1441 numPatchCoords,\n+ 1442 patchCoords->BindVBO(),\n+ 1443 patchTable->GetVaryingPatchArrays(),\n+ 1444 patchTable->GetVaryingPatchIndexTextureBuffer(),\n+ 1445 patchTable->GetPatchParamTextureBuffer());\n+ 1446 }\n+ 1447\n+ 1510 template \n+1512 static bool EvalPatchesVarying(\n+ 1513 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1514 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1515 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1516 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1517 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1518 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1519 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1520 int numPatchCoords,\n+ 1521 PATCHCOORD_BUFFER *patchCoords,\n+ 1522 PATCH_TABLE *patchTable,\n+ 1523 GLXFBEvaluator const *instance,\n+ 1524 void * deviceContext = NULL) {\n+ 1525\n+ 1526 if (instance) {\n+ 1527 return instance->EvalPatchesVarying(\n+ 1528 srcBuffer, srcDesc,\n+ 1529 dstBuffer, dstDesc,\n+ 1530 duBuffer, duDesc,\n+ 1531 dvBuffer, dvDesc,\n+ 1532 duuBuffer, duuDesc,\n+ 1533 duvBuffer, duvDesc,\n+ 1534 dvvBuffer, dvvDesc,\n+ 1535 numPatchCoords, patchCoords,\n+ 1536 patchTable);\n+ 1537 } else {\n+ 1538 // Create an instance on demand (slow)\n+ 1539 (void)deviceContext; // unused\n+ 1540 instance = Create(srcDesc, dstDesc,\n+ 1541 duDesc, dvDesc,\n+ 1542 duuDesc, duvDesc, dvvDesc);\n+ 1543 if (instance) {\n+ 1544 bool r = instance->EvalPatchesVarying(\n+ 1545 srcBuffer, srcDesc,\n+ 1546 dstBuffer, dstDesc,\n+ 1547 duBuffer, duDesc,\n+ 1548 dvBuffer, dvDesc,\n+ 1549 duuBuffer, duuDesc,\n+ 1550 duvBuffer, duvDesc,\n+ 1551 dvvBuffer, dvvDesc,\n+ 1552 numPatchCoords, patchCoords,\n+ 1553 patchTable);\n+ 1554 delete instance;\n+ 1555 return r;\n+ 1556 }\n+ 1557 return false;\n+ 1558 }\n+ 1559 }\n+ 1560\n+ 1615 template \n+1617 bool EvalPatchesVarying(\n+ 1618 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1619 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1620 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1621 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1622 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1623 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1624 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1625 int numPatchCoords,\n+ 1626 PATCHCOORD_BUFFER *patchCoords,\n+ 1627 PATCH_TABLE *patchTable) const {\n+ 1628\n+ 1629 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1630 dstBuffer->BindVBO(), dstDesc,\n+ 1631 duBuffer->BindVBO(), duDesc,\n+ 1632 dvBuffer->BindVBO(), dvDesc,\n+ 1633 duuBuffer->BindVBO(), duuDesc,\n+ 1634 duvBuffer->BindVBO(), duvDesc,\n+ 1635 dvvBuffer->BindVBO(), dvvDesc,\n+ 1636 numPatchCoords,\n+ 1637 patchCoords->BindVBO(),\n+ 1638 patchTable->GetVaryingPatchArrays(),\n+ 1639 patchTable->GetVaryingPatchIndexTextureBuffer(),\n+ 1640 patchTable->GetPatchParamTextureBuffer());\n+ 1641 }\n+ 1642\n+ 1677 template \n+1679 static bool EvalPatchesFaceVarying(\n+ 1680 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1681 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1682 int numPatchCoords,\n+ 1683 PATCHCOORD_BUFFER *patchCoords,\n+ 1684 PATCH_TABLE *patchTable,\n+ 1685 int fvarChannel,\n+ 1686 GLXFBEvaluator const *instance,\n+ 1687 void * deviceContext = NULL) {\n+ 1688\n+ 1689 if (instance) {\n+ 1690 return instance->EvalPatchesFaceVarying(\n+ 1691 srcBuffer, srcDesc,\n+ 1692 dstBuffer, dstDesc,\n+ 1693 numPatchCoords, patchCoords,\n+ 1694 patchTable, fvarChannel);\n+ 1695 } else {\n+ 1696 // Create an instance on demand (slow)\n+ 1697 (void)deviceContext; // unused\n+ 1698 instance = Create(srcDesc, dstDesc,\n+ 1699 BufferDescriptor(),\n+ 1700 BufferDescriptor());\n+ 1701 if (instance) {\n+ 1702 bool r = instance->EvalPatchesFaceVarying(\n+ 1703 srcBuffer, srcDesc,\n+ 1704 dstBuffer, dstDesc,\n+ 1705 numPatchCoords, patchCoords,\n+ 1706 patchTable, fvarChannel);\n+ 1707 delete instance;\n+ 1708 return r;\n+ 1709 }\n+ 1710 return false;\n+ 1711 }\n+ 1712 }\n+ 1713\n+ 1740 template \n+1742 bool EvalPatchesFaceVarying(\n+ 1743 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1744 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1745 int numPatchCoords,\n+ 1746 PATCHCOORD_BUFFER *patchCoords,\n+ 1747 PATCH_TABLE *patchTable,\n+ 1748 int fvarChannel = 0) const {\n+ 1749\n+ 1750 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1751 dstBuffer->BindVBO(), dstDesc,\n+ 1752 0, BufferDescriptor(),\n+ 1753 0, BufferDescriptor(),\n+ 1754 numPatchCoords,\n+ 1755 patchCoords->BindVBO(),\n+ 1756 patchTable->GetFVarPatchArrays(fvarChannel),\n+ 1757 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),\n+ 1758 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));\n+ 1759 }\n+ 1760\n+ 1807 template \n+1809 static bool EvalPatchesFaceVarying(\n+ 1810 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1811 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1812 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1813 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1814 int numPatchCoords,\n+ 1815 PATCHCOORD_BUFFER *patchCoords,\n+ 1816 PATCH_TABLE *patchTable,\n+ 1817 int fvarChannel,\n+ 1818 GLXFBEvaluator const *instance,\n+ 1819 void * deviceContext = NULL) {\n+ 1820\n+ 1821 if (instance) {\n+ 1822 return instance->EvalPatchesFaceVarying(\n+ 1823 srcBuffer, srcDesc,\n+ 1824 dstBuffer, dstDesc,\n+ 1825 duBuffer, duDesc,\n+ 1826 dvBuffer, dvDesc,\n+ 1827 numPatchCoords, patchCoords,\n+ 1828 patchTable, fvarChannel);\n+ 1829 } else {\n+ 1830 // Create an instance on demand (slow)\n+ 1831 (void)deviceContext; // unused\n+ 1832 instance = Create(srcDesc, dstDesc,\n+ 1833 duDesc, dvDesc);\n+ 1834 if (instance) {\n+ 1835 bool r = instance->EvalPatchesFaceVarying(\n+ 1836 srcBuffer, srcDesc,\n+ 1837 dstBuffer, dstDesc,\n+ 1838 duBuffer, duDesc,\n+ 1839 dvBuffer, dvDesc,\n+ 1840 numPatchCoords, patchCoords,\n+ 1841 patchTable, fvarChannel);\n+ 1842 delete instance;\n+ 1843 return r;\n+ 1844 }\n+ 1845 return false;\n+ 1846 }\n+ 1847 }\n+ 1848\n+ 1887 template \n+1889 bool EvalPatchesFaceVarying(\n+ 1890 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1891 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1892 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1893 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1894 int numPatchCoords,\n+ 1895 PATCHCOORD_BUFFER *patchCoords,\n+ 1896 PATCH_TABLE *patchTable,\n+ 1897 int fvarChannel = 0) const {\n+ 1898\n+ 1899 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1900 dstBuffer->BindVBO(), dstDesc,\n+ 1901 duBuffer->BindVBO(), duDesc,\n+ 1902 dvBuffer->BindVBO(), dvDesc,\n+ 1903 numPatchCoords,\n+ 1904 patchCoords->BindVBO(),\n+ 1905 patchTable->GetFVarPatchArrays(fvarChannel),\n+ 1906 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),\n+ 1907 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));\n+ 1908 }\n+ 1909\n+ 1974 template \n+1976 static bool EvalPatchesFaceVarying(\n+ 1977 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1978 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1979 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1980 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1981 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1982 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1983 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1984 int numPatchCoords,\n+ 1985 PATCHCOORD_BUFFER *patchCoords,\n+ 1986 PATCH_TABLE *patchTable,\n+ 1987 int fvarChannel,\n+ 1988 GLXFBEvaluator const *instance,\n+ 1989 void * deviceContext = NULL) {\n+ 1990\n+ 1991 if (instance) {\n+ 1992 return instance->EvalPatchesFaceVarying(\n+ 1993 srcBuffer, srcDesc,\n+ 1994 dstBuffer, dstDesc,\n+ 1995 duBuffer, duDesc,\n+ 1996 dvBuffer, dvDesc,\n+ 1997 duuBuffer, duuDesc,\n+ 1998 duvBuffer, duvDesc,\n+ 1999 dvvBuffer, dvvDesc,\n+ 2000 numPatchCoords, patchCoords,\n+ 2001 patchTable, fvarChannel);\n+ 2002 } else {\n+ 2003 // Create an instance on demand (slow)\n+ 2004 (void)deviceContext; // unused\n+ 2005 instance = Create(srcDesc, dstDesc,\n+ 2006 duDesc, dvDesc,\n+ 2007 duuDesc, duvDesc, dvvDesc);\n+ 2008 if (instance) {\n+ 2009 bool r = instance->EvalPatchesFaceVarying(\n+ 2010 srcBuffer, srcDesc,\n+ 2011 dstBuffer, dstDesc,\n+ 2012 duBuffer, duDesc,\n+ 2013 dvBuffer, dvDesc,\n+ 2014 duuBuffer, duuDesc,\n+ 2015 duvBuffer, duvDesc,\n+ 2016 dvvBuffer, dvvDesc,\n+ 2017 numPatchCoords, patchCoords,\n+ 2018 patchTable, fvarChannel);\n+ 2019 delete instance;\n+ 2020 return r;\n+ 2021 }\n+ 2022 return false;\n+ 2023 }\n+ 2024 }\n+ 2025\n+ 2082 template \n+2084 bool EvalPatchesFaceVarying(\n+ 2085 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 2086 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 2087 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 2088 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 2089 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 2090 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 2091 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 2092 int numPatchCoords,\n+ 2093 PATCHCOORD_BUFFER *patchCoords,\n+ 2094 PATCH_TABLE *patchTable,\n+ 2095 int fvarChannel = 0) const {\n+ 2096\n+ 2097 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 2098 dstBuffer->BindVBO(), dstDesc,\n+ 2099 duBuffer->BindVBO(), duDesc,\n+ 2100 dvBuffer->BindVBO(), dvDesc,\n+ 2101 duuBuffer->BindVBO(), duuDesc,\n+ 2102 duvBuffer->BindVBO(), duvDesc,\n+ 2103 dvvBuffer->BindVBO(), dvvDesc,\n+ 2104 numPatchCoords,\n+ 2105 patchCoords->BindVBO(),\n+ 2106 patchTable->GetFVarPatchArrays(fvarChannel),\n+ 2107 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),\n+ 2108 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));\n+ 2109 }\n+ 2110\n+ 2116\n+2119 bool Compile(BufferDescriptor const &srcDesc,\n+ 2120 BufferDescriptor const &dstDesc,\n+ 2121 BufferDescriptor const &duDesc = BufferDescriptor(),\n+ 2122 BufferDescriptor const &dvDesc = BufferDescriptor(),\n+ 2123 BufferDescriptor const &duuDesc = BufferDescriptor(),\n+ 2124 BufferDescriptor const &duvDesc = BufferDescriptor(),\n+ 2125 BufferDescriptor const &dvvDesc = BufferDescriptor());\n+ 2126\n+2128 static void Synchronize(void *kernel);\n+ 2129\n+ 2130private:\n+ 2131 GLuint _srcBufferTexture;\n+ 2132 GLuint _patchArraysUBO;\n+ 2133 bool _interleavedDerivativeBuffers;\n+ 2134\n+ 2135 struct _StencilKernel {\n+ 2136 _StencilKernel();\n+ 2137 ~_StencilKernel();\n+ 2138 bool Compile(BufferDescriptor const &srcDesc,\n+ 2139 BufferDescriptor const &dstDesc,\n+ 2140 BufferDescriptor const &duDesc,\n+ 2141 BufferDescriptor const &dvDesc,\n+ 2142 BufferDescriptor const &duuDesc,\n+ 2143 BufferDescriptor const &duvDesc,\n+ 2144 BufferDescriptor const &dvvDesc,\n+ 2145 bool interleavedDerivativeBuffers);\n+ 2146 GLuint program;\n+ 2147 GLint uniformSrcBufferTexture;\n+ 2148 GLint uniformSrcOffset; // src buffer offset (in elements)\n+ 2149\n+ 2150 GLint uniformSizesTexture;\n+ 2151 GLint uniformOffsetsTexture;\n+ 2152 GLint uniformIndicesTexture;\n+ 2153 GLint uniformWeightsTexture;\n+ 2154 GLint uniformDuWeightsTexture;\n+ 2155 GLint uniformDvWeightsTexture;\n+ 2156 GLint uniformDuuWeightsTexture;\n+ 2157 GLint uniformDuvWeightsTexture;\n+ 2158 GLint uniformDvvWeightsTexture;\n+ 2159 GLint uniformStart; // range\n+ 2160 GLint uniformEnd;\n+ 2161 } _stencilKernel;\n+ 2162\n+ 2163 struct _PatchKernel {\n+ 2164 _PatchKernel();\n+ 2165 ~_PatchKernel();\n+ 2166 bool Compile(BufferDescriptor const &srcDesc,\n+ 2167 BufferDescriptor const &dstDesc,\n+ 2168 BufferDescriptor const &duDesc,\n+ 2169 BufferDescriptor const &dvDesc,\n+ 2170 BufferDescriptor const &duuDesc,\n+ 2171 BufferDescriptor const &duvDesc,\n+ 2172 BufferDescriptor const &dvvDesc,\n+ 2173 bool interleavedDerivativeBuffers);\n+ 2174 GLuint program;\n+ 2175 GLint uniformSrcBufferTexture;\n+ 2176 GLint uniformSrcOffset; // src buffer offset (in elements)\n+ 2177\n+ 2178 GLint uniformPatchArraysUBOBinding;\n+ 2179 GLint uniformPatchParamTexture;\n+ 2180 GLint uniformPatchIndexTexture;\n+ 2181 } _patchKernel;\n+ 2182\n+ 2183};\n+ 2184\n+ 2185} // end namespace Osd\n+ 2186\n+ 2187} // end namespace OPENSUBDIV_VERSION\n+ 2188using namespace OPENSUBDIV_VERSION;\n+ 2189\n+ 2190} // end namespace OpenSubdiv\n+ 2191\n+ 2192\n+ 2193#endif // OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H\n OpenSubdiv\n Definition: limits.h:32\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n std::vector< PatchArray > PatchArrayVector\n Definition: types.h:132\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n-Container for arrays of parametric patches.\n-Definition: patchTable.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable\n-Definition: d3d11PatchTable.h:49\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::Create\n-static D3D11PatchTable * Create(Far::PatchTable const *farPatchTable,\n-DEVICE_CONTEXT context)\n-Definition: d3d11PatchTable.h:57\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchArrays\n-PatchArrayVector const & GetPatchArrays() const\n-Definition: d3d11PatchTable.h:65\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::allocate\n-bool allocate(Far::PatchTable const *farPatchTable, ID3D11DeviceContext\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n+Stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:273\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n+Limit stencil table class wrapping the template for compatibility.\n+Definition: stencilTable.h:583\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n+BufferDescriptor is a struct which describes buffer elements in interleaved\n+data buffers....\n+Definition: bufferDescriptor.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO\n+GL TextureBuffer stencil table.\n+Definition: glXFBEvaluator.h:52\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetSizesTexture\n+GLuint GetSizesTexture() const\n+Definition: glXFBEvaluator.h:72\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::Create\n+static GLStencilTableTBO * Create(Far::StencilTable const *stencilTable, void\n+*deviceContext=NULL)\n+Definition: glXFBEvaluator.h:54\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetDvvWeightsTexture\n+GLuint GetDvvWeightsTexture() const\n+Definition: glXFBEvaluator.h:80\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetNumStencils\n+int GetNumStencils() const\n+Definition: glXFBEvaluator.h:81\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetDuvWeightsTexture\n+GLuint GetDuvWeightsTexture() const\n+Definition: glXFBEvaluator.h:79\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetDuuWeightsTexture\n+GLuint GetDuuWeightsTexture() const\n+Definition: glXFBEvaluator.h:78\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetIndicesTexture\n+GLuint GetIndicesTexture() const\n+Definition: glXFBEvaluator.h:74\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::~GLStencilTableTBO\n+~GLStencilTableTBO()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GLStencilTableTBO\n+GLStencilTableTBO(Far::StencilTable const *stencilTable)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GLStencilTableTBO\n+GLStencilTableTBO(Far::LimitStencilTable const *limitStencilTable)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetWeightsTexture\n+GLuint GetWeightsTexture() const\n+Definition: glXFBEvaluator.h:75\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::Create\n+static GLStencilTableTBO * Create(Far::LimitStencilTable const\n+*limitStencilTable, void *deviceContext=NULL)\n+Definition: glXFBEvaluator.h:60\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetDvWeightsTexture\n+GLuint GetDvWeightsTexture() const\n+Definition: glXFBEvaluator.h:77\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetDuWeightsTexture\n+GLuint GetDuWeightsTexture() const\n+Definition: glXFBEvaluator.h:76\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetOffsetsTexture\n+GLuint GetOffsetsTexture() const\n+Definition: glXFBEvaluator.h:73\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator\n+Definition: glXFBEvaluator.h:98\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n+bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1287\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n+static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, bool interleavedDerivativeBuffers=false)\n+Definition: glXFBEvaluator.h:126\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, STENCIL_TABLE const *stencilTable) const\n+Generic stencil function.\n+Definition: glXFBEvaluator.h:506\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::GLXFBEvaluator\n+GLXFBEvaluator(bool interleavedDerivativeBuffers=false)\n+Constructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable) const\n+Generic stencil function.\n+Definition: glXFBEvaluator.h:575\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Instantiatable\n+bool Instantiatable\n+Definition: glXFBEvaluator.h:100\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n+dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n+const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint\n+duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor\n+const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, int\n+numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays,\n+GLuint patchIndexBuffer, GLuint patchParamsBuffer) const\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n+GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+Generic static stencil function. This function has a same signature as other\n+device kernels have so t...\n+Definition: glXFBEvaluator.h:396\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable) const\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: glXFBEvaluator.h:1144\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n+static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n+BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, bool\n+interleavedDerivativeBuffers=false)\n+Definition: glXFBEvaluator.h:170\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n+static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, void *deviceContext)\n+Specialization to allow creation without a device context.\n+Definition: glXFBEvaluator.h:117\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable,\n+GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+Generic static stencil function. This function has a same signature as other\n+device kernels have so t...\n+Definition: glXFBEvaluator.h:306\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void\n+*deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1351\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n+dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n+const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint\n+duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor\n+const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, GLuint\n+sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer,\n+GLuint duWeightsBuffer, GLuint dvWeightsBuffer, GLuint duuWeightsBuffer, GLuint\n+duvWeightsBuffer, GLuint dvvWeightsBuffer, int start, int end) const\n+Dispatch the GLSL XFB kernel on on GPU asynchronously returns false if the\n+kernel hasn't been compile...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:843\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n+static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n+BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc,\n+DEVICE_CONTEXT deviceContext)\n+Generic creator template.\n+Definition: glXFBEvaluator.h:140\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n+static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, DEVICE_CONTEXT deviceContext)\n+Generic creator template.\n+Definition: glXFBEvaluator.h:104\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Synchronize\n+static void Synchronize(void *kernel)\n+Wait the dispatched kernel finishes.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n+GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1227\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n+bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1617\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n+bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1889\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const\n+*instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:765\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLXFBEvaluator const\n+*instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1809\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1016\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void\n+*deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:943\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable) const\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: glXFBEvaluator.h:1070\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const\n+*stencilTable) const\n+Generic stencil function.\n+Definition: glXFBEvaluator.h:457\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n+dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n+const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, int\n+numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays,\n+GLuint patchIndexBuffer, GLuint patchParamsBuffer) const\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n+static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n+BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, void\n *deviceContext)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::VertexBufferBinding\n-ID3D11Buffer * VertexBufferBinding\n-Definition: d3d11PatchTable.h:51\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::~D3D11PatchTable\n-~D3D11PatchTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchIndexBuffer\n-ID3D11Buffer * GetPatchIndexBuffer() const\n-Returns the index buffer containing the patch control vertices.\n-Definition: d3d11PatchTable.h:70\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::_patchArrays\n-PatchArrayVector _patchArrays\n-Definition: d3d11PatchTable.h:84\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::_indexBuffer\n-ID3D11Buffer * _indexBuffer\n-Definition: d3d11PatchTable.h:86\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::Create\n-static D3D11PatchTable * Create(Far::PatchTable const *farPatchTable,\n-ID3D11DeviceContext *deviceContext)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::_patchParamBufferSRV\n-ID3D11ShaderResourceView * _patchParamBufferSRV\n-Definition: d3d11PatchTable.h:88\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::_patchParamBuffer\n-ID3D11Buffer * _patchParamBuffer\n-Definition: d3d11PatchTable.h:87\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::GetPatchParamSRV\n-ID3D11ShaderResourceView * GetPatchParamSRV() const\n-Returns the SRV containing the patch parameter.\n-Definition: d3d11PatchTable.h:75\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::D3D11PatchTable\n-D3D11PatchTable()\n+Specialization to allow creation without a device context.\n+Definition: glXFBEvaluator.h:157\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n+bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0)\n+const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:2084\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const\n+*instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1512\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n+dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n+const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint\n+sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer,\n+GLuint duWeightsBuffer, GLuint dvWeightsBuffer, int start, int end) const\n+Dispatch the GLSL XFB kernel on on GPU asynchronously returns false if the\n+kernel hasn't been compile...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n+fvarChannel, GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1679\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n+const *stencilTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+Generic static stencil function. This function has a same signature as other\n+device kernels have so t...\n+Definition: glXFBEvaluator.h:238\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::~GLXFBEvaluator\n+~GLXFBEvaluator()\n+Destructor. note that the GL context must be made current.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Compile\n+bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc,\n+BufferDescriptor const &duDesc=BufferDescriptor(), BufferDescriptor const\n+&dvDesc=BufferDescriptor(), BufferDescriptor const &duuDesc=BufferDescriptor(),\n+BufferDescriptor const &duvDesc=BufferDescriptor(), BufferDescriptor const\n+&dvvDesc=BufferDescriptor())\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n+bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1428\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n+bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n+fvarChannel=0) const\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1742\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n+GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glXFBEvaluator.h:1976\n * opensubdiv\n * osd\n- * d3d11PatchTable.h\n+ * glXFBEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00848.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00848.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glComputeEvaluator.h File Reference\n+OpenSubdiv: opensubdiv/osd/cpuD3D11VertexBuffer.h File Reference\n \n \n \n \n \n \n \n@@ -83,47 +83,41 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
glComputeEvaluator.h File Reference
\n+
cpuD3D11VertexBuffer.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../osd/opengl.h"
\n-#include "../osd/types.h"
\n-#include "../osd/bufferDescriptor.h"
\n+#include <cstddef>
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n-\n-\n+\n+\n \n

\n Classes

class  GLStencilTableSSBO
 GL stencil table (Shader Storage buffer) More...
 
class  GLComputeEvaluator
class  CpuD3D11VertexBuffer
 Concrete vertex buffer class for Cpu subdivision and DirectX drawing. More...
 
\n \n \n \n \n \n-\n-\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,32 +5,27 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-glComputeEvaluator.h File Reference\n+cpuD3D11VertexBuffer.h File Reference\n #include \"../version.h\"\n-#include \"../osd/opengl.h\"\n-#include \"../osd/types.h\"\n-#include \"../osd/bufferDescriptor.h\"\n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0GLStencilTableSSBO\n-\u00a0 GL stencil table (Shader Storage buffer) More...\n-\u00a0\n-class \u00a0GLComputeEvaluator\n+class \u00a0CpuD3D11VertexBuffer\n+\u00a0 Concrete vertex buffer class for Cpu subdivision and DirectX drawing.\n+ More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n-\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * glComputeEvaluator.h\n+ * cpuD3D11VertexBuffer.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00848.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00848.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,3 @@\n var a00848 = [\n- [\"GLStencilTableSSBO\", \"a01209.html\", \"a01209\"],\n- [\"GLComputeEvaluator\", \"a01213.html\", \"a01213\"]\n+ [\"CpuD3D11VertexBuffer\", \"a01153.html\", \"a01153\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00848_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00848_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glComputeEvaluator.h Source File\n+OpenSubdiv: opensubdiv/osd/cpuD3D11VertexBuffer.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
glComputeEvaluator.h
\n+
cpuD3D11VertexBuffer.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2015 Pixar
\n+
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,1072 +107,87 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H
\n-
26#define OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H
\n+
25#ifndef OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H
\n+
26#define OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H
\n
27
\n
28#include "../version.h"
\n-
29
\n-
30#include "../osd/opengl.h"
\n-
31#include "../osd/types.h"
\n-
32#include "../osd/bufferDescriptor.h"
\n-
33
\n-
34namespace OpenSubdiv {
\n-
35namespace OPENSUBDIV_VERSION {
\n-
36
\n-
37namespace Far {
\n-
38 class PatchTable;
\n-
39 class StencilTable;
\n-
40 class LimitStencilTable;
\n-
41}
\n-
42
\n-
43namespace Osd {
\n-
44
\n-
51class GLStencilTableSSBO {
\n-
52public:
\n-
53 static GLStencilTableSSBO *Create(Far::StencilTable const *stencilTable,
\n-
54 void *deviceContext = NULL) {
\n-
55 (void)deviceContext; // unused
\n-
56 return new GLStencilTableSSBO(stencilTable);
\n-
57 }
\n-
58 static GLStencilTableSSBO *Create(
\n-
59 Far::LimitStencilTable const *limitStencilTable,
\n-
60 void *deviceContext = NULL) {
\n-
61 (void)deviceContext; // unused
\n-
62 return new GLStencilTableSSBO(limitStencilTable);
\n-
63 }
\n+
29#include <cstddef>
\n+
30
\n+
31struct ID3D11Buffer;
\n+
32struct ID3D11Device;
\n+
33struct ID3D11DeviceContext;
\n+
34
\n+
35namespace OpenSubdiv {
\n+
36namespace OPENSUBDIV_VERSION {
\n+
37
\n+
38namespace Osd {
\n+
39
\n+
48class CpuD3D11VertexBuffer {
\n+
49public:
\n+
51 static CpuD3D11VertexBuffer * Create(int numElements, int numVertices,
\n+
52 ID3D11DeviceContext *deviceContext);
\n+
53
\n+
55 virtual ~CpuD3D11VertexBuffer();
\n+
56
\n+
59 void UpdateData(const float *src, int startVertex, int numVertices,
\n+
60 void *deviceContext = NULL);
\n+
61
\n+
63 int GetNumElements() const;
\n
64
\n-
65 explicit GLStencilTableSSBO(Far::StencilTable const *stencilTable);
\n-
66 explicit GLStencilTableSSBO(Far::LimitStencilTable const *limitStencilTable);
\n-
67 ~GLStencilTableSSBO();
\n-
68
\n-
69 // interfaces needed for GLSLComputeKernel
\n-
70 GLuint GetSizesBuffer() const { return _sizes; }
\n-
71 GLuint GetOffsetsBuffer() const { return _offsets; }
\n-
72 GLuint GetIndicesBuffer() const { return _indices; }
\n-
73 GLuint GetWeightsBuffer() const { return _weights; }
\n-
74 GLuint GetDuWeightsBuffer() const { return _duWeights; }
\n-
75 GLuint GetDvWeightsBuffer() const { return _dvWeights; }
\n-
76 GLuint GetDuuWeightsBuffer() const { return _duuWeights; }
\n-
77 GLuint GetDuvWeightsBuffer() const { return _duvWeights; }
\n-
78 GLuint GetDvvWeightsBuffer() const { return _dvvWeights; }
\n-
79 int GetNumStencils() const { return _numStencils; }
\n-
80
\n-
81private:
\n-
82 GLuint _sizes;
\n-
83 GLuint _offsets;
\n-
84 GLuint _indices;
\n-
85 GLuint _weights;
\n-
86 GLuint _duWeights;
\n-
87 GLuint _dvWeights;
\n-
88 GLuint _duuWeights;
\n-
89 GLuint _duvWeights;
\n-
90 GLuint _dvvWeights;
\n-
91 int _numStencils;
\n-
92};
\n+
66 int GetNumVertices() const;
\n+
67
\n+
69 float * BindCpuBuffer();
\n+
70
\n+
72 ID3D11Buffer *BindD3D11Buffer(ID3D11DeviceContext *deviceContext);
\n+
73
\n+
75 ID3D11Buffer *BindVBO(ID3D11DeviceContext *deviceContext) {
\n+
76 return BindD3D11Buffer(deviceContext);
\n+
77 }
\n+
78
\n+
79protected:
\n+
81 CpuD3D11VertexBuffer(int numElements, int numVertices);
\n+
82
\n+
83 bool allocate(ID3D11Device *device);
\n+
84
\n+
85private:
\n+
86 int _numElements;
\n+
87 int _numVertices;
\n+
88 ID3D11Buffer *_d3d11Buffer;
\n+
89 float *_cpuBuffer;
\n+
90};
\n+
91
\n+
92} // end namespace Osd
\n
93
\n-
94// ---------------------------------------------------------------------------
\n-
95
\n-
96class GLComputeEvaluator {
\n-
97public:
\n-
98 typedef bool Instantiatable;
\n-
99 static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n-
100 BufferDescriptor const &dstDesc,
\n-
101 BufferDescriptor const &duDesc,
\n-
102 BufferDescriptor const &dvDesc,
\n-
103 void * deviceContext = NULL) {
\n-
104 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n-
105 BufferDescriptor(),
\n-
106 BufferDescriptor(),
\n-
107 BufferDescriptor(),
\n-
108 deviceContext);
\n-
109 }
\n-
110
\n-
111 static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n-
112 BufferDescriptor const &dstDesc,
\n-
113 BufferDescriptor const &duDesc,
\n-
114 BufferDescriptor const &dvDesc,
\n-
115 BufferDescriptor const &duuDesc,
\n-
116 BufferDescriptor const &duvDesc,
\n-
117 BufferDescriptor const &dvvDesc,
\n-
118 void * deviceContext = NULL) {
\n-
119 (void)deviceContext; // not used
\n-
120 GLComputeEvaluator *instance = new GLComputeEvaluator();
\n-
121 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc,
\n-
122 duuDesc, duvDesc, dvvDesc))
\n-
123 return instance;
\n-
124 delete instance;
\n-
125 return NULL;
\n-
126 }
\n-
127
\n-
129 GLComputeEvaluator();
\n-
130
\n-
132 ~GLComputeEvaluator();
\n-
133
\n-
139
\n-
167 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
168 static bool EvalStencils(
\n-
169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
171 STENCIL_TABLE const *stencilTable,
\n-
172 GLComputeEvaluator const *instance,
\n-
173 void * deviceContext = NULL) {
\n-
174
\n-
175 if (instance) {
\n-
176 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
177 dstBuffer, dstDesc,
\n-
178 stencilTable);
\n-
179 } else {
\n-
180 // Create an instance on demand (slow)
\n-
181 (void)deviceContext; // unused
\n-
182 instance = Create(srcDesc, dstDesc,
\n-
183 BufferDescriptor(),
\n-
184 BufferDescriptor());
\n-
185 if (instance) {
\n-
186 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
187 dstBuffer, dstDesc,
\n-
188 stencilTable);
\n-
189 delete instance;
\n-
190 return r;
\n-
191 }
\n-
192 return false;
\n-
193 }
\n-
194 }
\n-
195
\n-
235 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
236 static bool EvalStencils(
\n-
237 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
238 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
239 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
240 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
241 STENCIL_TABLE const *stencilTable,
\n-
242 GLComputeEvaluator const *instance,
\n-
243 void * deviceContext = NULL) {
\n-
244
\n-
245 if (instance) {
\n-
246 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
247 dstBuffer, dstDesc,
\n-
248 duBuffer, duDesc,
\n-
249 dvBuffer, dvDesc,
\n-
250 stencilTable);
\n-
251 } else {
\n-
252 // Create an instance on demand (slow)
\n-
253 (void)deviceContext; // unused
\n-
254 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);
\n-
255 if (instance) {
\n-
256 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
257 dstBuffer, dstDesc,
\n-
258 duBuffer, duDesc,
\n-
259 dvBuffer, dvDesc,
\n-
260 stencilTable);
\n-
261 delete instance;
\n-
262 return r;
\n-
263 }
\n-
264 return false;
\n-
265 }
\n-
266 }
\n-
267
\n-
325 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
326 static bool EvalStencils(
\n-
327 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
328 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
329 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
330 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
331 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
332 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
333 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
334 STENCIL_TABLE const *stencilTable,
\n-
335 GLComputeEvaluator const *instance,
\n-
336 void * deviceContext = NULL) {
\n-
337
\n-
338 if (instance) {
\n-
339 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
340 dstBuffer, dstDesc,
\n-
341 duBuffer, duDesc,
\n-
342 dvBuffer, dvDesc,
\n-
343 duuBuffer, duuDesc,
\n-
344 duvBuffer, duvDesc,
\n-
345 dvvBuffer, dvvDesc,
\n-
346 stencilTable);
\n-
347 } else {
\n-
348 // Create an instance on demand (slow)
\n-
349 (void)deviceContext; // unused
\n-
350 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,
\n-
351 duuDesc, duvDesc, dvvDesc);
\n-
352 if (instance) {
\n-
353 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
354 dstBuffer, dstDesc,
\n-
355 duBuffer, duDesc,
\n-
356 dvBuffer, dvDesc,
\n-
357 duuBuffer, duuDesc,
\n-
358 duvBuffer, duvDesc,
\n-
359 dvvBuffer, dvvDesc,
\n-
360 stencilTable);
\n-
361 delete instance;
\n-
362 return r;
\n-
363 }
\n-
364 return false;
\n-
365 }
\n-
366 }
\n-
367
\n-
385 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
386 bool EvalStencils(
\n-
387 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
388 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
389 STENCIL_TABLE const *stencilTable) const {
\n-
390 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n-
391 dstBuffer->BindVBO(), dstDesc,
\n-
392 0, BufferDescriptor(),
\n-
393 0, BufferDescriptor(),
\n-
394 stencilTable->GetSizesBuffer(),
\n-
395 stencilTable->GetOffsetsBuffer(),
\n-
396 stencilTable->GetIndicesBuffer(),
\n-
397 stencilTable->GetWeightsBuffer(),
\n-
398 0,
\n-
399 0,
\n-
400 /* start = */ 0,
\n-
401 /* end = */ stencilTable->GetNumStencils());
\n-
402 }
\n-
403
\n-
433 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
434 bool EvalStencils(
\n-
435 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
436 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
437 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
438 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
439 STENCIL_TABLE const *stencilTable) const {
\n-
440 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n-
441 dstBuffer->BindVBO(), dstDesc,
\n-
442 duBuffer->BindVBO(), duDesc,
\n-
443 dvBuffer->BindVBO(), dvDesc,
\n-
444 stencilTable->GetSizesBuffer(),
\n-
445 stencilTable->GetOffsetsBuffer(),
\n-
446 stencilTable->GetIndicesBuffer(),
\n-
447 stencilTable->GetWeightsBuffer(),
\n-
448 stencilTable->GetDuWeightsBuffer(),
\n-
449 stencilTable->GetDvWeightsBuffer(),
\n-
450 /* start = */ 0,
\n-
451 /* end = */ stencilTable->GetNumStencils());
\n-
452 }
\n-
453
\n-
501 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
502 bool EvalStencils(
\n-
503 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
504 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
505 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
506 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
507 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
508 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
509 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
510 STENCIL_TABLE const *stencilTable) const {
\n-
511 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n-
512 dstBuffer->BindVBO(), dstDesc,
\n-
513 duBuffer->BindVBO(), duDesc,
\n-
514 dvBuffer->BindVBO(), dvDesc,
\n-
515 duuBuffer->BindVBO(), duuDesc,
\n-
516 duvBuffer->BindVBO(), duvDesc,
\n-
517 dvvBuffer->BindVBO(), dvvDesc,
\n-
518 stencilTable->GetSizesBuffer(),
\n-
519 stencilTable->GetOffsetsBuffer(),
\n-
520 stencilTable->GetIndicesBuffer(),
\n-
521 stencilTable->GetWeightsBuffer(),
\n-
522 stencilTable->GetDuWeightsBuffer(),
\n-
523 stencilTable->GetDvWeightsBuffer(),
\n-
524 stencilTable->GetDuuWeightsBuffer(),
\n-
525 stencilTable->GetDuvWeightsBuffer(),
\n-
526 stencilTable->GetDvvWeightsBuffer(),
\n-
527 /* start = */ 0,
\n-
528 /* end = */ stencilTable->GetNumStencils());
\n-
529 }
\n-
530
\n-
566 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n-
567 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n-
568 GLuint duBuffer, BufferDescriptor const &duDesc,
\n-
569 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n-
570 GLuint sizesBuffer,
\n-
571 GLuint offsetsBuffer,
\n-
572 GLuint indicesBuffer,
\n-
573 GLuint weightsBuffer,
\n-
574 GLuint duWeightsBuffer,
\n-
575 GLuint dvWeightsBuffer,
\n-
576 int start,
\n-
577 int end) const;
\n-
578
\n-
632 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n-
633 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n-
634 GLuint duBuffer, BufferDescriptor const &duDesc,
\n-
635 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n-
636 GLuint duuBuffer, BufferDescriptor const &duuDesc,
\n-
637 GLuint duvBuffer, BufferDescriptor const &duvDesc,
\n-
638 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
639 GLuint sizesBuffer,
\n-
640 GLuint offsetsBuffer,
\n-
641 GLuint indicesBuffer,
\n-
642 GLuint weightsBuffer,
\n-
643 GLuint duWeightsBuffer,
\n-
644 GLuint dvWeightsBuffer,
\n-
645 GLuint duuWeightsBuffer,
\n-
646 GLuint duvWeightsBuffer,
\n-
647 GLuint dvvWeightsBuffer,
\n-
648 int start,
\n-
649 int end) const;
\n-
650
\n-
656
\n-
689 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
690 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
691 static bool EvalPatches(
\n-
692 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
693 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
694 int numPatchCoords,
\n-
695 PATCHCOORD_BUFFER *patchCoords,
\n-
696 PATCH_TABLE *patchTable,
\n-
697 GLComputeEvaluator const *instance,
\n-
698 void * deviceContext = NULL) {
\n-
699
\n-
700 if (instance) {
\n-
701 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
702 dstBuffer, dstDesc,
\n-
703 numPatchCoords, patchCoords,
\n-
704 patchTable);
\n-
705 } else {
\n-
706 // Create an instance on demand (slow)
\n-
707 (void)deviceContext; // unused
\n-
708 instance = Create(srcDesc, dstDesc,
\n-
709 BufferDescriptor(),
\n-
710 BufferDescriptor());
\n-
711 if (instance) {
\n-
712 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
713 dstBuffer, dstDesc,
\n-
714 numPatchCoords, patchCoords,
\n-
715 patchTable);
\n-
716 delete instance;
\n-
717 return r;
\n-
718 }
\n-
719 return false;
\n-
720 }
\n-
721 }
\n-
722
\n-
767 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
768 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
769 static bool EvalPatches(
\n-
770 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
771 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
772 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
773 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
774 int numPatchCoords,
\n-
775 PATCHCOORD_BUFFER *patchCoords,
\n-
776 PATCH_TABLE *patchTable,
\n-
777 GLComputeEvaluator const *instance,
\n-
778 void * deviceContext = NULL) {
\n-
779
\n-
780 if (instance) {
\n-
781 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
782 dstBuffer, dstDesc,
\n-
783 duBuffer, duDesc,
\n-
784 dvBuffer, dvDesc,
\n-
785 numPatchCoords, patchCoords,
\n-
786 patchTable);
\n-
787 } else {
\n-
788 // Create an instance on demand (slow)
\n-
789 (void)deviceContext; // unused
\n-
790 instance = Create(srcDesc, dstDesc,
\n-
791 duDesc, dvDesc);
\n-
792 if (instance) {
\n-
793 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
794 dstBuffer, dstDesc,
\n-
795 duBuffer, duDesc,
\n-
796 dvBuffer, dvDesc,
\n-
797 numPatchCoords, patchCoords,
\n-
798 patchTable);
\n-
799 delete instance;
\n-
800 return r;
\n-
801 }
\n-
802 return false;
\n-
803 }
\n-
804 }
\n-
805
\n-
868 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
869 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
870 static bool EvalPatches(
\n-
871 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
872 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
873 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
874 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
875 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
876 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
877 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
878 int numPatchCoords,
\n-
879 PATCHCOORD_BUFFER *patchCoords,
\n-
880 PATCH_TABLE *patchTable,
\n-
881 GLComputeEvaluator const *instance,
\n-
882 void * deviceContext = NULL) {
\n-
883
\n-
884 if (instance) {
\n-
885 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
886 dstBuffer, dstDesc,
\n-
887 duBuffer, duDesc,
\n-
888 dvBuffer, dvDesc,
\n-
889 duuBuffer, duuDesc,
\n-
890 duvBuffer, duvDesc,
\n-
891 dvvBuffer, dvvDesc,
\n-
892 numPatchCoords, patchCoords,
\n-
893 patchTable);
\n-
894 } else {
\n-
895 // Create an instance on demand (slow)
\n-
896 (void)deviceContext; // unused
\n-
897 instance = Create(srcDesc, dstDesc,
\n-
898 duDesc, dvDesc,
\n-
899 duuDesc, duvDesc, dvvDesc);
\n-
900 if (instance) {
\n-
901 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
902 dstBuffer, dstDesc,
\n-
903 duBuffer, duDesc,
\n-
904 dvBuffer, dvDesc,
\n-
905 duuBuffer, duuDesc,
\n-
906 duvBuffer, duvDesc,
\n-
907 dvvBuffer, dvvDesc,
\n-
908 numPatchCoords, patchCoords,
\n-
909 patchTable);
\n-
910 delete instance;
\n-
911 return r;
\n-
912 }
\n-
913 return false;
\n-
914 }
\n-
915 }
\n-
916
\n-
941 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
942 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
943 bool EvalPatches(
\n-
944 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
945 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
946 int numPatchCoords,
\n-
947 PATCHCOORD_BUFFER *patchCoords,
\n-
948 PATCH_TABLE *patchTable) const {
\n-
949
\n-
950 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
951 dstBuffer->BindVBO(), dstDesc,
\n-
952 0, BufferDescriptor(),
\n-
953 0, BufferDescriptor(),
\n-
954 numPatchCoords,
\n-
955 patchCoords->BindVBO(),
\n-
956 patchTable->GetPatchArrays(),
\n-
957 patchTable->GetPatchIndexBuffer(),
\n-
958 patchTable->GetPatchParamBuffer());
\n-
959 }
\n-
960
\n-
995 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
996 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
997 bool EvalPatches(
\n-
998 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
999 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1000 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1001 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1002 int numPatchCoords,
\n-
1003 PATCHCOORD_BUFFER *patchCoords,
\n-
1004 PATCH_TABLE *patchTable) const {
\n-
1005
\n-
1006 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1007 dstBuffer->BindVBO(), dstDesc,
\n-
1008 duBuffer->BindVBO(), duDesc,
\n-
1009 dvBuffer->BindVBO(), dvDesc,
\n-
1010 numPatchCoords,
\n-
1011 patchCoords->BindVBO(),
\n-
1012 patchTable->GetPatchArrays(),
\n-
1013 patchTable->GetPatchIndexBuffer(),
\n-
1014 patchTable->GetPatchParamBuffer());
\n-
1015 }
\n-
1016
\n-
1069 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1070 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1071 bool EvalPatches(
\n-
1072 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1073 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1074 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1075 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1076 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1077 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1078 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1079 int numPatchCoords,
\n-
1080 PATCHCOORD_BUFFER *patchCoords,
\n-
1081 PATCH_TABLE *patchTable) const {
\n-
1082
\n-
1083 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1084 dstBuffer->BindVBO(), dstDesc,
\n-
1085 duBuffer->BindVBO(), duDesc,
\n-
1086 dvBuffer->BindVBO(), dvDesc,
\n-
1087 duuBuffer->BindVBO(), duuDesc,
\n-
1088 duvBuffer->BindVBO(), duvDesc,
\n-
1089 dvvBuffer->BindVBO(), dvvDesc,
\n-
1090 numPatchCoords,
\n-
1091 patchCoords->BindVBO(),
\n-
1092 patchTable->GetPatchArrays(),
\n-
1093 patchTable->GetPatchIndexBuffer(),
\n-
1094 patchTable->GetPatchParamBuffer());
\n-
1095 }
\n-
1096
\n-
1097 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n-
1098 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n-
1099 GLuint duBuffer, BufferDescriptor const &duDesc,
\n-
1100 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n-
1101 int numPatchCoords,
\n-
1102 GLuint patchCoordsBuffer,
\n-
1103 const PatchArrayVector &patchArrays,
\n-
1104 GLuint patchIndexBuffer,
\n-
1105 GLuint patchParamsBuffer) const;
\n-
1106
\n-
1107 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n-
1108 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n-
1109 GLuint duBuffer, BufferDescriptor const &duDesc,
\n-
1110 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n-
1111 GLuint duuBuffer, BufferDescriptor const &duuDesc,
\n-
1112 GLuint duvBuffer, BufferDescriptor const &duvDesc,
\n-
1113 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1114 int numPatchCoords,
\n-
1115 GLuint patchCoordsBuffer,
\n-
1116 const PatchArrayVector &patchArrays,
\n-
1117 GLuint patchIndexBuffer,
\n-
1118 GLuint patchParamsBuffer) const;
\n-
1119
\n-
1152 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1153 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1154 static bool EvalPatchesVarying(
\n-
1155 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1156 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1157 int numPatchCoords,
\n-
1158 PATCHCOORD_BUFFER *patchCoords,
\n-
1159 PATCH_TABLE *patchTable,
\n-
1160 GLComputeEvaluator const *instance,
\n-
1161 void * deviceContext = NULL) {
\n-
1162
\n-
1163 if (instance) {
\n-
1164 return instance->EvalPatchesVarying(
\n-
1165 srcBuffer, srcDesc,
\n-
1166 dstBuffer, dstDesc,
\n-
1167 numPatchCoords, patchCoords,
\n-
1168 patchTable);
\n-
1169 } else {
\n-
1170 // Create an instance on demand (slow)
\n-
1171 (void)deviceContext; // unused
\n-
1172 instance = Create(srcDesc, dstDesc,
\n-
1173 BufferDescriptor(),
\n-
1174 BufferDescriptor());
\n-
1175 if (instance) {
\n-
1176 bool r = instance->EvalPatchesVarying(
\n-
1177 srcBuffer, srcDesc,
\n-
1178 dstBuffer, dstDesc,
\n-
1179 numPatchCoords, patchCoords,
\n-
1180 patchTable);
\n-
1181 delete instance;
\n-
1182 return r;
\n-
1183 }
\n-
1184 return false;
\n-
1185 }
\n-
1186 }
\n-
1187
\n-
1212 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1213 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1214 bool EvalPatchesVarying(
\n-
1215 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1216 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1217 int numPatchCoords,
\n-
1218 PATCHCOORD_BUFFER *patchCoords,
\n-
1219 PATCH_TABLE *patchTable) const {
\n-
1220
\n-
1221 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1222 dstBuffer->BindVBO(), dstDesc,
\n-
1223 0, BufferDescriptor(),
\n-
1224 0, BufferDescriptor(),
\n-
1225 numPatchCoords,
\n-
1226 patchCoords->BindVBO(),
\n-
1227 patchTable->GetVaryingPatchArrays(),
\n-
1228 patchTable->GetVaryingPatchIndexBuffer(),
\n-
1229 patchTable->GetPatchParamBuffer());
\n-
1230 }
\n-
1231
\n-
1276 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1277 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1278 static bool EvalPatchesVarying(
\n-
1279 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1280 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1281 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1282 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1283 int numPatchCoords,
\n-
1284 PATCHCOORD_BUFFER *patchCoords,
\n-
1285 PATCH_TABLE *patchTable,
\n-
1286 GLComputeEvaluator const *instance,
\n-
1287 void * deviceContext = NULL) {
\n-
1288
\n-
1289 if (instance) {
\n-
1290 return instance->EvalPatchesVarying(
\n-
1291 srcBuffer, srcDesc,
\n-
1292 dstBuffer, dstDesc,
\n-
1293 duBuffer, duDesc,
\n-
1294 dvBuffer, dvDesc,
\n-
1295 numPatchCoords, patchCoords,
\n-
1296 patchTable);
\n-
1297 } else {
\n-
1298 // Create an instance on demand (slow)
\n-
1299 (void)deviceContext; // unused
\n-
1300 instance = Create(srcDesc, dstDesc,
\n-
1301 duDesc, dvDesc);
\n-
1302 if (instance) {
\n-
1303 bool r = instance->EvalPatchesVarying(
\n-
1304 srcBuffer, srcDesc,
\n-
1305 dstBuffer, dstDesc,
\n-
1306 duBuffer, duDesc,
\n-
1307 dvBuffer, dvDesc,
\n-
1308 numPatchCoords, patchCoords,
\n-
1309 patchTable);
\n-
1310 delete instance;
\n-
1311 return r;
\n-
1312 }
\n-
1313 return false;
\n-
1314 }
\n-
1315 }
\n-
1316
\n-
1353 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1354 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1355 bool EvalPatchesVarying(
\n-
1356 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1357 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1358 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1359 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1360 int numPatchCoords,
\n-
1361 PATCHCOORD_BUFFER *patchCoords,
\n-
1362 PATCH_TABLE *patchTable) const {
\n-
1363
\n-
1364 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1365 dstBuffer->BindVBO(), dstDesc,
\n-
1366 duBuffer->BindVBO(), duDesc,
\n-
1367 dvBuffer->BindVBO(), dvDesc,
\n-
1368 numPatchCoords,
\n-
1369 patchCoords->BindVBO(),
\n-
1370 patchTable->GetVaryingPatchArrays(),
\n-
1371 patchTable->GetVaryingPatchIndexBuffer(),
\n-
1372 patchTable->GetPatchParamBuffer());
\n-
1373 }
\n-
1374
\n-
1437 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1438 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1439 static bool EvalPatchesVarying(
\n-
1440 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1441 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1442 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1443 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1444 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1445 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1446 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1447 int numPatchCoords,
\n-
1448 PATCHCOORD_BUFFER *patchCoords,
\n-
1449 PATCH_TABLE *patchTable,
\n-
1450 GLComputeEvaluator const *instance,
\n-
1451 void * deviceContext = NULL) {
\n-
1452
\n-
1453 if (instance) {
\n-
1454 return instance->EvalPatchesVarying(
\n-
1455 srcBuffer, srcDesc,
\n-
1456 dstBuffer, dstDesc,
\n-
1457 duBuffer, duDesc,
\n-
1458 dvBuffer, dvDesc,
\n-
1459 duuBuffer, duuDesc,
\n-
1460 duvBuffer, duvDesc,
\n-
1461 dvvBuffer, dvvDesc,
\n-
1462 numPatchCoords, patchCoords,
\n-
1463 patchTable);
\n-
1464 } else {
\n-
1465 // Create an instance on demand (slow)
\n-
1466 (void)deviceContext; // unused
\n-
1467 instance = Create(srcDesc, dstDesc,
\n-
1468 duDesc, dvDesc,
\n-
1469 duuDesc, duvDesc, dvvDesc);
\n-
1470 if (instance) {
\n-
1471 bool r = instance->EvalPatchesVarying(
\n-
1472 srcBuffer, srcDesc,
\n-
1473 dstBuffer, dstDesc,
\n-
1474 duBuffer, duDesc,
\n-
1475 dvBuffer, dvDesc,
\n-
1476 duuBuffer, duuDesc,
\n-
1477 duvBuffer, duvDesc,
\n-
1478 dvvBuffer, dvvDesc,
\n-
1479 numPatchCoords, patchCoords,
\n-
1480 patchTable);
\n-
1481 delete instance;
\n-
1482 return r;
\n-
1483 }
\n-
1484 return false;
\n-
1485 }
\n-
1486 }
\n-
1487
\n-
1542 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1543 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1544 bool EvalPatchesVarying(
\n-
1545 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1546 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1547 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1548 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1549 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1550 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1551 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1552 int numPatchCoords,
\n-
1553 PATCHCOORD_BUFFER *patchCoords,
\n-
1554 PATCH_TABLE *patchTable) const {
\n-
1555
\n-
1556 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1557 dstBuffer->BindVBO(), dstDesc,
\n-
1558 duBuffer->BindVBO(), duDesc,
\n-
1559 dvBuffer->BindVBO(), dvDesc,
\n-
1560 duuBuffer->BindVBO(), duuDesc,
\n-
1561 duvBuffer->BindVBO(), duvDesc,
\n-
1562 dvvBuffer->BindVBO(), dvvDesc,
\n-
1563 numPatchCoords,
\n-
1564 patchCoords->BindVBO(),
\n-
1565 patchTable->GetVaryingPatchArrays(),
\n-
1566 patchTable->GetVaryingPatchIndexBuffer(),
\n-
1567 patchTable->GetPatchParamBuffer());
\n-
1568 }
\n-
1569
\n-
1604 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1605 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1606 static bool EvalPatchesFaceVarying(
\n-
1607 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1608 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1609 int numPatchCoords,
\n-
1610 PATCHCOORD_BUFFER *patchCoords,
\n-
1611 PATCH_TABLE *patchTable,
\n-
1612 int fvarChannel,
\n-
1613 GLComputeEvaluator const *instance,
\n-
1614 void * deviceContext = NULL) {
\n-
1615
\n-
1616 if (instance) {
\n-
1617 return instance->EvalPatchesFaceVarying(
\n-
1618 srcBuffer, srcDesc,
\n-
1619 dstBuffer, dstDesc,
\n-
1620 numPatchCoords, patchCoords,
\n-
1621 patchTable, fvarChannel);
\n-
1622 } else {
\n-
1623 // Create an instance on demand (slow)
\n-
1624 (void)deviceContext; // unused
\n-
1625 instance = Create(srcDesc, dstDesc,
\n-
1626 BufferDescriptor(),
\n-
1627 BufferDescriptor());
\n-
1628 if (instance) {
\n-
1629 bool r = instance->EvalPatchesFaceVarying(
\n-
1630 srcBuffer, srcDesc,
\n-
1631 dstBuffer, dstDesc,
\n-
1632 numPatchCoords, patchCoords,
\n-
1633 patchTable, fvarChannel);
\n-
1634 delete instance;
\n-
1635 return r;
\n-
1636 }
\n-
1637 return false;
\n-
1638 }
\n-
1639 }
\n-
1640
\n-
1667 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1668 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1669 bool EvalPatchesFaceVarying(
\n-
1670 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1671 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1672 int numPatchCoords,
\n-
1673 PATCHCOORD_BUFFER *patchCoords,
\n-
1674 PATCH_TABLE *patchTable,
\n-
1675 int fvarChannel = 0) const {
\n-
1676
\n-
1677 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1678 dstBuffer->BindVBO(), dstDesc,
\n-
1679 0, BufferDescriptor(),
\n-
1680 0, BufferDescriptor(),
\n-
1681 numPatchCoords,
\n-
1682 patchCoords->BindVBO(),
\n-
1683 patchTable->GetFVarPatchArrays(fvarChannel),
\n-
1684 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1685 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
1686 }
\n-
1687
\n-
1734 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1735 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1736 static bool EvalPatchesFaceVarying(
\n-
1737 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1738 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1739 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1740 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1741 int numPatchCoords,
\n-
1742 PATCHCOORD_BUFFER *patchCoords,
\n-
1743 PATCH_TABLE *patchTable,
\n-
1744 int fvarChannel,
\n-
1745 GLComputeEvaluator const *instance,
\n-
1746 void * deviceContext = NULL) {
\n-
1747
\n-
1748 if (instance) {
\n-
1749 return instance->EvalPatchesFaceVarying(
\n-
1750 srcBuffer, srcDesc,
\n-
1751 dstBuffer, dstDesc,
\n-
1752 duBuffer, duDesc,
\n-
1753 dvBuffer, dvDesc,
\n-
1754 numPatchCoords, patchCoords,
\n-
1755 patchTable, fvarChannel);
\n-
1756 } else {
\n-
1757 // Create an instance on demand (slow)
\n-
1758 (void)deviceContext; // unused
\n-
1759 instance = Create(srcDesc, dstDesc,
\n-
1760 duDesc, dvDesc);
\n-
1761 if (instance) {
\n-
1762 bool r = instance->EvalPatchesFaceVarying(
\n-
1763 srcBuffer, srcDesc,
\n-
1764 dstBuffer, dstDesc,
\n-
1765 duBuffer, duDesc,
\n-
1766 dvBuffer, dvDesc,
\n-
1767 numPatchCoords, patchCoords,
\n-
1768 patchTable, fvarChannel);
\n-
1769 delete instance;
\n-
1770 return r;
\n-
1771 }
\n-
1772 return false;
\n-
1773 }
\n-
1774 }
\n-
1775
\n-
1814 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1815 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1816 bool EvalPatchesFaceVarying(
\n-
1817 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1818 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1819 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1820 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1821 int numPatchCoords,
\n-
1822 PATCHCOORD_BUFFER *patchCoords,
\n-
1823 PATCH_TABLE *patchTable,
\n-
1824 int fvarChannel = 0) const {
\n-
1825
\n-
1826 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1827 dstBuffer->BindVBO(), dstDesc,
\n-
1828 duBuffer->BindVBO(), duDesc,
\n-
1829 dvBuffer->BindVBO(), dvDesc,
\n-
1830 numPatchCoords,
\n-
1831 patchCoords->BindVBO(),
\n-
1832 patchTable->GetFVarPatchArrays(fvarChannel),
\n-
1833 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1834 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
1835 }
\n-
1836
\n-
1901 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1902 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1903 static bool EvalPatchesFaceVarying(
\n-
1904 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1905 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1906 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1907 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1908 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1909 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1910 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1911 int numPatchCoords,
\n-
1912 PATCHCOORD_BUFFER *patchCoords,
\n-
1913 PATCH_TABLE *patchTable,
\n-
1914 int fvarChannel,
\n-
1915 GLComputeEvaluator const *instance,
\n-
1916 void * deviceContext = NULL) {
\n-
1917
\n-
1918 if (instance) {
\n-
1919 return instance->EvalPatchesFaceVarying(
\n-
1920 srcBuffer, srcDesc,
\n-
1921 dstBuffer, dstDesc,
\n-
1922 duBuffer, duDesc,
\n-
1923 dvBuffer, dvDesc,
\n-
1924 duuBuffer, duuDesc,
\n-
1925 duvBuffer, duvDesc,
\n-
1926 dvvBuffer, dvvDesc,
\n-
1927 numPatchCoords, patchCoords,
\n-
1928 patchTable, fvarChannel);
\n-
1929 } else {
\n-
1930 // Create an instance on demand (slow)
\n-
1931 (void)deviceContext; // unused
\n-
1932 instance = Create(srcDesc, dstDesc,
\n-
1933 duDesc, dvDesc,
\n-
1934 duuDesc, duvDesc, dvvDesc);
\n-
1935 if (instance) {
\n-
1936 bool r = instance->EvalPatchesFaceVarying(
\n-
1937 srcBuffer, srcDesc,
\n-
1938 dstBuffer, dstDesc,
\n-
1939 duBuffer, duDesc,
\n-
1940 dvBuffer, dvDesc,
\n-
1941 duuBuffer, duuDesc,
\n-
1942 duvBuffer, duvDesc,
\n-
1943 dvvBuffer, dvvDesc,
\n-
1944 numPatchCoords, patchCoords,
\n-
1945 patchTable, fvarChannel);
\n-
1946 delete instance;
\n-
1947 return r;
\n-
1948 }
\n-
1949 return false;
\n-
1950 }
\n-
1951 }
\n-
1952
\n-
2009 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
2010 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
2011 bool EvalPatchesFaceVarying(
\n-
2012 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
2013 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
2014 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
2015 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
2016 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
2017 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
2018 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
2019 int numPatchCoords,
\n-
2020 PATCHCOORD_BUFFER *patchCoords,
\n-
2021 PATCH_TABLE *patchTable,
\n-
2022 int fvarChannel = 0) const {
\n-
2023
\n-
2024 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
2025 dstBuffer->BindVBO(), dstDesc,
\n-
2026 duBuffer->BindVBO(), duDesc,
\n-
2027 dvBuffer->BindVBO(), dvDesc,
\n-
2028 duuBuffer->BindVBO(), duuDesc,
\n-
2029 duvBuffer->BindVBO(), duvDesc,
\n-
2030 dvvBuffer->BindVBO(), dvvDesc,
\n-
2031 numPatchCoords,
\n-
2032 patchCoords->BindVBO(),
\n-
2033 patchTable->GetFVarPatchArrays(fvarChannel),
\n-
2034 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
2035 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
2036 }
\n-
2037
\n-
2043
\n-
2046 bool Compile(BufferDescriptor const &srcDesc,
\n-
2047 BufferDescriptor const &dstDesc,
\n-
2048 BufferDescriptor const &duDesc = BufferDescriptor(),
\n-
2049 BufferDescriptor const &dvDesc = BufferDescriptor(),
\n-
2050 BufferDescriptor const &duuDesc = BufferDescriptor(),
\n-
2051 BufferDescriptor const &duvDesc = BufferDescriptor(),
\n-
2052 BufferDescriptor const &dvvDesc = BufferDescriptor());
\n-
2053
\n-
2055 static void Synchronize(void *deviceContext);
\n-
2056
\n-
2057private:
\n-
2058 struct _StencilKernel {
\n-
2059 _StencilKernel();
\n-
2060 ~_StencilKernel();
\n-
2061 bool Compile(BufferDescriptor const &srcDesc,
\n-
2062 BufferDescriptor const &dstDesc,
\n-
2063 BufferDescriptor const &duDesc,
\n-
2064 BufferDescriptor const &dvDesc,
\n-
2065 BufferDescriptor const &duuDesc,
\n-
2066 BufferDescriptor const &duvDesc,
\n-
2067 BufferDescriptor const &dvvDesc,
\n-
2068 int workGroupSize);
\n-
2069 GLuint program;
\n-
2070 GLuint uniformStart;
\n-
2071 GLuint uniformEnd;
\n-
2072 GLuint uniformSrcOffset;
\n-
2073 GLuint uniformDstOffset;
\n-
2074 GLuint uniformDuDesc;
\n-
2075 GLuint uniformDvDesc;
\n-
2076 GLuint uniformDuuDesc;
\n-
2077 GLuint uniformDuvDesc;
\n-
2078 GLuint uniformDvvDesc;
\n-
2079 } _stencilKernel;
\n-
2080
\n-
2081 struct _PatchKernel {
\n-
2082 _PatchKernel();
\n-
2083 ~_PatchKernel();
\n-
2084 bool Compile(BufferDescriptor const &srcDesc,
\n-
2085 BufferDescriptor const &dstDesc,
\n-
2086 BufferDescriptor const &duDesc,
\n-
2087 BufferDescriptor const &dvDesc,
\n-
2088 BufferDescriptor const &duuDesc,
\n-
2089 BufferDescriptor const &duvDesc,
\n-
2090 BufferDescriptor const &dvvDesc,
\n-
2091 int workGroupSize);
\n-
2092 GLuint program;
\n-
2093 GLuint uniformSrcOffset;
\n-
2094 GLuint uniformDstOffset;
\n-
2095 GLuint uniformPatchArray;
\n-
2096 GLuint uniformDuDesc;
\n-
2097 GLuint uniformDvDesc;
\n-
2098 GLuint uniformDuuDesc;
\n-
2099 GLuint uniformDuvDesc;
\n-
2100 GLuint uniformDvvDesc;
\n-
2101 } _patchKernel;
\n-
2102
\n-
2103 int _workGroupSize;
\n-
2104 GLuint _patchArraysSSBO;
\n-
2105};
\n-
2106
\n-
2107} // end namespace Osd
\n-
2108
\n-
2109} // end namespace OPENSUBDIV_VERSION
\n-
2110using namespace OPENSUBDIV_VERSION;
\n-
2111
\n-
2112} // end namespace OpenSubdiv
\n-
2113
\n-
2114
\n-
2115#endif // OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H
\n+
94} // end namespace OPENSUBDIV_VERSION
\n+
95using namespace OPENSUBDIV_VERSION;
\n+
96
\n+
97} // end namespace OpenSubdiv
\n+
98
\n+
99#endif // OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO
GL stencil table (Shader Storage buffer)
Definition: glComputeEvaluator.h:51
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GLStencilTableSSBO
GLStencilTableSSBO(Far::LimitStencilTable const *limitStencilTable)
\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GLStencilTableSSBO
GLStencilTableSSBO(Far::StencilTable const *stencilTable)
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::Create
static GLStencilTableSSBO * Create(Far::LimitStencilTable const *limitStencilTable, void *deviceContext=NULL)
Definition: glComputeEvaluator.h:58
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::Create
static GLStencilTableSSBO * Create(Far::StencilTable const *stencilTable, void *deviceContext=NULL)
Definition: glComputeEvaluator.h:53
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1214
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1736
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glComputeEvaluator.h:434
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1154
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1606
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glComputeEvaluator.h:502
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1278
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays, GLuint patchIndexBuffer, GLuint patchParamsBuffer) const
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::~GLComputeEvaluator
~GLComputeEvaluator()
Destructor. note that the GL context must be made current.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: glComputeEvaluator.h:1071
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glComputeEvaluator.h:236
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, GLuint sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer, GLuint duWeightsBuffer, GLuint dvWeightsBuffer, GLuint duuWeightsBuffer, GLuint duvWeightsBuffer, GLuint dvvWeightsBuffer, int start, int end) const
Dispatch the GLSL compute kernel on GPU asynchronously returns false if the kernel hasn't been compil...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glComputeEvaluator.h:168
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1439
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:769
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1903
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1544
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1816
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:943
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: glComputeEvaluator.h:997
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:691
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glComputeEvaluator.h:386
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays, GLuint patchIndexBuffer, GLuint patchParamsBuffer) const
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:2011
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Create
static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, void *deviceContext=NULL)
Definition: glComputeEvaluator.h:99
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer, GLuint duWeightsBuffer, GLuint dvWeightsBuffer, int start, int end) const
Dispatch the GLSL compute kernel on GPU asynchronously returns false if the kernel hasn't been compil...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glComputeEvaluator.h:326
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Synchronize
static void Synchronize(void *deviceContext)
Wait the dispatched kernel finishes.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Compile
bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc=BufferDescriptor(), BufferDescriptor const &dvDesc=BufferDescriptor(), BufferDescriptor const &duuDesc=BufferDescriptor(), BufferDescriptor const &duvDesc=BufferDescriptor(), BufferDescriptor const &dvvDesc=BufferDescriptor())
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1355
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Create
static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, void *deviceContext=NULL)
Definition: glComputeEvaluator.h:111
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1669
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:870
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer
Concrete vertex buffer class for Cpu subdivision and DirectX drawing.
Definition: cpuD3D11VertexBuffer.h:48
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::Create
static CpuD3D11VertexBuffer * Create(int numElements, int numVertices, ID3D11DeviceContext *deviceContext)
Creator. Returns NULL if error.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::CpuD3D11VertexBuffer
CpuD3D11VertexBuffer(int numElements, int numVertices)
Constructor.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::GetNumElements
int GetNumElements() const
Returns how many elements defined in this vertex buffer.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::UpdateData
void UpdateData(const float *src, int startVertex, int numVertices, void *deviceContext=NULL)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::GetNumVertices
int GetNumVertices() const
Returns how many vertices allocated in this vertex buffer.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindD3D11Buffer
ID3D11Buffer * BindD3D11Buffer(ID3D11DeviceContext *deviceContext)
Returns the D3D11 buffer object.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindCpuBuffer
float * BindCpuBuffer()
Returns the address of CPU buffer.
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindVBO
ID3D11Buffer * BindVBO(ID3D11DeviceContext *deviceContext)
Returns the D3D11 buffer object (for Osd::Mesh interface)
Definition: cpuD3D11VertexBuffer.h:75
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-glComputeEvaluator.h\n+cpuD3D11VertexBuffer.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2015 Pixar\n+ 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,1365 +30,105 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H\n- 26#define OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H\n+ 25#ifndef OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H\n+ 26#define OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H\n 27\n 28#include \"../version.h\"\n- 29\n- 30#include \"../osd/opengl.h\"\n- 31#include \"../osd/types.h\"\n- 32#include \"../osd/bufferDescriptor.h\"\n- 33\n- 34namespace OpenSubdiv {\n- 35namespace OPENSUBDIV_VERSION {\n- 36\n- 37namespace Far {\n- 38 class PatchTable;\n- 39 class StencilTable;\n- 40 class LimitStencilTable;\n- 41}\n- 42\n- 43namespace Osd {\n- 44\n-51class GLStencilTableSSBO {\n- 52public:\n-53 static GLStencilTableSSBO *Create(Far::StencilTable const *stencilTable,\n- 54 void *deviceContext = NULL) {\n- 55 (void)deviceContext; // unused\n- 56 return new GLStencilTableSSBO(stencilTable);\n- 57 }\n-58 static GLStencilTableSSBO *Create(\n- 59 Far::LimitStencilTable const *limitStencilTable,\n- 60 void *deviceContext = NULL) {\n- 61 (void)deviceContext; // unused\n- 62 return new GLStencilTableSSBO(limitStencilTable);\n- 63 }\n+ 29#include \n+ 30\n+ 31struct ID3D11Buffer;\n+ 32struct ID3D11Device;\n+ 33struct ID3D11DeviceContext;\n+ 34\n+ 35namespace OpenSubdiv {\n+ 36namespace OPENSUBDIV_VERSION {\n+ 37\n+ 38namespace Osd {\n+ 39\n+48class CpuD3D11VertexBuffer {\n+ 49public:\n+51 static CpuD3D11VertexBuffer * Create(int numElements, int numVertices,\n+ 52 ID3D11DeviceContext *deviceContext);\n+ 53\n+55 virtual ~CpuD3D11VertexBuffer();\n+ 56\n+59 void UpdateData(const float *src, int startVertex, int numVertices,\n+ 60 void *deviceContext = NULL);\n+ 61\n+63 int GetNumElements() const;\n 64\n-65 explicit GLStencilTableSSBO(Far::StencilTable const *stencilTable);\n-66 explicit GLStencilTableSSBO(Far::LimitStencilTable const\n-*limitStencilTable);\n-67 ~GLStencilTableSSBO();\n- 68\n- 69 // interfaces needed for GLSLComputeKernel\n-70 GLuint GetSizesBuffer() const { return _sizes; }\n-71 GLuint GetOffsetsBuffer() const { return _offsets; }\n-72 GLuint GetIndicesBuffer() const { return _indices; }\n-73 GLuint GetWeightsBuffer() const { return _weights; }\n-74 GLuint GetDuWeightsBuffer() const { return _duWeights; }\n-75 GLuint GetDvWeightsBuffer() const { return _dvWeights; }\n-76 GLuint GetDuuWeightsBuffer() const { return _duuWeights; }\n-77 GLuint GetDuvWeightsBuffer() const { return _duvWeights; }\n-78 GLuint GetDvvWeightsBuffer() const { return _dvvWeights; }\n-79 int GetNumStencils() const { return _numStencils; }\n- 80\n- 81private:\n- 82 GLuint _sizes;\n- 83 GLuint _offsets;\n- 84 GLuint _indices;\n- 85 GLuint _weights;\n- 86 GLuint _duWeights;\n- 87 GLuint _dvWeights;\n- 88 GLuint _duuWeights;\n- 89 GLuint _duvWeights;\n- 90 GLuint _dvvWeights;\n- 91 int _numStencils;\n- 92};\n+66 int GetNumVertices() const;\n+ 67\n+69 float * BindCpuBuffer();\n+ 70\n+72 ID3D11Buffer *BindD3D11Buffer(ID3D11DeviceContext *deviceContext);\n+ 73\n+75 ID3D11Buffer *BindVBO(ID3D11DeviceContext *deviceContext) {\n+ 76 return BindD3D11Buffer(deviceContext);\n+ 77 }\n+ 78\n+ 79protected:\n+81 CpuD3D11VertexBuffer(int numElements, int numVertices);\n+ 82\n+83 bool allocate(ID3D11Device *device);\n+ 84\n+ 85private:\n+ 86 int _numElements;\n+ 87 int _numVertices;\n+ 88 ID3D11Buffer *_d3d11Buffer;\n+ 89 float *_cpuBuffer;\n+ 90};\n+ 91\n+ 92} // end namespace Osd\n 93\n- 94// -------------------------------------------------------------------------\n---\n- 95\n-96class GLComputeEvaluator {\n- 97public:\n-98 typedef bool Instantiatable;\n-99 static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n- 100 BufferDescriptor const &dstDesc,\n- 101 BufferDescriptor const &duDesc,\n- 102 BufferDescriptor const &dvDesc,\n- 103 void * deviceContext = NULL) {\n- 104 return Create(srcDesc, dstDesc, duDesc, dvDesc,\n- 105 BufferDescriptor(),\n- 106 BufferDescriptor(),\n- 107 BufferDescriptor(),\n- 108 deviceContext);\n- 109 }\n- 110\n-111 static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n- 112 BufferDescriptor const &dstDesc,\n- 113 BufferDescriptor const &duDesc,\n- 114 BufferDescriptor const &dvDesc,\n- 115 BufferDescriptor const &duuDesc,\n- 116 BufferDescriptor const &duvDesc,\n- 117 BufferDescriptor const &dvvDesc,\n- 118 void * deviceContext = NULL) {\n- 119 (void)deviceContext; // not used\n- 120 GLComputeEvaluator *instance = new GLComputeEvaluator();\n- 121 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc,\n- 122 duuDesc, duvDesc, dvvDesc))\n- 123 return instance;\n- 124 delete instance;\n- 125 return NULL;\n- 126 }\n- 127\n-129 GLComputeEvaluator();\n- 130\n-132 ~GLComputeEvaluator();\n- 133\n- 139\n- 167 template \n-168 static bool EvalStencils(\n- 169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 171 STENCIL_TABLE const *stencilTable,\n- 172 GLComputeEvaluator const *instance,\n- 173 void * deviceContext = NULL) {\n- 174\n- 175 if (instance) {\n- 176 return instance->EvalStencils(srcBuffer, srcDesc,\n- 177 dstBuffer, dstDesc,\n- 178 stencilTable);\n- 179 } else {\n- 180 // Create an instance on demand (slow)\n- 181 (void)deviceContext; // unused\n- 182 instance = Create(srcDesc, dstDesc,\n- 183 BufferDescriptor(),\n- 184 BufferDescriptor());\n- 185 if (instance) {\n- 186 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 187 dstBuffer, dstDesc,\n- 188 stencilTable);\n- 189 delete instance;\n- 190 return r;\n- 191 }\n- 192 return false;\n- 193 }\n- 194 }\n- 195\n- 235 template \n-236 static bool EvalStencils(\n- 237 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 238 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 239 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 240 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 241 STENCIL_TABLE const *stencilTable,\n- 242 GLComputeEvaluator const *instance,\n- 243 void * deviceContext = NULL) {\n- 244\n- 245 if (instance) {\n- 246 return instance->EvalStencils(srcBuffer, srcDesc,\n- 247 dstBuffer, dstDesc,\n- 248 duBuffer, duDesc,\n- 249 dvBuffer, dvDesc,\n- 250 stencilTable);\n- 251 } else {\n- 252 // Create an instance on demand (slow)\n- 253 (void)deviceContext; // unused\n- 254 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);\n- 255 if (instance) {\n- 256 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 257 dstBuffer, dstDesc,\n- 258 duBuffer, duDesc,\n- 259 dvBuffer, dvDesc,\n- 260 stencilTable);\n- 261 delete instance;\n- 262 return r;\n- 263 }\n- 264 return false;\n- 265 }\n- 266 }\n- 267\n- 325 template \n-326 static bool EvalStencils(\n- 327 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 328 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 329 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 330 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 331 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 332 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 333 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 334 STENCIL_TABLE const *stencilTable,\n- 335 GLComputeEvaluator const *instance,\n- 336 void * deviceContext = NULL) {\n- 337\n- 338 if (instance) {\n- 339 return instance->EvalStencils(srcBuffer, srcDesc,\n- 340 dstBuffer, dstDesc,\n- 341 duBuffer, duDesc,\n- 342 dvBuffer, dvDesc,\n- 343 duuBuffer, duuDesc,\n- 344 duvBuffer, duvDesc,\n- 345 dvvBuffer, dvvDesc,\n- 346 stencilTable);\n- 347 } else {\n- 348 // Create an instance on demand (slow)\n- 349 (void)deviceContext; // unused\n- 350 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,\n- 351 duuDesc, duvDesc, dvvDesc);\n- 352 if (instance) {\n- 353 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 354 dstBuffer, dstDesc,\n- 355 duBuffer, duDesc,\n- 356 dvBuffer, dvDesc,\n- 357 duuBuffer, duuDesc,\n- 358 duvBuffer, duvDesc,\n- 359 dvvBuffer, dvvDesc,\n- 360 stencilTable);\n- 361 delete instance;\n- 362 return r;\n- 363 }\n- 364 return false;\n- 365 }\n- 366 }\n- 367\n- 385 template \n-386 bool EvalStencils(\n- 387 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 388 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 389 STENCIL_TABLE const *stencilTable) const {\n- 390 return EvalStencils(srcBuffer->BindVBO(), srcDesc,\n- 391 dstBuffer->BindVBO(), dstDesc,\n- 392 0, BufferDescriptor(),\n- 393 0, BufferDescriptor(),\n- 394 stencilTable->GetSizesBuffer(),\n- 395 stencilTable->GetOffsetsBuffer(),\n- 396 stencilTable->GetIndicesBuffer(),\n- 397 stencilTable->GetWeightsBuffer(),\n- 398 0,\n- 399 0,\n- 400 /* start = */ 0,\n- 401 /* end = */ stencilTable->GetNumStencils());\n- 402 }\n- 403\n- 433 template \n-434 bool EvalStencils(\n- 435 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 436 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 437 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 438 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 439 STENCIL_TABLE const *stencilTable) const {\n- 440 return EvalStencils(srcBuffer->BindVBO(), srcDesc,\n- 441 dstBuffer->BindVBO(), dstDesc,\n- 442 duBuffer->BindVBO(), duDesc,\n- 443 dvBuffer->BindVBO(), dvDesc,\n- 444 stencilTable->GetSizesBuffer(),\n- 445 stencilTable->GetOffsetsBuffer(),\n- 446 stencilTable->GetIndicesBuffer(),\n- 447 stencilTable->GetWeightsBuffer(),\n- 448 stencilTable->GetDuWeightsBuffer(),\n- 449 stencilTable->GetDvWeightsBuffer(),\n- 450 /* start = */ 0,\n- 451 /* end = */ stencilTable->GetNumStencils());\n- 452 }\n- 453\n- 501 template \n-502 bool EvalStencils(\n- 503 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 504 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 505 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 506 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 507 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 508 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 509 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 510 STENCIL_TABLE const *stencilTable) const {\n- 511 return EvalStencils(srcBuffer->BindVBO(), srcDesc,\n- 512 dstBuffer->BindVBO(), dstDesc,\n- 513 duBuffer->BindVBO(), duDesc,\n- 514 dvBuffer->BindVBO(), dvDesc,\n- 515 duuBuffer->BindVBO(), duuDesc,\n- 516 duvBuffer->BindVBO(), duvDesc,\n- 517 dvvBuffer->BindVBO(), dvvDesc,\n- 518 stencilTable->GetSizesBuffer(),\n- 519 stencilTable->GetOffsetsBuffer(),\n- 520 stencilTable->GetIndicesBuffer(),\n- 521 stencilTable->GetWeightsBuffer(),\n- 522 stencilTable->GetDuWeightsBuffer(),\n- 523 stencilTable->GetDvWeightsBuffer(),\n- 524 stencilTable->GetDuuWeightsBuffer(),\n- 525 stencilTable->GetDuvWeightsBuffer(),\n- 526 stencilTable->GetDvvWeightsBuffer(),\n- 527 /* start = */ 0,\n- 528 /* end = */ stencilTable->GetNumStencils());\n- 529 }\n- 530\n-566 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n- 567 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n- 568 GLuint duBuffer, BufferDescriptor const &duDesc,\n- 569 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n- 570 GLuint sizesBuffer,\n- 571 GLuint offsetsBuffer,\n- 572 GLuint indicesBuffer,\n- 573 GLuint weightsBuffer,\n- 574 GLuint duWeightsBuffer,\n- 575 GLuint dvWeightsBuffer,\n- 576 int start,\n- 577 int end) const;\n- 578\n-632 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n- 633 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n- 634 GLuint duBuffer, BufferDescriptor const &duDesc,\n- 635 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n- 636 GLuint duuBuffer, BufferDescriptor const &duuDesc,\n- 637 GLuint duvBuffer, BufferDescriptor const &duvDesc,\n- 638 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,\n- 639 GLuint sizesBuffer,\n- 640 GLuint offsetsBuffer,\n- 641 GLuint indicesBuffer,\n- 642 GLuint weightsBuffer,\n- 643 GLuint duWeightsBuffer,\n- 644 GLuint dvWeightsBuffer,\n- 645 GLuint duuWeightsBuffer,\n- 646 GLuint duvWeightsBuffer,\n- 647 GLuint dvvWeightsBuffer,\n- 648 int start,\n- 649 int end) const;\n- 650\n- 656\n- 689 template \n-691 static bool EvalPatches(\n- 692 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 693 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 694 int numPatchCoords,\n- 695 PATCHCOORD_BUFFER *patchCoords,\n- 696 PATCH_TABLE *patchTable,\n- 697 GLComputeEvaluator const *instance,\n- 698 void * deviceContext = NULL) {\n- 699\n- 700 if (instance) {\n- 701 return instance->EvalPatches(srcBuffer, srcDesc,\n- 702 dstBuffer, dstDesc,\n- 703 numPatchCoords, patchCoords,\n- 704 patchTable);\n- 705 } else {\n- 706 // Create an instance on demand (slow)\n- 707 (void)deviceContext; // unused\n- 708 instance = Create(srcDesc, dstDesc,\n- 709 BufferDescriptor(),\n- 710 BufferDescriptor());\n- 711 if (instance) {\n- 712 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 713 dstBuffer, dstDesc,\n- 714 numPatchCoords, patchCoords,\n- 715 patchTable);\n- 716 delete instance;\n- 717 return r;\n- 718 }\n- 719 return false;\n- 720 }\n- 721 }\n- 722\n- 767 template \n-769 static bool EvalPatches(\n- 770 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 771 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 772 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 773 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 774 int numPatchCoords,\n- 775 PATCHCOORD_BUFFER *patchCoords,\n- 776 PATCH_TABLE *patchTable,\n- 777 GLComputeEvaluator const *instance,\n- 778 void * deviceContext = NULL) {\n- 779\n- 780 if (instance) {\n- 781 return instance->EvalPatches(srcBuffer, srcDesc,\n- 782 dstBuffer, dstDesc,\n- 783 duBuffer, duDesc,\n- 784 dvBuffer, dvDesc,\n- 785 numPatchCoords, patchCoords,\n- 786 patchTable);\n- 787 } else {\n- 788 // Create an instance on demand (slow)\n- 789 (void)deviceContext; // unused\n- 790 instance = Create(srcDesc, dstDesc,\n- 791 duDesc, dvDesc);\n- 792 if (instance) {\n- 793 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 794 dstBuffer, dstDesc,\n- 795 duBuffer, duDesc,\n- 796 dvBuffer, dvDesc,\n- 797 numPatchCoords, patchCoords,\n- 798 patchTable);\n- 799 delete instance;\n- 800 return r;\n- 801 }\n- 802 return false;\n- 803 }\n- 804 }\n- 805\n- 868 template \n-870 static bool EvalPatches(\n- 871 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 872 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 873 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 874 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 875 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 876 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 877 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 878 int numPatchCoords,\n- 879 PATCHCOORD_BUFFER *patchCoords,\n- 880 PATCH_TABLE *patchTable,\n- 881 GLComputeEvaluator const *instance,\n- 882 void * deviceContext = NULL) {\n- 883\n- 884 if (instance) {\n- 885 return instance->EvalPatches(srcBuffer, srcDesc,\n- 886 dstBuffer, dstDesc,\n- 887 duBuffer, duDesc,\n- 888 dvBuffer, dvDesc,\n- 889 duuBuffer, duuDesc,\n- 890 duvBuffer, duvDesc,\n- 891 dvvBuffer, dvvDesc,\n- 892 numPatchCoords, patchCoords,\n- 893 patchTable);\n- 894 } else {\n- 895 // Create an instance on demand (slow)\n- 896 (void)deviceContext; // unused\n- 897 instance = Create(srcDesc, dstDesc,\n- 898 duDesc, dvDesc,\n- 899 duuDesc, duvDesc, dvvDesc);\n- 900 if (instance) {\n- 901 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 902 dstBuffer, dstDesc,\n- 903 duBuffer, duDesc,\n- 904 dvBuffer, dvDesc,\n- 905 duuBuffer, duuDesc,\n- 906 duvBuffer, duvDesc,\n- 907 dvvBuffer, dvvDesc,\n- 908 numPatchCoords, patchCoords,\n- 909 patchTable);\n- 910 delete instance;\n- 911 return r;\n- 912 }\n- 913 return false;\n- 914 }\n- 915 }\n- 916\n- 941 template \n-943 bool EvalPatches(\n- 944 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 945 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 946 int numPatchCoords,\n- 947 PATCHCOORD_BUFFER *patchCoords,\n- 948 PATCH_TABLE *patchTable) const {\n- 949\n- 950 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 951 dstBuffer->BindVBO(), dstDesc,\n- 952 0, BufferDescriptor(),\n- 953 0, BufferDescriptor(),\n- 954 numPatchCoords,\n- 955 patchCoords->BindVBO(),\n- 956 patchTable->GetPatchArrays(),\n- 957 patchTable->GetPatchIndexBuffer(),\n- 958 patchTable->GetPatchParamBuffer());\n- 959 }\n- 960\n- 995 template \n-997 bool EvalPatches(\n- 998 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 999 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1000 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1001 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1002 int numPatchCoords,\n- 1003 PATCHCOORD_BUFFER *patchCoords,\n- 1004 PATCH_TABLE *patchTable) const {\n- 1005\n- 1006 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1007 dstBuffer->BindVBO(), dstDesc,\n- 1008 duBuffer->BindVBO(), duDesc,\n- 1009 dvBuffer->BindVBO(), dvDesc,\n- 1010 numPatchCoords,\n- 1011 patchCoords->BindVBO(),\n- 1012 patchTable->GetPatchArrays(),\n- 1013 patchTable->GetPatchIndexBuffer(),\n- 1014 patchTable->GetPatchParamBuffer());\n- 1015 }\n- 1016\n- 1069 template \n-1071 bool EvalPatches(\n- 1072 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1073 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1074 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1075 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1076 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1077 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1078 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1079 int numPatchCoords,\n- 1080 PATCHCOORD_BUFFER *patchCoords,\n- 1081 PATCH_TABLE *patchTable) const {\n- 1082\n- 1083 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1084 dstBuffer->BindVBO(), dstDesc,\n- 1085 duBuffer->BindVBO(), duDesc,\n- 1086 dvBuffer->BindVBO(), dvDesc,\n- 1087 duuBuffer->BindVBO(), duuDesc,\n- 1088 duvBuffer->BindVBO(), duvDesc,\n- 1089 dvvBuffer->BindVBO(), dvvDesc,\n- 1090 numPatchCoords,\n- 1091 patchCoords->BindVBO(),\n- 1092 patchTable->GetPatchArrays(),\n- 1093 patchTable->GetPatchIndexBuffer(),\n- 1094 patchTable->GetPatchParamBuffer());\n- 1095 }\n- 1096\n-1097 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n- 1098 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n- 1099 GLuint duBuffer, BufferDescriptor const &duDesc,\n- 1100 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n- 1101 int numPatchCoords,\n- 1102 GLuint patchCoordsBuffer,\n- 1103 const PatchArrayVector &patchArrays,\n- 1104 GLuint patchIndexBuffer,\n- 1105 GLuint patchParamsBuffer) const;\n- 1106\n-1107 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n- 1108 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n- 1109 GLuint duBuffer, BufferDescriptor const &duDesc,\n- 1110 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n- 1111 GLuint duuBuffer, BufferDescriptor const &duuDesc,\n- 1112 GLuint duvBuffer, BufferDescriptor const &duvDesc,\n- 1113 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1114 int numPatchCoords,\n- 1115 GLuint patchCoordsBuffer,\n- 1116 const PatchArrayVector &patchArrays,\n- 1117 GLuint patchIndexBuffer,\n- 1118 GLuint patchParamsBuffer) const;\n- 1119\n- 1152 template \n-1154 static bool EvalPatchesVarying(\n- 1155 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1156 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1157 int numPatchCoords,\n- 1158 PATCHCOORD_BUFFER *patchCoords,\n- 1159 PATCH_TABLE *patchTable,\n- 1160 GLComputeEvaluator const *instance,\n- 1161 void * deviceContext = NULL) {\n- 1162\n- 1163 if (instance) {\n- 1164 return instance->EvalPatchesVarying(\n- 1165 srcBuffer, srcDesc,\n- 1166 dstBuffer, dstDesc,\n- 1167 numPatchCoords, patchCoords,\n- 1168 patchTable);\n- 1169 } else {\n- 1170 // Create an instance on demand (slow)\n- 1171 (void)deviceContext; // unused\n- 1172 instance = Create(srcDesc, dstDesc,\n- 1173 BufferDescriptor(),\n- 1174 BufferDescriptor());\n- 1175 if (instance) {\n- 1176 bool r = instance->EvalPatchesVarying(\n- 1177 srcBuffer, srcDesc,\n- 1178 dstBuffer, dstDesc,\n- 1179 numPatchCoords, patchCoords,\n- 1180 patchTable);\n- 1181 delete instance;\n- 1182 return r;\n- 1183 }\n- 1184 return false;\n- 1185 }\n- 1186 }\n- 1187\n- 1212 template \n-1214 bool EvalPatchesVarying(\n- 1215 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1216 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1217 int numPatchCoords,\n- 1218 PATCHCOORD_BUFFER *patchCoords,\n- 1219 PATCH_TABLE *patchTable) const {\n- 1220\n- 1221 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1222 dstBuffer->BindVBO(), dstDesc,\n- 1223 0, BufferDescriptor(),\n- 1224 0, BufferDescriptor(),\n- 1225 numPatchCoords,\n- 1226 patchCoords->BindVBO(),\n- 1227 patchTable->GetVaryingPatchArrays(),\n- 1228 patchTable->GetVaryingPatchIndexBuffer(),\n- 1229 patchTable->GetPatchParamBuffer());\n- 1230 }\n- 1231\n- 1276 template \n-1278 static bool EvalPatchesVarying(\n- 1279 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1280 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1281 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1282 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1283 int numPatchCoords,\n- 1284 PATCHCOORD_BUFFER *patchCoords,\n- 1285 PATCH_TABLE *patchTable,\n- 1286 GLComputeEvaluator const *instance,\n- 1287 void * deviceContext = NULL) {\n- 1288\n- 1289 if (instance) {\n- 1290 return instance->EvalPatchesVarying(\n- 1291 srcBuffer, srcDesc,\n- 1292 dstBuffer, dstDesc,\n- 1293 duBuffer, duDesc,\n- 1294 dvBuffer, dvDesc,\n- 1295 numPatchCoords, patchCoords,\n- 1296 patchTable);\n- 1297 } else {\n- 1298 // Create an instance on demand (slow)\n- 1299 (void)deviceContext; // unused\n- 1300 instance = Create(srcDesc, dstDesc,\n- 1301 duDesc, dvDesc);\n- 1302 if (instance) {\n- 1303 bool r = instance->EvalPatchesVarying(\n- 1304 srcBuffer, srcDesc,\n- 1305 dstBuffer, dstDesc,\n- 1306 duBuffer, duDesc,\n- 1307 dvBuffer, dvDesc,\n- 1308 numPatchCoords, patchCoords,\n- 1309 patchTable);\n- 1310 delete instance;\n- 1311 return r;\n- 1312 }\n- 1313 return false;\n- 1314 }\n- 1315 }\n- 1316\n- 1353 template \n-1355 bool EvalPatchesVarying(\n- 1356 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1357 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1358 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1359 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1360 int numPatchCoords,\n- 1361 PATCHCOORD_BUFFER *patchCoords,\n- 1362 PATCH_TABLE *patchTable) const {\n- 1363\n- 1364 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1365 dstBuffer->BindVBO(), dstDesc,\n- 1366 duBuffer->BindVBO(), duDesc,\n- 1367 dvBuffer->BindVBO(), dvDesc,\n- 1368 numPatchCoords,\n- 1369 patchCoords->BindVBO(),\n- 1370 patchTable->GetVaryingPatchArrays(),\n- 1371 patchTable->GetVaryingPatchIndexBuffer(),\n- 1372 patchTable->GetPatchParamBuffer());\n- 1373 }\n- 1374\n- 1437 template \n-1439 static bool EvalPatchesVarying(\n- 1440 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1441 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1442 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1443 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1444 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1445 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1446 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1447 int numPatchCoords,\n- 1448 PATCHCOORD_BUFFER *patchCoords,\n- 1449 PATCH_TABLE *patchTable,\n- 1450 GLComputeEvaluator const *instance,\n- 1451 void * deviceContext = NULL) {\n- 1452\n- 1453 if (instance) {\n- 1454 return instance->EvalPatchesVarying(\n- 1455 srcBuffer, srcDesc,\n- 1456 dstBuffer, dstDesc,\n- 1457 duBuffer, duDesc,\n- 1458 dvBuffer, dvDesc,\n- 1459 duuBuffer, duuDesc,\n- 1460 duvBuffer, duvDesc,\n- 1461 dvvBuffer, dvvDesc,\n- 1462 numPatchCoords, patchCoords,\n- 1463 patchTable);\n- 1464 } else {\n- 1465 // Create an instance on demand (slow)\n- 1466 (void)deviceContext; // unused\n- 1467 instance = Create(srcDesc, dstDesc,\n- 1468 duDesc, dvDesc,\n- 1469 duuDesc, duvDesc, dvvDesc);\n- 1470 if (instance) {\n- 1471 bool r = instance->EvalPatchesVarying(\n- 1472 srcBuffer, srcDesc,\n- 1473 dstBuffer, dstDesc,\n- 1474 duBuffer, duDesc,\n- 1475 dvBuffer, dvDesc,\n- 1476 duuBuffer, duuDesc,\n- 1477 duvBuffer, duvDesc,\n- 1478 dvvBuffer, dvvDesc,\n- 1479 numPatchCoords, patchCoords,\n- 1480 patchTable);\n- 1481 delete instance;\n- 1482 return r;\n- 1483 }\n- 1484 return false;\n- 1485 }\n- 1486 }\n- 1487\n- 1542 template \n-1544 bool EvalPatchesVarying(\n- 1545 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1546 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1547 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1548 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1549 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1550 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1551 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1552 int numPatchCoords,\n- 1553 PATCHCOORD_BUFFER *patchCoords,\n- 1554 PATCH_TABLE *patchTable) const {\n- 1555\n- 1556 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1557 dstBuffer->BindVBO(), dstDesc,\n- 1558 duBuffer->BindVBO(), duDesc,\n- 1559 dvBuffer->BindVBO(), dvDesc,\n- 1560 duuBuffer->BindVBO(), duuDesc,\n- 1561 duvBuffer->BindVBO(), duvDesc,\n- 1562 dvvBuffer->BindVBO(), dvvDesc,\n- 1563 numPatchCoords,\n- 1564 patchCoords->BindVBO(),\n- 1565 patchTable->GetVaryingPatchArrays(),\n- 1566 patchTable->GetVaryingPatchIndexBuffer(),\n- 1567 patchTable->GetPatchParamBuffer());\n- 1568 }\n- 1569\n- 1604 template \n-1606 static bool EvalPatchesFaceVarying(\n- 1607 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1608 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1609 int numPatchCoords,\n- 1610 PATCHCOORD_BUFFER *patchCoords,\n- 1611 PATCH_TABLE *patchTable,\n- 1612 int fvarChannel,\n- 1613 GLComputeEvaluator const *instance,\n- 1614 void * deviceContext = NULL) {\n- 1615\n- 1616 if (instance) {\n- 1617 return instance->EvalPatchesFaceVarying(\n- 1618 srcBuffer, srcDesc,\n- 1619 dstBuffer, dstDesc,\n- 1620 numPatchCoords, patchCoords,\n- 1621 patchTable, fvarChannel);\n- 1622 } else {\n- 1623 // Create an instance on demand (slow)\n- 1624 (void)deviceContext; // unused\n- 1625 instance = Create(srcDesc, dstDesc,\n- 1626 BufferDescriptor(),\n- 1627 BufferDescriptor());\n- 1628 if (instance) {\n- 1629 bool r = instance->EvalPatchesFaceVarying(\n- 1630 srcBuffer, srcDesc,\n- 1631 dstBuffer, dstDesc,\n- 1632 numPatchCoords, patchCoords,\n- 1633 patchTable, fvarChannel);\n- 1634 delete instance;\n- 1635 return r;\n- 1636 }\n- 1637 return false;\n- 1638 }\n- 1639 }\n- 1640\n- 1667 template \n-1669 bool EvalPatchesFaceVarying(\n- 1670 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1671 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1672 int numPatchCoords,\n- 1673 PATCHCOORD_BUFFER *patchCoords,\n- 1674 PATCH_TABLE *patchTable,\n- 1675 int fvarChannel = 0) const {\n- 1676\n- 1677 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1678 dstBuffer->BindVBO(), dstDesc,\n- 1679 0, BufferDescriptor(),\n- 1680 0, BufferDescriptor(),\n- 1681 numPatchCoords,\n- 1682 patchCoords->BindVBO(),\n- 1683 patchTable->GetFVarPatchArrays(fvarChannel),\n- 1684 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1685 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 1686 }\n- 1687\n- 1734 template \n-1736 static bool EvalPatchesFaceVarying(\n- 1737 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1738 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1739 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1740 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1741 int numPatchCoords,\n- 1742 PATCHCOORD_BUFFER *patchCoords,\n- 1743 PATCH_TABLE *patchTable,\n- 1744 int fvarChannel,\n- 1745 GLComputeEvaluator const *instance,\n- 1746 void * deviceContext = NULL) {\n- 1747\n- 1748 if (instance) {\n- 1749 return instance->EvalPatchesFaceVarying(\n- 1750 srcBuffer, srcDesc,\n- 1751 dstBuffer, dstDesc,\n- 1752 duBuffer, duDesc,\n- 1753 dvBuffer, dvDesc,\n- 1754 numPatchCoords, patchCoords,\n- 1755 patchTable, fvarChannel);\n- 1756 } else {\n- 1757 // Create an instance on demand (slow)\n- 1758 (void)deviceContext; // unused\n- 1759 instance = Create(srcDesc, dstDesc,\n- 1760 duDesc, dvDesc);\n- 1761 if (instance) {\n- 1762 bool r = instance->EvalPatchesFaceVarying(\n- 1763 srcBuffer, srcDesc,\n- 1764 dstBuffer, dstDesc,\n- 1765 duBuffer, duDesc,\n- 1766 dvBuffer, dvDesc,\n- 1767 numPatchCoords, patchCoords,\n- 1768 patchTable, fvarChannel);\n- 1769 delete instance;\n- 1770 return r;\n- 1771 }\n- 1772 return false;\n- 1773 }\n- 1774 }\n- 1775\n- 1814 template \n-1816 bool EvalPatchesFaceVarying(\n- 1817 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1818 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1819 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1820 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1821 int numPatchCoords,\n- 1822 PATCHCOORD_BUFFER *patchCoords,\n- 1823 PATCH_TABLE *patchTable,\n- 1824 int fvarChannel = 0) const {\n- 1825\n- 1826 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1827 dstBuffer->BindVBO(), dstDesc,\n- 1828 duBuffer->BindVBO(), duDesc,\n- 1829 dvBuffer->BindVBO(), dvDesc,\n- 1830 numPatchCoords,\n- 1831 patchCoords->BindVBO(),\n- 1832 patchTable->GetFVarPatchArrays(fvarChannel),\n- 1833 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1834 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 1835 }\n- 1836\n- 1901 template \n-1903 static bool EvalPatchesFaceVarying(\n- 1904 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1905 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1906 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1907 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1908 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1909 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1910 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1911 int numPatchCoords,\n- 1912 PATCHCOORD_BUFFER *patchCoords,\n- 1913 PATCH_TABLE *patchTable,\n- 1914 int fvarChannel,\n- 1915 GLComputeEvaluator const *instance,\n- 1916 void * deviceContext = NULL) {\n- 1917\n- 1918 if (instance) {\n- 1919 return instance->EvalPatchesFaceVarying(\n- 1920 srcBuffer, srcDesc,\n- 1921 dstBuffer, dstDesc,\n- 1922 duBuffer, duDesc,\n- 1923 dvBuffer, dvDesc,\n- 1924 duuBuffer, duuDesc,\n- 1925 duvBuffer, duvDesc,\n- 1926 dvvBuffer, dvvDesc,\n- 1927 numPatchCoords, patchCoords,\n- 1928 patchTable, fvarChannel);\n- 1929 } else {\n- 1930 // Create an instance on demand (slow)\n- 1931 (void)deviceContext; // unused\n- 1932 instance = Create(srcDesc, dstDesc,\n- 1933 duDesc, dvDesc,\n- 1934 duuDesc, duvDesc, dvvDesc);\n- 1935 if (instance) {\n- 1936 bool r = instance->EvalPatchesFaceVarying(\n- 1937 srcBuffer, srcDesc,\n- 1938 dstBuffer, dstDesc,\n- 1939 duBuffer, duDesc,\n- 1940 dvBuffer, dvDesc,\n- 1941 duuBuffer, duuDesc,\n- 1942 duvBuffer, duvDesc,\n- 1943 dvvBuffer, dvvDesc,\n- 1944 numPatchCoords, patchCoords,\n- 1945 patchTable, fvarChannel);\n- 1946 delete instance;\n- 1947 return r;\n- 1948 }\n- 1949 return false;\n- 1950 }\n- 1951 }\n- 1952\n- 2009 template \n-2011 bool EvalPatchesFaceVarying(\n- 2012 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 2013 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 2014 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 2015 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 2016 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 2017 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 2018 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 2019 int numPatchCoords,\n- 2020 PATCHCOORD_BUFFER *patchCoords,\n- 2021 PATCH_TABLE *patchTable,\n- 2022 int fvarChannel = 0) const {\n- 2023\n- 2024 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 2025 dstBuffer->BindVBO(), dstDesc,\n- 2026 duBuffer->BindVBO(), duDesc,\n- 2027 dvBuffer->BindVBO(), dvDesc,\n- 2028 duuBuffer->BindVBO(), duuDesc,\n- 2029 duvBuffer->BindVBO(), duvDesc,\n- 2030 dvvBuffer->BindVBO(), dvvDesc,\n- 2031 numPatchCoords,\n- 2032 patchCoords->BindVBO(),\n- 2033 patchTable->GetFVarPatchArrays(fvarChannel),\n- 2034 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 2035 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 2036 }\n- 2037\n- 2043\n-2046 bool Compile(BufferDescriptor const &srcDesc,\n- 2047 BufferDescriptor const &dstDesc,\n- 2048 BufferDescriptor const &duDesc = BufferDescriptor(),\n- 2049 BufferDescriptor const &dvDesc = BufferDescriptor(),\n- 2050 BufferDescriptor const &duuDesc = BufferDescriptor(),\n- 2051 BufferDescriptor const &duvDesc = BufferDescriptor(),\n- 2052 BufferDescriptor const &dvvDesc = BufferDescriptor());\n- 2053\n-2055 static void Synchronize(void *deviceContext);\n- 2056\n- 2057private:\n- 2058 struct _StencilKernel {\n- 2059 _StencilKernel();\n- 2060 ~_StencilKernel();\n- 2061 bool Compile(BufferDescriptor const &srcDesc,\n- 2062 BufferDescriptor const &dstDesc,\n- 2063 BufferDescriptor const &duDesc,\n- 2064 BufferDescriptor const &dvDesc,\n- 2065 BufferDescriptor const &duuDesc,\n- 2066 BufferDescriptor const &duvDesc,\n- 2067 BufferDescriptor const &dvvDesc,\n- 2068 int workGroupSize);\n- 2069 GLuint program;\n- 2070 GLuint uniformStart;\n- 2071 GLuint uniformEnd;\n- 2072 GLuint uniformSrcOffset;\n- 2073 GLuint uniformDstOffset;\n- 2074 GLuint uniformDuDesc;\n- 2075 GLuint uniformDvDesc;\n- 2076 GLuint uniformDuuDesc;\n- 2077 GLuint uniformDuvDesc;\n- 2078 GLuint uniformDvvDesc;\n- 2079 } _stencilKernel;\n- 2080\n- 2081 struct _PatchKernel {\n- 2082 _PatchKernel();\n- 2083 ~_PatchKernel();\n- 2084 bool Compile(BufferDescriptor const &srcDesc,\n- 2085 BufferDescriptor const &dstDesc,\n- 2086 BufferDescriptor const &duDesc,\n- 2087 BufferDescriptor const &dvDesc,\n- 2088 BufferDescriptor const &duuDesc,\n- 2089 BufferDescriptor const &duvDesc,\n- 2090 BufferDescriptor const &dvvDesc,\n- 2091 int workGroupSize);\n- 2092 GLuint program;\n- 2093 GLuint uniformSrcOffset;\n- 2094 GLuint uniformDstOffset;\n- 2095 GLuint uniformPatchArray;\n- 2096 GLuint uniformDuDesc;\n- 2097 GLuint uniformDvDesc;\n- 2098 GLuint uniformDuuDesc;\n- 2099 GLuint uniformDuvDesc;\n- 2100 GLuint uniformDvvDesc;\n- 2101 } _patchKernel;\n- 2102\n- 2103 int _workGroupSize;\n- 2104 GLuint _patchArraysSSBO;\n- 2105};\n- 2106\n- 2107} // end namespace Osd\n- 2108\n- 2109} // end namespace OPENSUBDIV_VERSION\n- 2110using namespace OPENSUBDIV_VERSION;\n- 2111\n- 2112} // end namespace OpenSubdiv\n- 2113\n- 2114\n- 2115#endif // OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H\n+ 94} // end namespace OPENSUBDIV_VERSION\n+ 95using namespace OPENSUBDIV_VERSION;\n+ 96\n+ 97} // end namespace OpenSubdiv\n+ 98\n+ 99#endif // OPENSUBDIV3_OSD_CPU_D3D11_VERTEX_BUFFER_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n-std::vector< PatchArray > PatchArrayVector\n-Definition: types.h:132\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n-Stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:273\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n-Limit stencil table class wrapping the template for compatibility.\n-Definition: stencilTable.h:583\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n-BufferDescriptor is a struct which describes buffer elements in interleaved\n-data buffers....\n-Definition: bufferDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO\n-GL stencil table (Shader Storage buffer)\n-Definition: glComputeEvaluator.h:51\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetDuvWeightsBuffer\n-GLuint GetDuvWeightsBuffer() const\n-Definition: glComputeEvaluator.h:77\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetWeightsBuffer\n-GLuint GetWeightsBuffer() const\n-Definition: glComputeEvaluator.h:73\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GLStencilTableSSBO\n-GLStencilTableSSBO(Far::LimitStencilTable const *limitStencilTable)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetDuWeightsBuffer\n-GLuint GetDuWeightsBuffer() const\n-Definition: glComputeEvaluator.h:74\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::~GLStencilTableSSBO\n-~GLStencilTableSSBO()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetNumStencils\n-int GetNumStencils() const\n-Definition: glComputeEvaluator.h:79\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetIndicesBuffer\n-GLuint GetIndicesBuffer() const\n-Definition: glComputeEvaluator.h:72\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetOffsetsBuffer\n-GLuint GetOffsetsBuffer() const\n-Definition: glComputeEvaluator.h:71\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GLStencilTableSSBO\n-GLStencilTableSSBO(Far::StencilTable const *stencilTable)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetDvvWeightsBuffer\n-GLuint GetDvvWeightsBuffer() const\n-Definition: glComputeEvaluator.h:78\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetSizesBuffer\n-GLuint GetSizesBuffer() const\n-Definition: glComputeEvaluator.h:70\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::Create\n-static GLStencilTableSSBO * Create(Far::LimitStencilTable const\n-*limitStencilTable, void *deviceContext=NULL)\n-Definition: glComputeEvaluator.h:58\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetDuuWeightsBuffer\n-GLuint GetDuuWeightsBuffer() const\n-Definition: glComputeEvaluator.h:76\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetDvWeightsBuffer\n-GLuint GetDvWeightsBuffer() const\n-Definition: glComputeEvaluator.h:75\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::Create\n-static GLStencilTableSSBO * Create(Far::StencilTable const *stencilTable, void\n-*deviceContext=NULL)\n-Definition: glComputeEvaluator.h:53\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator\n-Definition: glComputeEvaluator.h:96\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n-bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:1214\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLComputeEvaluator\n-const *instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:1736\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n-bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, STENCIL_TABLE const *stencilTable) const\n-Generic stencil function.\n-Definition: glComputeEvaluator.h:434\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n-GLComputeEvaluator const *instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:1154\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n-fvarChannel, GLComputeEvaluator const *instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:1606\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n-bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable) const\n-Generic stencil function.\n-Definition: glComputeEvaluator.h:502\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void\n-*deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:1278\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Instantiatable\n-bool Instantiatable\n-Definition: glComputeEvaluator.h:98\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n-bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n-dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n-const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint\n-duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor\n-const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, int\n-numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays,\n-GLuint patchIndexBuffer, GLuint patchParamsBuffer) const\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::~GLComputeEvaluator\n-~GLComputeEvaluator()\n-Destructor. note that the GL context must be made current.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n-bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable) const\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: glComputeEvaluator.h:1071\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable,\n-GLComputeEvaluator const *instance, void *deviceContext=NULL)\n-Generic static stencil function. This function has a same signature as other\n-device kernels have so t...\n-Definition: glComputeEvaluator.h:236\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n-bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n-dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n-const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint\n-duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor\n-const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, GLuint\n-sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer,\n-GLuint duWeightsBuffer, GLuint dvWeightsBuffer, GLuint duuWeightsBuffer, GLuint\n-duvWeightsBuffer, GLuint dvvWeightsBuffer, int start, int end) const\n-Dispatch the GLSL compute kernel on GPU asynchronously returns false if the\n-kernel hasn't been compil...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n-const *stencilTable, GLComputeEvaluator const *instance, void\n-*deviceContext=NULL)\n-Generic static stencil function. This function has a same signature as other\n-device kernels have so t...\n-Definition: glComputeEvaluator.h:168\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator\n-const *instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:1439\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:769\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n-GLComputeEvaluator const *instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:1903\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n-bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:1544\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n-bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:1816\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n-bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:943\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::GLComputeEvaluator\n-GLComputeEvaluator()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer\n+Concrete vertex buffer class for Cpu subdivision and DirectX drawing.\n+Definition: cpuD3D11VertexBuffer.h:48\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::Create\n+static CpuD3D11VertexBuffer * Create(int numElements, int numVertices,\n+ID3D11DeviceContext *deviceContext)\n+Creator. Returns NULL if error.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::CpuD3D11VertexBuffer\n+CpuD3D11VertexBuffer(int numElements, int numVertices)\n Constructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n-bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable) const\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: glComputeEvaluator.h:997\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator\n-const *instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:691\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n-bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const\n-*stencilTable) const\n-Generic stencil function.\n-Definition: glComputeEvaluator.h:386\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n-bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n-dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n-const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, int\n-numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays,\n-GLuint patchIndexBuffer, GLuint patchParamsBuffer) const\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n-bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0)\n-const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:2011\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Create\n-static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, void *deviceContext=NULL)\n-Definition: glComputeEvaluator.h:99\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n-bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n-dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n-const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint\n-sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer,\n-GLuint duWeightsBuffer, GLuint dvWeightsBuffer, int start, int end) const\n-Dispatch the GLSL compute kernel on GPU asynchronously returns false if the\n-kernel hasn't been compil...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n-GLComputeEvaluator const *instance, void *deviceContext=NULL)\n-Generic static stencil function. This function has a same signature as other\n-device kernels have so t...\n-Definition: glComputeEvaluator.h:326\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Synchronize\n-static void Synchronize(void *deviceContext)\n-Wait the dispatched kernel finishes.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Compile\n-bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc,\n-BufferDescriptor const &duDesc=BufferDescriptor(), BufferDescriptor const\n-&dvDesc=BufferDescriptor(), BufferDescriptor const &duuDesc=BufferDescriptor(),\n-BufferDescriptor const &duvDesc=BufferDescriptor(), BufferDescriptor const\n-&dvvDesc=BufferDescriptor())\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n-bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:1355\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Create\n-static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n-BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, void\n-*deviceContext=NULL)\n-Definition: glComputeEvaluator.h:111\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n-bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n-fvarChannel=0) const\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:1669\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::GetNumElements\n+int GetNumElements() const\n+Returns how many elements defined in this vertex buffer.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::UpdateData\n+void UpdateData(const float *src, int startVertex, int numVertices, void\n *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glComputeEvaluator.h:870\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::GetNumVertices\n+int GetNumVertices() const\n+Returns how many vertices allocated in this vertex buffer.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindD3D11Buffer\n+ID3D11Buffer * BindD3D11Buffer(ID3D11DeviceContext *deviceContext)\n+Returns the D3D11 buffer object.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindCpuBuffer\n+float * BindCpuBuffer()\n+Returns the address of CPU buffer.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::allocate\n+bool allocate(ID3D11Device *device)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::\n+~CpuD3D11VertexBuffer\n+virtual ~CpuD3D11VertexBuffer()\n+Destructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindVBO\n+ID3D11Buffer * BindVBO(ID3D11DeviceContext *deviceContext)\n+Returns the D3D11 buffer object (for Osd::Mesh interface)\n+Definition: cpuD3D11VertexBuffer.h:75\n * opensubdiv\n * osd\n- * glComputeEvaluator.h\n+ * cpuD3D11VertexBuffer.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00851.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00851.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glXFBEvaluator.h File Reference\n+OpenSubdiv: opensubdiv/osd/glComputeEvaluator.h File Reference\n \n \n \n \n \n \n \n@@ -83,30 +83,30 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
glXFBEvaluator.h File Reference
\n+
glComputeEvaluator.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../osd/opengl.h"
\n+#include "../osd/opengl.h"
\n #include "../osd/types.h"
\n-#include "../osd/bufferDescriptor.h"
\n+#include "../osd/bufferDescriptor.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n+\n+\n \n-\n+\n \n

\n Classes

class  GLStencilTableTBO
 GL TextureBuffer stencil table. More...
class  GLStencilTableSSBO
 GL stencil table (Shader Storage buffer) More...
 
class  GLXFBEvaluator
class  GLComputeEvaluator
 
\n \n \n \n \n@@ -117,13 +117,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,32 +5,32 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-glXFBEvaluator.h File Reference\n+glComputeEvaluator.h File Reference\n #include \"../version.h\"\n #include \"../osd/opengl.h\"\n #include \"../osd/types.h\"\n #include \"../osd/bufferDescriptor.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0GLStencilTableTBO\n-\u00a0 GL TextureBuffer stencil table. More...\n+class \u00a0GLStencilTableSSBO\n+\u00a0 GL stencil table (Shader Storage buffer) More...\n \u00a0\n-class \u00a0GLXFBEvaluator\n+class \u00a0GLComputeEvaluator\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * glXFBEvaluator.h\n+ * glComputeEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00851.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00851.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,4 @@\n var a00851 = [\n- [\"GLStencilTableTBO\", \"a01241.html\", \"a01241\"],\n- [\"GLXFBEvaluator\", \"a01245.html\", \"a01245\"]\n+ [\"GLStencilTableSSBO\", \"a01209.html\", \"a01209\"],\n+ [\"GLComputeEvaluator\", \"a01213.html\", \"a01213\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00851_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00851_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glXFBEvaluator.h Source File\n+OpenSubdiv: opensubdiv/osd/glComputeEvaluator.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
glXFBEvaluator.h
\n+
glComputeEvaluator.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,16 +107,16 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H
\n-
26#define OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H
\n+
25#ifndef OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H
\n+
26#define OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H
\n
27
\n
28#include "../version.h"
\n
29
\n
30#include "../osd/opengl.h"
\n
31#include "../osd/types.h"
\n
32#include "../osd/bufferDescriptor.h"
\n
33
\n@@ -127,1117 +127,1052 @@\n
38 class PatchTable;
\n
39 class StencilTable;
\n
40 class LimitStencilTable;
\n
41}
\n
42
\n
43namespace Osd {
\n
44
\n-
52class GLStencilTableTBO {
\n-
53public:
\n-
54 static GLStencilTableTBO *Create(
\n-
55 Far::StencilTable const *stencilTable, void *deviceContext = NULL) {
\n-
56 (void)deviceContext; // unused
\n-
57 return new GLStencilTableTBO(stencilTable);
\n-
58 }
\n-
59
\n-
60 static GLStencilTableTBO *Create(
\n-
61 Far::LimitStencilTable const *limitStencilTable,
\n-
62 void *deviceContext = NULL) {
\n-
63 (void)deviceContext; // unused
\n-
64 return new GLStencilTableTBO(limitStencilTable);
\n-
65 }
\n-
66
\n-
67 explicit GLStencilTableTBO(Far::StencilTable const *stencilTable);
\n-
68 explicit GLStencilTableTBO(Far::LimitStencilTable const *limitStencilTable);
\n-
69 ~GLStencilTableTBO();
\n-
70
\n-
71 // interfaces needed for GLSLTransformFeedbackKernel
\n-
72 GLuint GetSizesTexture() const { return _sizes; }
\n-
73 GLuint GetOffsetsTexture() const { return _offsets; }
\n-
74 GLuint GetIndicesTexture() const { return _indices; }
\n-
75 GLuint GetWeightsTexture() const { return _weights; }
\n-
76 GLuint GetDuWeightsTexture() const { return _duWeights; }
\n-
77 GLuint GetDvWeightsTexture() const { return _dvWeights; }
\n-
78 GLuint GetDuuWeightsTexture() const { return _duuWeights; }
\n-
79 GLuint GetDuvWeightsTexture() const { return _duvWeights; }
\n-
80 GLuint GetDvvWeightsTexture() const { return _dvvWeights; }
\n-
81 int GetNumStencils() const { return _numStencils; }
\n-
82
\n-
83private:
\n-
84 GLuint _sizes;
\n-
85 GLuint _offsets;
\n-
86 GLuint _indices;
\n-
87 GLuint _weights;
\n-
88 GLuint _duWeights;
\n-
89 GLuint _dvWeights;
\n-
90 GLuint _duuWeights;
\n-
91 GLuint _duvWeights;
\n-
92 GLuint _dvvWeights;
\n-
93 int _numStencils;
\n-
94};
\n+
51class GLStencilTableSSBO {
\n+
52public:
\n+
53 static GLStencilTableSSBO *Create(Far::StencilTable const *stencilTable,
\n+
54 void *deviceContext = NULL) {
\n+
55 (void)deviceContext; // unused
\n+
56 return new GLStencilTableSSBO(stencilTable);
\n+
57 }
\n+
58 static GLStencilTableSSBO *Create(
\n+
59 Far::LimitStencilTable const *limitStencilTable,
\n+
60 void *deviceContext = NULL) {
\n+
61 (void)deviceContext; // unused
\n+
62 return new GLStencilTableSSBO(limitStencilTable);
\n+
63 }
\n+
64
\n+
65 explicit GLStencilTableSSBO(Far::StencilTable const *stencilTable);
\n+
66 explicit GLStencilTableSSBO(Far::LimitStencilTable const *limitStencilTable);
\n+
67 ~GLStencilTableSSBO();
\n+
68
\n+
69 // interfaces needed for GLSLComputeKernel
\n+
70 GLuint GetSizesBuffer() const { return _sizes; }
\n+
71 GLuint GetOffsetsBuffer() const { return _offsets; }
\n+
72 GLuint GetIndicesBuffer() const { return _indices; }
\n+
73 GLuint GetWeightsBuffer() const { return _weights; }
\n+
74 GLuint GetDuWeightsBuffer() const { return _duWeights; }
\n+
75 GLuint GetDvWeightsBuffer() const { return _dvWeights; }
\n+
76 GLuint GetDuuWeightsBuffer() const { return _duuWeights; }
\n+
77 GLuint GetDuvWeightsBuffer() const { return _duvWeights; }
\n+
78 GLuint GetDvvWeightsBuffer() const { return _dvvWeights; }
\n+
79 int GetNumStencils() const { return _numStencils; }
\n+
80
\n+
81private:
\n+
82 GLuint _sizes;
\n+
83 GLuint _offsets;
\n+
84 GLuint _indices;
\n+
85 GLuint _weights;
\n+
86 GLuint _duWeights;
\n+
87 GLuint _dvWeights;
\n+
88 GLuint _duuWeights;
\n+
89 GLuint _duvWeights;
\n+
90 GLuint _dvvWeights;
\n+
91 int _numStencils;
\n+
92};
\n+
93
\n+
94// ---------------------------------------------------------------------------
\n
95
\n-
96// ---------------------------------------------------------------------------
\n-
97
\n-
98class GLXFBEvaluator {
\n-
99public:
\n-
100 typedef bool Instantiatable;
\n-
101
\n-
103 template <typename DEVICE_CONTEXT>
\n-
104 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,
\n-
105 BufferDescriptor const &dstDesc,
\n-
106 BufferDescriptor const &duDesc,
\n-
107 BufferDescriptor const &dvDesc,
\n-
108 DEVICE_CONTEXT deviceContext) {
\n-
109 bool interleavedDerivativeBuffers = deviceContext
\n-
110 ? deviceContext->AreInterleavedDerivativeBuffers()
\n-
111 : false;
\n-
112 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n-
113 interleavedDerivativeBuffers);
\n-
114 }
\n-
115
\n-
117 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,
\n-
118 BufferDescriptor const &dstDesc,
\n-
119 BufferDescriptor const &duDesc,
\n-
120 BufferDescriptor const &dvDesc,
\n-
121 void * deviceContext) {
\n-
122 (void)deviceContext; // not used
\n-
123 return Create(srcDesc, dstDesc, duDesc, dvDesc);
\n-
124 }
\n-
125
\n-
126 static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,
\n-
127 BufferDescriptor const &dstDesc,
\n-
128 BufferDescriptor const &duDesc,
\n-
129 BufferDescriptor const &dvDesc,
\n-
130 bool interleavedDerivativeBuffers = false) {
\n-
131 GLXFBEvaluator *instance = new GLXFBEvaluator(interleavedDerivativeBuffers);
\n-
132 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc))
\n-
133 return instance;
\n-
134 delete instance;
\n-
135 return NULL;
\n-
136 }
\n-
137
\n-
139 template <typename DEVICE_CONTEXT>
\n-
140 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,
\n-
141 BufferDescriptor const &dstDesc,
\n-
142 BufferDescriptor const &duDesc,
\n-
143 BufferDescriptor const &dvDesc,
\n-
144 BufferDescriptor const &duuDesc,
\n-
145 BufferDescriptor const &duvDesc,
\n-
146 BufferDescriptor const &dvvDesc,
\n-
147 DEVICE_CONTEXT deviceContext) {
\n-
148 bool interleavedDerivativeBuffers = deviceContext
\n-
149 ? deviceContext->AreInterleavedDerivativeBuffers()
\n-
150 : false;
\n-
151 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n-
152 duuDesc, duvDesc, dvvDesc,
\n-
153 interleavedDerivativeBuffers);
\n-
154 }
\n-
155
\n-
157 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,
\n-
158 BufferDescriptor const &dstDesc,
\n-
159 BufferDescriptor const &duDesc,
\n-
160 BufferDescriptor const &dvDesc,
\n-
161 BufferDescriptor const &duuDesc,
\n-
162 BufferDescriptor const &duvDesc,
\n-
163 BufferDescriptor const &dvvDesc,
\n-
164 void * deviceContext) {
\n-
165 (void)deviceContext; // not used
\n-
166 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n-
167 duuDesc, duvDesc, dvvDesc);
\n-
168 }
\n-
169
\n-
170 static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,
\n-
171 BufferDescriptor const &dstDesc,
\n-
172 BufferDescriptor const &duDesc,
\n-
173 BufferDescriptor const &dvDesc,
\n-
174 BufferDescriptor const &duuDesc,
\n-
175 BufferDescriptor const &duvDesc,
\n-
176 BufferDescriptor const &dvvDesc,
\n-
177 bool interleavedDerivativeBuffers = false) {
\n-
178 GLXFBEvaluator *instance = new GLXFBEvaluator(interleavedDerivativeBuffers);
\n-
179 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc,
\n-
180 duuDesc, duvDesc, dvvDesc))
\n-
181 return instance;
\n-
182 delete instance;
\n-
183 return NULL;
\n-
184 }
\n-
185
\n-
199 GLXFBEvaluator(bool interleavedDerivativeBuffers = false);
\n-
200
\n-
202 ~GLXFBEvaluator();
\n-
203
\n-
209
\n-
237 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
238 static bool EvalStencils(
\n-
239 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
240 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
96class GLComputeEvaluator {
\n+
97public:
\n+
98 typedef bool Instantiatable;
\n+
99 static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n+
100 BufferDescriptor const &dstDesc,
\n+
101 BufferDescriptor const &duDesc,
\n+
102 BufferDescriptor const &dvDesc,
\n+
103 void * deviceContext = NULL) {
\n+
104 return Create(srcDesc, dstDesc, duDesc, dvDesc,
\n+
105 BufferDescriptor(),
\n+
106 BufferDescriptor(),
\n+
107 BufferDescriptor(),
\n+
108 deviceContext);
\n+
109 }
\n+
110
\n+
111 static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc,
\n+
112 BufferDescriptor const &dstDesc,
\n+
113 BufferDescriptor const &duDesc,
\n+
114 BufferDescriptor const &dvDesc,
\n+
115 BufferDescriptor const &duuDesc,
\n+
116 BufferDescriptor const &duvDesc,
\n+
117 BufferDescriptor const &dvvDesc,
\n+
118 void * deviceContext = NULL) {
\n+
119 (void)deviceContext; // not used
\n+
120 GLComputeEvaluator *instance = new GLComputeEvaluator();
\n+
121 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc,
\n+
122 duuDesc, duvDesc, dvvDesc))
\n+
123 return instance;
\n+
124 delete instance;
\n+
125 return NULL;
\n+
126 }
\n+
127
\n+
129 GLComputeEvaluator();
\n+
130
\n+
132 ~GLComputeEvaluator();
\n+
133
\n+
139
\n+
167 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
168 static bool EvalStencils(
\n+
169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
171 STENCIL_TABLE const *stencilTable,
\n+
172 GLComputeEvaluator const *instance,
\n+
173 void * deviceContext = NULL) {
\n+
174
\n+
175 if (instance) {
\n+
176 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
177 dstBuffer, dstDesc,
\n+
178 stencilTable);
\n+
179 } else {
\n+
180 // Create an instance on demand (slow)
\n+
181 (void)deviceContext; // unused
\n+
182 instance = Create(srcDesc, dstDesc,
\n+
183 BufferDescriptor(),
\n+
184 BufferDescriptor());
\n+
185 if (instance) {
\n+
186 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
187 dstBuffer, dstDesc,
\n+
188 stencilTable);
\n+
189 delete instance;
\n+
190 return r;
\n+
191 }
\n+
192 return false;
\n+
193 }
\n+
194 }
\n+
195
\n+
235 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
236 static bool EvalStencils(
\n+
237 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
238 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
239 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
240 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n
241 STENCIL_TABLE const *stencilTable,
\n-
242 GLXFBEvaluator const *instance,
\n+
242 GLComputeEvaluator const *instance,
\n
243 void * deviceContext = NULL) {
\n
244
\n
245 if (instance) {
\n-
246 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
246 return instance->EvalStencils(srcBuffer, srcDesc,
\n
247 dstBuffer, dstDesc,
\n-
248 stencilTable);
\n-
249 } else {
\n-
250 // Create an instance on demand (slow)
\n-
251 (void)deviceContext; // unused
\n-
252 instance = Create(srcDesc, dstDesc,
\n-
253 BufferDescriptor(),
\n-
254 BufferDescriptor());
\n+
248 duBuffer, duDesc,
\n+
249 dvBuffer, dvDesc,
\n+
250 stencilTable);
\n+
251 } else {
\n+
252 // Create an instance on demand (slow)
\n+
253 (void)deviceContext; // unused
\n+
254 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);
\n
255 if (instance) {
\n-
256 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
256 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n
257 dstBuffer, dstDesc,
\n-
258 stencilTable);
\n-
259 delete instance;
\n-
260 return r;
\n-
261 }
\n-
262 return false;
\n-
263 }
\n-
264 }
\n-
265
\n-
305 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
306 static bool EvalStencils(
\n-
307 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
308 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
309 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
310 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
311 STENCIL_TABLE const *stencilTable,
\n-
312 GLXFBEvaluator const *instance,
\n-
313 void * deviceContext = NULL) {
\n-
314
\n-
315 if (instance) {
\n-
316 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
317 dstBuffer, dstDesc,
\n-
318 duBuffer, duDesc,
\n-
319 dvBuffer, dvDesc,
\n-
320 stencilTable);
\n-
321 } else {
\n-
322 // Create an instance on demand (slow)
\n-
323 (void)deviceContext; // unused
\n-
324 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);
\n-
325 if (instance) {
\n-
326 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
327 dstBuffer, dstDesc,
\n-
328 duBuffer, duDesc,
\n-
329 dvBuffer, dvDesc,
\n-
330 stencilTable);
\n-
331 delete instance;
\n-
332 return r;
\n-
333 }
\n-
334 return false;
\n-
335 }
\n-
336 }
\n+
258 duBuffer, duDesc,
\n+
259 dvBuffer, dvDesc,
\n+
260 stencilTable);
\n+
261 delete instance;
\n+
262 return r;
\n+
263 }
\n+
264 return false;
\n+
265 }
\n+
266 }
\n+
267
\n+
325 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
326 static bool EvalStencils(
\n+
327 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
328 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
329 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
330 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
331 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
332 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
333 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
334 STENCIL_TABLE const *stencilTable,
\n+
335 GLComputeEvaluator const *instance,
\n+
336 void * deviceContext = NULL) {
\n
337
\n-
395 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
396 static bool EvalStencils(
\n-
397 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
398 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
399 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
400 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
401 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
402 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
403 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
404 STENCIL_TABLE const *stencilTable,
\n-
405 GLXFBEvaluator const *instance,
\n-
406 void * deviceContext = NULL) {
\n-
407
\n-
408 if (instance) {
\n-
409 return instance->EvalStencils(srcBuffer, srcDesc,
\n-
410 dstBuffer, dstDesc,
\n-
411 duBuffer, duDesc,
\n-
412 dvBuffer, dvDesc,
\n-
413 duuBuffer, duuDesc,
\n-
414 duvBuffer, duvDesc,
\n-
415 dvvBuffer, dvvDesc,
\n-
416 stencilTable);
\n-
417 } else {
\n-
418 // Create an instance on demand (slow)
\n-
419 (void)deviceContext; // unused
\n-
420 instance = Create(srcDesc, dstDesc,
\n-
421 duDesc, dvDesc,
\n-
422 duuDesc, duvDesc, dvvDesc);
\n-
423 if (instance) {
\n-
424 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n-
425 dstBuffer, dstDesc,
\n-
426 duBuffer, duDesc,
\n-
427 dvBuffer, dvDesc,
\n-
428 duuBuffer, duuDesc,
\n-
429 duvBuffer, duvDesc,
\n-
430 dvvBuffer, dvvDesc,
\n-
431 stencilTable);
\n-
432 delete instance;
\n-
433 return r;
\n-
434 }
\n-
435 return false;
\n-
436 }
\n-
437 }
\n-
438
\n-
456 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
457 bool EvalStencils(
\n-
458 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
459 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
460 STENCIL_TABLE const *stencilTable) const {
\n-
461
\n-
462 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n-
463 dstBuffer->BindVBO(), dstDesc,
\n-
464 0, BufferDescriptor(),
\n-
465 0, BufferDescriptor(),
\n-
466 stencilTable->GetSizesTexture(),
\n-
467 stencilTable->GetOffsetsTexture(),
\n-
468 stencilTable->GetIndicesTexture(),
\n-
469 stencilTable->GetWeightsTexture(),
\n-
470 0,
\n-
471 0,
\n-
472 /* start = */ 0,
\n-
473 /* end = */ stencilTable->GetNumStencils());
\n-
474 }
\n-
475
\n-
505 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
506 bool EvalStencils(
\n-
507 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
508 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
509 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
510 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
511 STENCIL_TABLE const *stencilTable) const {
\n-
512
\n-
513 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n-
514 dstBuffer->BindVBO(), dstDesc,
\n-
515 duBuffer->BindVBO(), duDesc,
\n-
516 dvBuffer->BindVBO(), dvDesc,
\n-
517 stencilTable->GetSizesTexture(),
\n-
518 stencilTable->GetOffsetsTexture(),
\n-
519 stencilTable->GetIndicesTexture(),
\n-
520 stencilTable->GetWeightsTexture(),
\n-
521 stencilTable->GetDuWeightsTexture(),
\n-
522 stencilTable->GetDvWeightsTexture(),
\n-
523 /* start = */ 0,
\n-
524 /* end = */ stencilTable->GetNumStencils());
\n-
525 }
\n-
526
\n-
574 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
575 bool EvalStencils(
\n-
576 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
577 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
578 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
579 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
580 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
581 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
582 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
583 STENCIL_TABLE const *stencilTable) const {
\n-
584
\n-
585 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n-
586 dstBuffer->BindVBO(), dstDesc,
\n-
587 duBuffer->BindVBO(), duDesc,
\n-
588 dvBuffer->BindVBO(), dvDesc,
\n-
589 duuBuffer->BindVBO(), duuDesc,
\n-
590 duvBuffer->BindVBO(), duvDesc,
\n-
591 dvvBuffer->BindVBO(), dvvDesc,
\n-
592 stencilTable->GetSizesTexture(),
\n-
593 stencilTable->GetOffsetsTexture(),
\n-
594 stencilTable->GetIndicesTexture(),
\n-
595 stencilTable->GetWeightsTexture(),
\n-
596 stencilTable->GetDuWeightsTexture(),
\n-
597 stencilTable->GetDvWeightsTexture(),
\n-
598 stencilTable->GetDuuWeightsTexture(),
\n-
599 stencilTable->GetDuvWeightsTexture(),
\n-
600 stencilTable->GetDvvWeightsTexture(),
\n-
601 /* start = */ 0,
\n-
602 /* end = */ stencilTable->GetNumStencils());
\n-
603 }
\n-
604
\n-
640 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n-
641 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n-
642 GLuint duBuffer, BufferDescriptor const &duDesc,
\n-
643 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n-
644 GLuint sizesBuffer,
\n-
645 GLuint offsetsBuffer,
\n-
646 GLuint indicesBuffer,
\n-
647 GLuint weightsBuffer,
\n-
648 GLuint duWeightsBuffer,
\n-
649 GLuint dvWeightsBuffer,
\n-
650 int start,
\n-
651 int end) const;
\n-
652
\n-
706 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n-
707 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n-
708 GLuint duBuffer, BufferDescriptor const &duDesc,
\n-
709 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n-
710 GLuint duuBuffer, BufferDescriptor const &duuDesc,
\n-
711 GLuint duvBuffer, BufferDescriptor const &duvDesc,
\n-
712 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
713 GLuint sizesBuffer,
\n-
714 GLuint offsetsBuffer,
\n-
715 GLuint indicesBuffer,
\n-
716 GLuint weightsBuffer,
\n-
717 GLuint duWeightsBuffer,
\n-
718 GLuint dvWeightsBuffer,
\n-
719 GLuint duuWeightsBuffer,
\n-
720 GLuint duvWeightsBuffer,
\n-
721 GLuint dvvWeightsBuffer,
\n-
722 int start,
\n-
723 int end) const;
\n-
724
\n-
730
\n-
763 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
764 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
765 static bool EvalPatches(
\n-
766 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
767 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
768 int numPatchCoords,
\n-
769 PATCHCOORD_BUFFER *patchCoords,
\n-
770 PATCH_TABLE *patchTable,
\n-
771 GLXFBEvaluator const *instance,
\n-
772 void * deviceContext = NULL) {
\n-
773
\n-
774 if (instance) {
\n-
775 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
776 dstBuffer, dstDesc,
\n-
777 numPatchCoords, patchCoords,
\n-
778 patchTable);
\n-
779 } else {
\n-
780 // Create an instance on demand (slow)
\n-
781 (void)deviceContext; // unused
\n-
782 instance = Create(srcDesc, dstDesc,
\n-
783 BufferDescriptor(),
\n-
784 BufferDescriptor());
\n-
785 if (instance) {
\n-
786 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
787 dstBuffer, dstDesc,
\n-
788 numPatchCoords, patchCoords,
\n-
789 patchTable);
\n-
790 delete instance;
\n-
791 return r;
\n-
792 }
\n-
793 return false;
\n-
794 }
\n-
795 }
\n-
796
\n-
841 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
842 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
843 static bool EvalPatches(
\n-
844 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
845 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
846 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
847 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
848 int numPatchCoords,
\n-
849 PATCHCOORD_BUFFER *patchCoords,
\n-
850 PATCH_TABLE *patchTable,
\n-
851 GLXFBEvaluator const *instance,
\n-
852 void * deviceContext = NULL) {
\n-
853
\n-
854 if (instance) {
\n-
855 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
856 dstBuffer, dstDesc,
\n-
857 duBuffer, duDesc,
\n-
858 dvBuffer, dvDesc,
\n-
859 numPatchCoords, patchCoords,
\n-
860 patchTable);
\n-
861 } else {
\n-
862 // Create an instance on demand (slow)
\n-
863 (void)deviceContext; // unused
\n-
864 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);
\n-
865 if (instance) {
\n-
866 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
867 dstBuffer, dstDesc,
\n-
868 duBuffer, duDesc,
\n-
869 dvBuffer, dvDesc,
\n-
870 numPatchCoords, patchCoords,
\n-
871 patchTable);
\n-
872 delete instance;
\n-
873 return r;
\n-
874 }
\n-
875 return false;
\n-
876 }
\n-
877 }
\n-
878
\n+
338 if (instance) {
\n+
339 return instance->EvalStencils(srcBuffer, srcDesc,
\n+
340 dstBuffer, dstDesc,
\n+
341 duBuffer, duDesc,
\n+
342 dvBuffer, dvDesc,
\n+
343 duuBuffer, duuDesc,
\n+
344 duvBuffer, duvDesc,
\n+
345 dvvBuffer, dvvDesc,
\n+
346 stencilTable);
\n+
347 } else {
\n+
348 // Create an instance on demand (slow)
\n+
349 (void)deviceContext; // unused
\n+
350 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,
\n+
351 duuDesc, duvDesc, dvvDesc);
\n+
352 if (instance) {
\n+
353 bool r = instance->EvalStencils(srcBuffer, srcDesc,
\n+
354 dstBuffer, dstDesc,
\n+
355 duBuffer, duDesc,
\n+
356 dvBuffer, dvDesc,
\n+
357 duuBuffer, duuDesc,
\n+
358 duvBuffer, duvDesc,
\n+
359 dvvBuffer, dvvDesc,
\n+
360 stencilTable);
\n+
361 delete instance;
\n+
362 return r;
\n+
363 }
\n+
364 return false;
\n+
365 }
\n+
366 }
\n+
367
\n+
385 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
386 bool EvalStencils(
\n+
387 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
388 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
389 STENCIL_TABLE const *stencilTable) const {
\n+
390 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n+
391 dstBuffer->BindVBO(), dstDesc,
\n+
392 0, BufferDescriptor(),
\n+
393 0, BufferDescriptor(),
\n+
394 stencilTable->GetSizesBuffer(),
\n+
395 stencilTable->GetOffsetsBuffer(),
\n+
396 stencilTable->GetIndicesBuffer(),
\n+
397 stencilTable->GetWeightsBuffer(),
\n+
398 0,
\n+
399 0,
\n+
400 /* start = */ 0,
\n+
401 /* end = */ stencilTable->GetNumStencils());
\n+
402 }
\n+
403
\n+
433 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
434 bool EvalStencils(
\n+
435 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
436 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
437 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
438 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
439 STENCIL_TABLE const *stencilTable) const {
\n+
440 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n+
441 dstBuffer->BindVBO(), dstDesc,
\n+
442 duBuffer->BindVBO(), duDesc,
\n+
443 dvBuffer->BindVBO(), dvDesc,
\n+
444 stencilTable->GetSizesBuffer(),
\n+
445 stencilTable->GetOffsetsBuffer(),
\n+
446 stencilTable->GetIndicesBuffer(),
\n+
447 stencilTable->GetWeightsBuffer(),
\n+
448 stencilTable->GetDuWeightsBuffer(),
\n+
449 stencilTable->GetDvWeightsBuffer(),
\n+
450 /* start = */ 0,
\n+
451 /* end = */ stencilTable->GetNumStencils());
\n+
452 }
\n+
453
\n+
501 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
502 bool EvalStencils(
\n+
503 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
504 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
505 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
506 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
507 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
508 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
509 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
510 STENCIL_TABLE const *stencilTable) const {
\n+
511 return EvalStencils(srcBuffer->BindVBO(), srcDesc,
\n+
512 dstBuffer->BindVBO(), dstDesc,
\n+
513 duBuffer->BindVBO(), duDesc,
\n+
514 dvBuffer->BindVBO(), dvDesc,
\n+
515 duuBuffer->BindVBO(), duuDesc,
\n+
516 duvBuffer->BindVBO(), duvDesc,
\n+
517 dvvBuffer->BindVBO(), dvvDesc,
\n+
518 stencilTable->GetSizesBuffer(),
\n+
519 stencilTable->GetOffsetsBuffer(),
\n+
520 stencilTable->GetIndicesBuffer(),
\n+
521 stencilTable->GetWeightsBuffer(),
\n+
522 stencilTable->GetDuWeightsBuffer(),
\n+
523 stencilTable->GetDvWeightsBuffer(),
\n+
524 stencilTable->GetDuuWeightsBuffer(),
\n+
525 stencilTable->GetDuvWeightsBuffer(),
\n+
526 stencilTable->GetDvvWeightsBuffer(),
\n+
527 /* start = */ 0,
\n+
528 /* end = */ stencilTable->GetNumStencils());
\n+
529 }
\n+
530
\n+
566 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n+
567 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n+
568 GLuint duBuffer, BufferDescriptor const &duDesc,
\n+
569 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n+
570 GLuint sizesBuffer,
\n+
571 GLuint offsetsBuffer,
\n+
572 GLuint indicesBuffer,
\n+
573 GLuint weightsBuffer,
\n+
574 GLuint duWeightsBuffer,
\n+
575 GLuint dvWeightsBuffer,
\n+
576 int start,
\n+
577 int end) const;
\n+
578
\n+
632 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n+
633 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n+
634 GLuint duBuffer, BufferDescriptor const &duDesc,
\n+
635 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n+
636 GLuint duuBuffer, BufferDescriptor const &duuDesc,
\n+
637 GLuint duvBuffer, BufferDescriptor const &duvDesc,
\n+
638 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
639 GLuint sizesBuffer,
\n+
640 GLuint offsetsBuffer,
\n+
641 GLuint indicesBuffer,
\n+
642 GLuint weightsBuffer,
\n+
643 GLuint duWeightsBuffer,
\n+
644 GLuint dvWeightsBuffer,
\n+
645 GLuint duuWeightsBuffer,
\n+
646 GLuint duvWeightsBuffer,
\n+
647 GLuint dvvWeightsBuffer,
\n+
648 int start,
\n+
649 int end) const;
\n+
650
\n+
656
\n+
689 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
690 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
691 static bool EvalPatches(
\n+
692 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
693 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
694 int numPatchCoords,
\n+
695 PATCHCOORD_BUFFER *patchCoords,
\n+
696 PATCH_TABLE *patchTable,
\n+
697 GLComputeEvaluator const *instance,
\n+
698 void * deviceContext = NULL) {
\n+
699
\n+
700 if (instance) {
\n+
701 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
702 dstBuffer, dstDesc,
\n+
703 numPatchCoords, patchCoords,
\n+
704 patchTable);
\n+
705 } else {
\n+
706 // Create an instance on demand (slow)
\n+
707 (void)deviceContext; // unused
\n+
708 instance = Create(srcDesc, dstDesc,
\n+
709 BufferDescriptor(),
\n+
710 BufferDescriptor());
\n+
711 if (instance) {
\n+
712 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
713 dstBuffer, dstDesc,
\n+
714 numPatchCoords, patchCoords,
\n+
715 patchTable);
\n+
716 delete instance;
\n+
717 return r;
\n+
718 }
\n+
719 return false;
\n+
720 }
\n+
721 }
\n+
722
\n+
767 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
768 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
769 static bool EvalPatches(
\n+
770 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
771 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
772 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
773 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
774 int numPatchCoords,
\n+
775 PATCHCOORD_BUFFER *patchCoords,
\n+
776 PATCH_TABLE *patchTable,
\n+
777 GLComputeEvaluator const *instance,
\n+
778 void * deviceContext = NULL) {
\n+
779
\n+
780 if (instance) {
\n+
781 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
782 dstBuffer, dstDesc,
\n+
783 duBuffer, duDesc,
\n+
784 dvBuffer, dvDesc,
\n+
785 numPatchCoords, patchCoords,
\n+
786 patchTable);
\n+
787 } else {
\n+
788 // Create an instance on demand (slow)
\n+
789 (void)deviceContext; // unused
\n+
790 instance = Create(srcDesc, dstDesc,
\n+
791 duDesc, dvDesc);
\n+
792 if (instance) {
\n+
793 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
794 dstBuffer, dstDesc,
\n+
795 duBuffer, duDesc,
\n+
796 dvBuffer, dvDesc,
\n+
797 numPatchCoords, patchCoords,
\n+
798 patchTable);
\n+
799 delete instance;
\n+
800 return r;
\n+
801 }
\n+
802 return false;
\n+
803 }
\n+
804 }
\n+
805
\n+
868 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
869 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
870 static bool EvalPatches(
\n+
871 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
872 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
873 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
874 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
875 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
876 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
877 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
878 int numPatchCoords,
\n+
879 PATCHCOORD_BUFFER *patchCoords,
\n+
880 PATCH_TABLE *patchTable,
\n+
881 GLComputeEvaluator const *instance,
\n+
882 void * deviceContext = NULL) {
\n+
883
\n+
884 if (instance) {
\n+
885 return instance->EvalPatches(srcBuffer, srcDesc,
\n+
886 dstBuffer, dstDesc,
\n+
887 duBuffer, duDesc,
\n+
888 dvBuffer, dvDesc,
\n+
889 duuBuffer, duuDesc,
\n+
890 duvBuffer, duvDesc,
\n+
891 dvvBuffer, dvvDesc,
\n+
892 numPatchCoords, patchCoords,
\n+
893 patchTable);
\n+
894 } else {
\n+
895 // Create an instance on demand (slow)
\n+
896 (void)deviceContext; // unused
\n+
897 instance = Create(srcDesc, dstDesc,
\n+
898 duDesc, dvDesc,
\n+
899 duuDesc, duvDesc, dvvDesc);
\n+
900 if (instance) {
\n+
901 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n+
902 dstBuffer, dstDesc,
\n+
903 duBuffer, duDesc,
\n+
904 dvBuffer, dvDesc,
\n+
905 duuBuffer, duuDesc,
\n+
906 duvBuffer, duvDesc,
\n+
907 dvvBuffer, dvvDesc,
\n+
908 numPatchCoords, patchCoords,
\n+
909 patchTable);
\n+
910 delete instance;
\n+
911 return r;
\n+
912 }
\n+
913 return false;
\n+
914 }
\n+
915 }
\n+
916
\n
941 template <typename SRC_BUFFER, typename DST_BUFFER,
\n
942 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
943 static bool EvalPatches(
\n+
943 bool EvalPatches(
\n
944 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n
945 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
946 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
947 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
948 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
949 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
950 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
951 int numPatchCoords,
\n-
952 PATCHCOORD_BUFFER *patchCoords,
\n-
953 PATCH_TABLE *patchTable,
\n-
954 GLXFBEvaluator const *instance,
\n-
955 void * deviceContext = NULL) {
\n-
956
\n-
957 if (instance) {
\n-
958 return instance->EvalPatches(srcBuffer, srcDesc,
\n-
959 dstBuffer, dstDesc,
\n-
960 duBuffer, duDesc,
\n-
961 dvBuffer, dvDesc,
\n-
962 duuBuffer, duuDesc,
\n-
963 duvBuffer, duvDesc,
\n-
964 dvvBuffer, dvvDesc,
\n-
965 numPatchCoords, patchCoords,
\n-
966 patchTable);
\n-
967 } else {
\n-
968 // Create an instance on demand (slow)
\n-
969 (void)deviceContext; // unused
\n-
970 instance = Create(srcDesc, dstDesc,
\n-
971 duDesc, dvDesc,
\n-
972 duuDesc, duvDesc, dvvDesc);
\n-
973 if (instance) {
\n-
974 bool r = instance->EvalPatches(srcBuffer, srcDesc,
\n-
975 dstBuffer, dstDesc,
\n-
976 duBuffer, duDesc,
\n-
977 dvBuffer, dvDesc,
\n-
978 duuBuffer, duuDesc,
\n-
979 duvBuffer, duvDesc,
\n-
980 dvvBuffer, dvvDesc,
\n-
981 numPatchCoords, patchCoords,
\n-
982 patchTable);
\n-
983 delete instance;
\n-
984 return r;
\n-
985 }
\n-
986 return false;
\n-
987 }
\n-
988 }
\n-
989
\n-
1014 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1015 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1016 bool EvalPatches(
\n-
1017 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1018 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1019 int numPatchCoords,
\n-
1020 PATCHCOORD_BUFFER *patchCoords,
\n-
1021 PATCH_TABLE *patchTable) const {
\n-
1022
\n-
1023 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1024 dstBuffer->BindVBO(), dstDesc,
\n-
1025 0, BufferDescriptor(),
\n-
1026 0, BufferDescriptor(),
\n-
1027 numPatchCoords,
\n-
1028 patchCoords->BindVBO(),
\n-
1029 patchTable->GetPatchArrays(),
\n-
1030 patchTable->GetPatchIndexTextureBuffer(),
\n-
1031 patchTable->GetPatchParamTextureBuffer());
\n-
1032 }
\n-
1033
\n-
1068 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1069 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1070 bool EvalPatches(
\n-
1071 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1072 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1073 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1074 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1075 int numPatchCoords,
\n-
1076 PATCHCOORD_BUFFER *patchCoords,
\n-
1077 PATCH_TABLE *patchTable) const {
\n-
1078
\n-
1079 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1080 dstBuffer->BindVBO(), dstDesc,
\n-
1081 duBuffer->BindVBO(), duDesc,
\n-
1082 dvBuffer->BindVBO(), dvDesc,
\n-
1083 numPatchCoords,
\n-
1084 patchCoords->BindVBO(),
\n-
1085 patchTable->GetPatchArrays(),
\n-
1086 patchTable->GetPatchIndexTextureBuffer(),
\n-
1087 patchTable->GetPatchParamTextureBuffer());
\n-
1088 }
\n-
1089
\n-
1142 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1143 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1144 bool EvalPatches(
\n-
1145 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1146 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1147 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1148 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1149 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1150 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1151 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1152 int numPatchCoords,
\n-
1153 PATCHCOORD_BUFFER *patchCoords,
\n-
1154 PATCH_TABLE *patchTable) const {
\n-
1155
\n-
1156 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1157 dstBuffer->BindVBO(), dstDesc,
\n-
1158 duBuffer->BindVBO(), duDesc,
\n-
1159 dvBuffer->BindVBO(), dvDesc,
\n-
1160 duuBuffer->BindVBO(), duuDesc,
\n-
1161 duvBuffer->BindVBO(), duvDesc,
\n-
1162 dvvBuffer->BindVBO(), dvvDesc,
\n-
1163 numPatchCoords,
\n-
1164 patchCoords->BindVBO(),
\n-
1165 patchTable->GetPatchArrays(),
\n-
1166 patchTable->GetPatchIndexTextureBuffer(),
\n-
1167 patchTable->GetPatchParamTextureBuffer());
\n-
1168 }
\n-
1169
\n-
1170 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n-
1171 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n-
1172 GLuint duBuffer, BufferDescriptor const &duDesc,
\n-
1173 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n-
1174 int numPatchCoords,
\n-
1175 GLuint patchCoordsBuffer,
\n-
1176 const PatchArrayVector &patchArrays,
\n-
1177 GLuint patchIndexBuffer,
\n-
1178 GLuint patchParamsBuffer) const;
\n-
1179
\n-
1180 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n-
1181 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n-
1182 GLuint duBuffer, BufferDescriptor const &duDesc,
\n-
1183 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n-
1184 GLuint duuBuffer, BufferDescriptor const &duuDesc,
\n-
1185 GLuint duvBuffer, BufferDescriptor const &duvDesc,
\n-
1186 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1187 int numPatchCoords,
\n-
1188 GLuint patchCoordsBuffer,
\n-
1189 const PatchArrayVector &patchArrays,
\n-
1190 GLuint patchIndexBuffer,
\n-
1191 GLuint patchParamsBuffer) const;
\n-
1192
\n-
1225 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1226 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1227 static bool EvalPatchesVarying(
\n-
1228 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1229 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1230 int numPatchCoords,
\n-
1231 PATCHCOORD_BUFFER *patchCoords,
\n-
1232 PATCH_TABLE *patchTable,
\n-
1233 GLXFBEvaluator const *instance,
\n-
1234 void * deviceContext = NULL) {
\n-
1235
\n-
1236 if (instance) {
\n-
1237 return instance->EvalPatchesVarying(
\n-
1238 srcBuffer, srcDesc,
\n-
1239 dstBuffer, dstDesc,
\n-
1240 numPatchCoords, patchCoords,
\n-
1241 patchTable);
\n-
1242 } else {
\n-
1243 // Create an instance on demand (slow)
\n-
1244 (void)deviceContext; // unused
\n-
1245 instance = Create(srcDesc, dstDesc,
\n-
1246 BufferDescriptor(),
\n-
1247 BufferDescriptor());
\n-
1248 if (instance) {
\n-
1249 bool r = instance->EvalPatchesVarying(
\n-
1250 srcBuffer, srcDesc,
\n-
1251 dstBuffer, dstDesc,
\n-
1252 numPatchCoords, patchCoords,
\n-
1253 patchTable);
\n-
1254 delete instance;
\n-
1255 return r;
\n-
1256 }
\n-
1257 return false;
\n-
1258 }
\n-
1259 }
\n-
1260
\n-
1285 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1286 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1287 bool EvalPatchesVarying(
\n-
1288 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1289 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1290 int numPatchCoords,
\n-
1291 PATCHCOORD_BUFFER *patchCoords,
\n-
1292 PATCH_TABLE *patchTable) const {
\n-
1293
\n-
1294 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1295 dstBuffer->BindVBO(), dstDesc,
\n-
1296 0, BufferDescriptor(),
\n-
1297 0, BufferDescriptor(),
\n-
1298 numPatchCoords,
\n-
1299 patchCoords->BindVBO(),
\n-
1300 patchTable->GetVaryingPatchArrays(),
\n-
1301 patchTable->GetVaryingPatchIndexTextureBuffer(),
\n-
1302 patchTable->GetPatchParamTextureBuffer());
\n-
1303 }
\n-
1304
\n-
1349 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1350 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1351 static bool EvalPatchesVarying(
\n-
1352 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1353 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1354 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1355 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1356 int numPatchCoords,
\n-
1357 PATCHCOORD_BUFFER *patchCoords,
\n-
1358 PATCH_TABLE *patchTable,
\n-
1359 GLXFBEvaluator const *instance,
\n-
1360 void * deviceContext = NULL) {
\n-
1361
\n-
1362 if (instance) {
\n-
1363 return instance->EvalPatchesVarying(
\n-
1364 srcBuffer, srcDesc,
\n-
1365 dstBuffer, dstDesc,
\n-
1366 duBuffer, duDesc,
\n-
1367 dvBuffer, dvDesc,
\n-
1368 numPatchCoords, patchCoords,
\n-
1369 patchTable);
\n-
1370 } else {
\n-
1371 // Create an instance on demand (slow)
\n-
1372 (void)deviceContext; // unused
\n-
1373 instance = Create(srcDesc, dstDesc,
\n-
1374 duDesc, dvDesc);
\n-
1375 if (instance) {
\n-
1376 bool r = instance->EvalPatchesVarying(
\n-
1377 srcBuffer, srcDesc,
\n-
1378 dstBuffer, dstDesc,
\n-
1379 duBuffer, duDesc,
\n-
1380 dvBuffer, dvDesc,
\n-
1381 numPatchCoords, patchCoords,
\n-
1382 patchTable);
\n-
1383 delete instance;
\n-
1384 return r;
\n-
1385 }
\n-
1386 return false;
\n-
1387 }
\n-
1388 }
\n-
1389
\n-
1426 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1427 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1428 bool EvalPatchesVarying(
\n-
1429 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1430 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1431 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1432 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1433 int numPatchCoords,
\n-
1434 PATCHCOORD_BUFFER *patchCoords,
\n-
1435 PATCH_TABLE *patchTable) const {
\n-
1436
\n-
1437 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1438 dstBuffer->BindVBO(), dstDesc,
\n-
1439 duBuffer->BindVBO(), duDesc,
\n-
1440 dvBuffer->BindVBO(), dvDesc,
\n-
1441 numPatchCoords,
\n-
1442 patchCoords->BindVBO(),
\n-
1443 patchTable->GetVaryingPatchArrays(),
\n-
1444 patchTable->GetVaryingPatchIndexTextureBuffer(),
\n-
1445 patchTable->GetPatchParamTextureBuffer());
\n-
1446 }
\n-
1447
\n-
1510 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1511 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1512 static bool EvalPatchesVarying(
\n-
1513 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1514 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1515 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1516 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1517 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1518 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1519 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1520 int numPatchCoords,
\n-
1521 PATCHCOORD_BUFFER *patchCoords,
\n-
1522 PATCH_TABLE *patchTable,
\n-
1523 GLXFBEvaluator const *instance,
\n-
1524 void * deviceContext = NULL) {
\n-
1525
\n-
1526 if (instance) {
\n-
1527 return instance->EvalPatchesVarying(
\n-
1528 srcBuffer, srcDesc,
\n-
1529 dstBuffer, dstDesc,
\n-
1530 duBuffer, duDesc,
\n-
1531 dvBuffer, dvDesc,
\n-
1532 duuBuffer, duuDesc,
\n-
1533 duvBuffer, duvDesc,
\n-
1534 dvvBuffer, dvvDesc,
\n-
1535 numPatchCoords, patchCoords,
\n-
1536 patchTable);
\n-
1537 } else {
\n-
1538 // Create an instance on demand (slow)
\n-
1539 (void)deviceContext; // unused
\n-
1540 instance = Create(srcDesc, dstDesc,
\n-
1541 duDesc, dvDesc,
\n-
1542 duuDesc, duvDesc, dvvDesc);
\n-
1543 if (instance) {
\n-
1544 bool r = instance->EvalPatchesVarying(
\n-
1545 srcBuffer, srcDesc,
\n-
1546 dstBuffer, dstDesc,
\n-
1547 duBuffer, duDesc,
\n-
1548 dvBuffer, dvDesc,
\n-
1549 duuBuffer, duuDesc,
\n-
1550 duvBuffer, duvDesc,
\n-
1551 dvvBuffer, dvvDesc,
\n-
1552 numPatchCoords, patchCoords,
\n-
1553 patchTable);
\n-
1554 delete instance;
\n-
1555 return r;
\n-
1556 }
\n-
1557 return false;
\n-
1558 }
\n-
1559 }
\n-
1560
\n-
1615 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1616 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1617 bool EvalPatchesVarying(
\n-
1618 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1619 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1620 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1621 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1622 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1623 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1624 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1625 int numPatchCoords,
\n-
1626 PATCHCOORD_BUFFER *patchCoords,
\n-
1627 PATCH_TABLE *patchTable) const {
\n-
1628
\n-
1629 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1630 dstBuffer->BindVBO(), dstDesc,
\n-
1631 duBuffer->BindVBO(), duDesc,
\n-
1632 dvBuffer->BindVBO(), dvDesc,
\n-
1633 duuBuffer->BindVBO(), duuDesc,
\n-
1634 duvBuffer->BindVBO(), duvDesc,
\n-
1635 dvvBuffer->BindVBO(), dvvDesc,
\n-
1636 numPatchCoords,
\n-
1637 patchCoords->BindVBO(),
\n-
1638 patchTable->GetVaryingPatchArrays(),
\n-
1639 patchTable->GetVaryingPatchIndexTextureBuffer(),
\n-
1640 patchTable->GetPatchParamTextureBuffer());
\n-
1641 }
\n-
1642
\n-
1677 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1678 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1679 static bool EvalPatchesFaceVarying(
\n-
1680 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1681 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1682 int numPatchCoords,
\n-
1683 PATCHCOORD_BUFFER *patchCoords,
\n-
1684 PATCH_TABLE *patchTable,
\n-
1685 int fvarChannel,
\n-
1686 GLXFBEvaluator const *instance,
\n-
1687 void * deviceContext = NULL) {
\n-
1688
\n-
1689 if (instance) {
\n-
1690 return instance->EvalPatchesFaceVarying(
\n-
1691 srcBuffer, srcDesc,
\n-
1692 dstBuffer, dstDesc,
\n-
1693 numPatchCoords, patchCoords,
\n-
1694 patchTable, fvarChannel);
\n-
1695 } else {
\n-
1696 // Create an instance on demand (slow)
\n-
1697 (void)deviceContext; // unused
\n-
1698 instance = Create(srcDesc, dstDesc,
\n-
1699 BufferDescriptor(),
\n-
1700 BufferDescriptor());
\n-
1701 if (instance) {
\n-
1702 bool r = instance->EvalPatchesFaceVarying(
\n-
1703 srcBuffer, srcDesc,
\n-
1704 dstBuffer, dstDesc,
\n-
1705 numPatchCoords, patchCoords,
\n-
1706 patchTable, fvarChannel);
\n-
1707 delete instance;
\n-
1708 return r;
\n-
1709 }
\n-
1710 return false;
\n-
1711 }
\n-
1712 }
\n-
1713
\n-
1740 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1741 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1742 bool EvalPatchesFaceVarying(
\n-
1743 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1744 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1745 int numPatchCoords,
\n-
1746 PATCHCOORD_BUFFER *patchCoords,
\n-
1747 PATCH_TABLE *patchTable,
\n-
1748 int fvarChannel = 0) const {
\n-
1749
\n-
1750 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1751 dstBuffer->BindVBO(), dstDesc,
\n-
1752 0, BufferDescriptor(),
\n-
1753 0, BufferDescriptor(),
\n-
1754 numPatchCoords,
\n-
1755 patchCoords->BindVBO(),
\n-
1756 patchTable->GetFVarPatchArrays(fvarChannel),
\n-
1757 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),
\n-
1758 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));
\n-
1759 }
\n-
1760
\n-
1807 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1808 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1809 static bool EvalPatchesFaceVarying(
\n-
1810 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1811 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1812 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1813 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1814 int numPatchCoords,
\n-
1815 PATCHCOORD_BUFFER *patchCoords,
\n-
1816 PATCH_TABLE *patchTable,
\n-
1817 int fvarChannel,
\n-
1818 GLXFBEvaluator const *instance,
\n-
1819 void * deviceContext = NULL) {
\n-
1820
\n-
1821 if (instance) {
\n-
1822 return instance->EvalPatchesFaceVarying(
\n-
1823 srcBuffer, srcDesc,
\n-
1824 dstBuffer, dstDesc,
\n-
1825 duBuffer, duDesc,
\n-
1826 dvBuffer, dvDesc,
\n-
1827 numPatchCoords, patchCoords,
\n-
1828 patchTable, fvarChannel);
\n-
1829 } else {
\n-
1830 // Create an instance on demand (slow)
\n-
1831 (void)deviceContext; // unused
\n-
1832 instance = Create(srcDesc, dstDesc,
\n-
1833 duDesc, dvDesc);
\n-
1834 if (instance) {
\n-
1835 bool r = instance->EvalPatchesFaceVarying(
\n-
1836 srcBuffer, srcDesc,
\n-
1837 dstBuffer, dstDesc,
\n-
1838 duBuffer, duDesc,
\n-
1839 dvBuffer, dvDesc,
\n-
1840 numPatchCoords, patchCoords,
\n-
1841 patchTable, fvarChannel);
\n-
1842 delete instance;
\n-
1843 return r;
\n-
1844 }
\n-
1845 return false;
\n-
1846 }
\n-
1847 }
\n-
1848
\n-
1887 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1888 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1889 bool EvalPatchesFaceVarying(
\n-
1890 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1891 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1892 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1893 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1894 int numPatchCoords,
\n-
1895 PATCHCOORD_BUFFER *patchCoords,
\n-
1896 PATCH_TABLE *patchTable,
\n-
1897 int fvarChannel = 0) const {
\n-
1898
\n-
1899 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
1900 dstBuffer->BindVBO(), dstDesc,
\n-
1901 duBuffer->BindVBO(), duDesc,
\n-
1902 dvBuffer->BindVBO(), dvDesc,
\n-
1903 numPatchCoords,
\n-
1904 patchCoords->BindVBO(),
\n-
1905 patchTable->GetFVarPatchArrays(fvarChannel),
\n-
1906 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),
\n-
1907 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));
\n-
1908 }
\n-
1909
\n-
1974 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1975 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1976 static bool EvalPatchesFaceVarying(
\n-
1977 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1978 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1979 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1980 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1981 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1982 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1983 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1984 int numPatchCoords,
\n-
1985 PATCHCOORD_BUFFER *patchCoords,
\n-
1986 PATCH_TABLE *patchTable,
\n-
1987 int fvarChannel,
\n-
1988 GLXFBEvaluator const *instance,
\n-
1989 void * deviceContext = NULL) {
\n-
1990
\n-
1991 if (instance) {
\n-
1992 return instance->EvalPatchesFaceVarying(
\n-
1993 srcBuffer, srcDesc,
\n-
1994 dstBuffer, dstDesc,
\n-
1995 duBuffer, duDesc,
\n-
1996 dvBuffer, dvDesc,
\n-
1997 duuBuffer, duuDesc,
\n-
1998 duvBuffer, duvDesc,
\n-
1999 dvvBuffer, dvvDesc,
\n-
2000 numPatchCoords, patchCoords,
\n-
2001 patchTable, fvarChannel);
\n-
2002 } else {
\n-
2003 // Create an instance on demand (slow)
\n-
2004 (void)deviceContext; // unused
\n-
2005 instance = Create(srcDesc, dstDesc,
\n-
2006 duDesc, dvDesc,
\n-
2007 duuDesc, duvDesc, dvvDesc);
\n-
2008 if (instance) {
\n-
2009 bool r = instance->EvalPatchesFaceVarying(
\n-
2010 srcBuffer, srcDesc,
\n-
2011 dstBuffer, dstDesc,
\n-
2012 duBuffer, duDesc,
\n-
2013 dvBuffer, dvDesc,
\n-
2014 duuBuffer, duuDesc,
\n-
2015 duvBuffer, duvDesc,
\n-
2016 dvvBuffer, dvvDesc,
\n-
2017 numPatchCoords, patchCoords,
\n-
2018 patchTable, fvarChannel);
\n-
2019 delete instance;
\n-
2020 return r;
\n-
2021 }
\n-
2022 return false;
\n-
2023 }
\n-
2024 }
\n-
2025
\n-
2082 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
2083 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
2084 bool EvalPatchesFaceVarying(
\n-
2085 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
2086 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
2087 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
2088 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
2089 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
2090 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
2091 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
2092 int numPatchCoords,
\n-
2093 PATCHCOORD_BUFFER *patchCoords,
\n-
2094 PATCH_TABLE *patchTable,
\n-
2095 int fvarChannel = 0) const {
\n-
2096
\n-
2097 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n-
2098 dstBuffer->BindVBO(), dstDesc,
\n-
2099 duBuffer->BindVBO(), duDesc,
\n-
2100 dvBuffer->BindVBO(), dvDesc,
\n-
2101 duuBuffer->BindVBO(), duuDesc,
\n-
2102 duvBuffer->BindVBO(), duvDesc,
\n-
2103 dvvBuffer->BindVBO(), dvvDesc,
\n-
2104 numPatchCoords,
\n-
2105 patchCoords->BindVBO(),
\n-
2106 patchTable->GetFVarPatchArrays(fvarChannel),
\n-
2107 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),
\n-
2108 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));
\n-
2109 }
\n-
2110
\n-
2116
\n-
2119 bool Compile(BufferDescriptor const &srcDesc,
\n-
2120 BufferDescriptor const &dstDesc,
\n-
2121 BufferDescriptor const &duDesc = BufferDescriptor(),
\n-
2122 BufferDescriptor const &dvDesc = BufferDescriptor(),
\n-
2123 BufferDescriptor const &duuDesc = BufferDescriptor(),
\n-
2124 BufferDescriptor const &duvDesc = BufferDescriptor(),
\n-
2125 BufferDescriptor const &dvvDesc = BufferDescriptor());
\n-
2126
\n-
2128 static void Synchronize(void *kernel);
\n-
2129
\n-
2130private:
\n-
2131 GLuint _srcBufferTexture;
\n-
2132 GLuint _patchArraysUBO;
\n-
2133 bool _interleavedDerivativeBuffers;
\n-
2134
\n-
2135 struct _StencilKernel {
\n-
2136 _StencilKernel();
\n-
2137 ~_StencilKernel();
\n-
2138 bool Compile(BufferDescriptor const &srcDesc,
\n-
2139 BufferDescriptor const &dstDesc,
\n-
2140 BufferDescriptor const &duDesc,
\n-
2141 BufferDescriptor const &dvDesc,
\n-
2142 BufferDescriptor const &duuDesc,
\n-
2143 BufferDescriptor const &duvDesc,
\n-
2144 BufferDescriptor const &dvvDesc,
\n-
2145 bool interleavedDerivativeBuffers);
\n-
2146 GLuint program;
\n-
2147 GLint uniformSrcBufferTexture;
\n-
2148 GLint uniformSrcOffset; // src buffer offset (in elements)
\n-
2149
\n-
2150 GLint uniformSizesTexture;
\n-
2151 GLint uniformOffsetsTexture;
\n-
2152 GLint uniformIndicesTexture;
\n-
2153 GLint uniformWeightsTexture;
\n-
2154 GLint uniformDuWeightsTexture;
\n-
2155 GLint uniformDvWeightsTexture;
\n-
2156 GLint uniformDuuWeightsTexture;
\n-
2157 GLint uniformDuvWeightsTexture;
\n-
2158 GLint uniformDvvWeightsTexture;
\n-
2159 GLint uniformStart; // range
\n-
2160 GLint uniformEnd;
\n-
2161 } _stencilKernel;
\n-
2162
\n-
2163 struct _PatchKernel {
\n-
2164 _PatchKernel();
\n-
2165 ~_PatchKernel();
\n-
2166 bool Compile(BufferDescriptor const &srcDesc,
\n-
2167 BufferDescriptor const &dstDesc,
\n-
2168 BufferDescriptor const &duDesc,
\n-
2169 BufferDescriptor const &dvDesc,
\n-
2170 BufferDescriptor const &duuDesc,
\n-
2171 BufferDescriptor const &duvDesc,
\n-
2172 BufferDescriptor const &dvvDesc,
\n-
2173 bool interleavedDerivativeBuffers);
\n-
2174 GLuint program;
\n-
2175 GLint uniformSrcBufferTexture;
\n-
2176 GLint uniformSrcOffset; // src buffer offset (in elements)
\n-
2177
\n-
2178 GLint uniformPatchArraysUBOBinding;
\n-
2179 GLint uniformPatchParamTexture;
\n-
2180 GLint uniformPatchIndexTexture;
\n-
2181 } _patchKernel;
\n-
2182
\n-
2183};
\n-
2184
\n-
2185} // end namespace Osd
\n-
2186
\n-
2187} // end namespace OPENSUBDIV_VERSION
\n-
2188using namespace OPENSUBDIV_VERSION;
\n-
2189
\n-
2190} // end namespace OpenSubdiv
\n-
2191
\n-
2192
\n-
2193#endif // OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H
\n+
946 int numPatchCoords,
\n+
947 PATCHCOORD_BUFFER *patchCoords,
\n+
948 PATCH_TABLE *patchTable) const {
\n+
949
\n+
950 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
951 dstBuffer->BindVBO(), dstDesc,
\n+
952 0, BufferDescriptor(),
\n+
953 0, BufferDescriptor(),
\n+
954 numPatchCoords,
\n+
955 patchCoords->BindVBO(),
\n+
956 patchTable->GetPatchArrays(),
\n+
957 patchTable->GetPatchIndexBuffer(),
\n+
958 patchTable->GetPatchParamBuffer());
\n+
959 }
\n+
960
\n+
995 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
996 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
997 bool EvalPatches(
\n+
998 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
999 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1000 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1001 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1002 int numPatchCoords,
\n+
1003 PATCHCOORD_BUFFER *patchCoords,
\n+
1004 PATCH_TABLE *patchTable) const {
\n+
1005
\n+
1006 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1007 dstBuffer->BindVBO(), dstDesc,
\n+
1008 duBuffer->BindVBO(), duDesc,
\n+
1009 dvBuffer->BindVBO(), dvDesc,
\n+
1010 numPatchCoords,
\n+
1011 patchCoords->BindVBO(),
\n+
1012 patchTable->GetPatchArrays(),
\n+
1013 patchTable->GetPatchIndexBuffer(),
\n+
1014 patchTable->GetPatchParamBuffer());
\n+
1015 }
\n+
1016
\n+
1069 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1070 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1071 bool EvalPatches(
\n+
1072 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1073 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1074 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1075 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1076 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1077 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1078 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1079 int numPatchCoords,
\n+
1080 PATCHCOORD_BUFFER *patchCoords,
\n+
1081 PATCH_TABLE *patchTable) const {
\n+
1082
\n+
1083 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1084 dstBuffer->BindVBO(), dstDesc,
\n+
1085 duBuffer->BindVBO(), duDesc,
\n+
1086 dvBuffer->BindVBO(), dvDesc,
\n+
1087 duuBuffer->BindVBO(), duuDesc,
\n+
1088 duvBuffer->BindVBO(), duvDesc,
\n+
1089 dvvBuffer->BindVBO(), dvvDesc,
\n+
1090 numPatchCoords,
\n+
1091 patchCoords->BindVBO(),
\n+
1092 patchTable->GetPatchArrays(),
\n+
1093 patchTable->GetPatchIndexBuffer(),
\n+
1094 patchTable->GetPatchParamBuffer());
\n+
1095 }
\n+
1096
\n+
1097 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n+
1098 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n+
1099 GLuint duBuffer, BufferDescriptor const &duDesc,
\n+
1100 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n+
1101 int numPatchCoords,
\n+
1102 GLuint patchCoordsBuffer,
\n+
1103 const PatchArrayVector &patchArrays,
\n+
1104 GLuint patchIndexBuffer,
\n+
1105 GLuint patchParamsBuffer) const;
\n+
1106
\n+
1107 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,
\n+
1108 GLuint dstBuffer, BufferDescriptor const &dstDesc,
\n+
1109 GLuint duBuffer, BufferDescriptor const &duDesc,
\n+
1110 GLuint dvBuffer, BufferDescriptor const &dvDesc,
\n+
1111 GLuint duuBuffer, BufferDescriptor const &duuDesc,
\n+
1112 GLuint duvBuffer, BufferDescriptor const &duvDesc,
\n+
1113 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1114 int numPatchCoords,
\n+
1115 GLuint patchCoordsBuffer,
\n+
1116 const PatchArrayVector &patchArrays,
\n+
1117 GLuint patchIndexBuffer,
\n+
1118 GLuint patchParamsBuffer) const;
\n+
1119
\n+
1152 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1153 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1154 static bool EvalPatchesVarying(
\n+
1155 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1156 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1157 int numPatchCoords,
\n+
1158 PATCHCOORD_BUFFER *patchCoords,
\n+
1159 PATCH_TABLE *patchTable,
\n+
1160 GLComputeEvaluator const *instance,
\n+
1161 void * deviceContext = NULL) {
\n+
1162
\n+
1163 if (instance) {
\n+
1164 return instance->EvalPatchesVarying(
\n+
1165 srcBuffer, srcDesc,
\n+
1166 dstBuffer, dstDesc,
\n+
1167 numPatchCoords, patchCoords,
\n+
1168 patchTable);
\n+
1169 } else {
\n+
1170 // Create an instance on demand (slow)
\n+
1171 (void)deviceContext; // unused
\n+
1172 instance = Create(srcDesc, dstDesc,
\n+
1173 BufferDescriptor(),
\n+
1174 BufferDescriptor());
\n+
1175 if (instance) {
\n+
1176 bool r = instance->EvalPatchesVarying(
\n+
1177 srcBuffer, srcDesc,
\n+
1178 dstBuffer, dstDesc,
\n+
1179 numPatchCoords, patchCoords,
\n+
1180 patchTable);
\n+
1181 delete instance;
\n+
1182 return r;
\n+
1183 }
\n+
1184 return false;
\n+
1185 }
\n+
1186 }
\n+
1187
\n+
1212 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1213 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1214 bool EvalPatchesVarying(
\n+
1215 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1216 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1217 int numPatchCoords,
\n+
1218 PATCHCOORD_BUFFER *patchCoords,
\n+
1219 PATCH_TABLE *patchTable) const {
\n+
1220
\n+
1221 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1222 dstBuffer->BindVBO(), dstDesc,
\n+
1223 0, BufferDescriptor(),
\n+
1224 0, BufferDescriptor(),
\n+
1225 numPatchCoords,
\n+
1226 patchCoords->BindVBO(),
\n+
1227 patchTable->GetVaryingPatchArrays(),
\n+
1228 patchTable->GetVaryingPatchIndexBuffer(),
\n+
1229 patchTable->GetPatchParamBuffer());
\n+
1230 }
\n+
1231
\n+
1276 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1277 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1278 static bool EvalPatchesVarying(
\n+
1279 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1280 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1281 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1282 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1283 int numPatchCoords,
\n+
1284 PATCHCOORD_BUFFER *patchCoords,
\n+
1285 PATCH_TABLE *patchTable,
\n+
1286 GLComputeEvaluator const *instance,
\n+
1287 void * deviceContext = NULL) {
\n+
1288
\n+
1289 if (instance) {
\n+
1290 return instance->EvalPatchesVarying(
\n+
1291 srcBuffer, srcDesc,
\n+
1292 dstBuffer, dstDesc,
\n+
1293 duBuffer, duDesc,
\n+
1294 dvBuffer, dvDesc,
\n+
1295 numPatchCoords, patchCoords,
\n+
1296 patchTable);
\n+
1297 } else {
\n+
1298 // Create an instance on demand (slow)
\n+
1299 (void)deviceContext; // unused
\n+
1300 instance = Create(srcDesc, dstDesc,
\n+
1301 duDesc, dvDesc);
\n+
1302 if (instance) {
\n+
1303 bool r = instance->EvalPatchesVarying(
\n+
1304 srcBuffer, srcDesc,
\n+
1305 dstBuffer, dstDesc,
\n+
1306 duBuffer, duDesc,
\n+
1307 dvBuffer, dvDesc,
\n+
1308 numPatchCoords, patchCoords,
\n+
1309 patchTable);
\n+
1310 delete instance;
\n+
1311 return r;
\n+
1312 }
\n+
1313 return false;
\n+
1314 }
\n+
1315 }
\n+
1316
\n+
1353 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1354 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1355 bool EvalPatchesVarying(
\n+
1356 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1357 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1358 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1359 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1360 int numPatchCoords,
\n+
1361 PATCHCOORD_BUFFER *patchCoords,
\n+
1362 PATCH_TABLE *patchTable) const {
\n+
1363
\n+
1364 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1365 dstBuffer->BindVBO(), dstDesc,
\n+
1366 duBuffer->BindVBO(), duDesc,
\n+
1367 dvBuffer->BindVBO(), dvDesc,
\n+
1368 numPatchCoords,
\n+
1369 patchCoords->BindVBO(),
\n+
1370 patchTable->GetVaryingPatchArrays(),
\n+
1371 patchTable->GetVaryingPatchIndexBuffer(),
\n+
1372 patchTable->GetPatchParamBuffer());
\n+
1373 }
\n+
1374
\n+
1437 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1438 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1439 static bool EvalPatchesVarying(
\n+
1440 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1441 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1442 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1443 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1444 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1445 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1446 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1447 int numPatchCoords,
\n+
1448 PATCHCOORD_BUFFER *patchCoords,
\n+
1449 PATCH_TABLE *patchTable,
\n+
1450 GLComputeEvaluator const *instance,
\n+
1451 void * deviceContext = NULL) {
\n+
1452
\n+
1453 if (instance) {
\n+
1454 return instance->EvalPatchesVarying(
\n+
1455 srcBuffer, srcDesc,
\n+
1456 dstBuffer, dstDesc,
\n+
1457 duBuffer, duDesc,
\n+
1458 dvBuffer, dvDesc,
\n+
1459 duuBuffer, duuDesc,
\n+
1460 duvBuffer, duvDesc,
\n+
1461 dvvBuffer, dvvDesc,
\n+
1462 numPatchCoords, patchCoords,
\n+
1463 patchTable);
\n+
1464 } else {
\n+
1465 // Create an instance on demand (slow)
\n+
1466 (void)deviceContext; // unused
\n+
1467 instance = Create(srcDesc, dstDesc,
\n+
1468 duDesc, dvDesc,
\n+
1469 duuDesc, duvDesc, dvvDesc);
\n+
1470 if (instance) {
\n+
1471 bool r = instance->EvalPatchesVarying(
\n+
1472 srcBuffer, srcDesc,
\n+
1473 dstBuffer, dstDesc,
\n+
1474 duBuffer, duDesc,
\n+
1475 dvBuffer, dvDesc,
\n+
1476 duuBuffer, duuDesc,
\n+
1477 duvBuffer, duvDesc,
\n+
1478 dvvBuffer, dvvDesc,
\n+
1479 numPatchCoords, patchCoords,
\n+
1480 patchTable);
\n+
1481 delete instance;
\n+
1482 return r;
\n+
1483 }
\n+
1484 return false;
\n+
1485 }
\n+
1486 }
\n+
1487
\n+
1542 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1543 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1544 bool EvalPatchesVarying(
\n+
1545 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1546 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1547 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1548 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1549 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1550 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1551 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1552 int numPatchCoords,
\n+
1553 PATCHCOORD_BUFFER *patchCoords,
\n+
1554 PATCH_TABLE *patchTable) const {
\n+
1555
\n+
1556 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1557 dstBuffer->BindVBO(), dstDesc,
\n+
1558 duBuffer->BindVBO(), duDesc,
\n+
1559 dvBuffer->BindVBO(), dvDesc,
\n+
1560 duuBuffer->BindVBO(), duuDesc,
\n+
1561 duvBuffer->BindVBO(), duvDesc,
\n+
1562 dvvBuffer->BindVBO(), dvvDesc,
\n+
1563 numPatchCoords,
\n+
1564 patchCoords->BindVBO(),
\n+
1565 patchTable->GetVaryingPatchArrays(),
\n+
1566 patchTable->GetVaryingPatchIndexBuffer(),
\n+
1567 patchTable->GetPatchParamBuffer());
\n+
1568 }
\n+
1569
\n+
1604 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1605 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1606 static bool EvalPatchesFaceVarying(
\n+
1607 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1608 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1609 int numPatchCoords,
\n+
1610 PATCHCOORD_BUFFER *patchCoords,
\n+
1611 PATCH_TABLE *patchTable,
\n+
1612 int fvarChannel,
\n+
1613 GLComputeEvaluator const *instance,
\n+
1614 void * deviceContext = NULL) {
\n+
1615
\n+
1616 if (instance) {
\n+
1617 return instance->EvalPatchesFaceVarying(
\n+
1618 srcBuffer, srcDesc,
\n+
1619 dstBuffer, dstDesc,
\n+
1620 numPatchCoords, patchCoords,
\n+
1621 patchTable, fvarChannel);
\n+
1622 } else {
\n+
1623 // Create an instance on demand (slow)
\n+
1624 (void)deviceContext; // unused
\n+
1625 instance = Create(srcDesc, dstDesc,
\n+
1626 BufferDescriptor(),
\n+
1627 BufferDescriptor());
\n+
1628 if (instance) {
\n+
1629 bool r = instance->EvalPatchesFaceVarying(
\n+
1630 srcBuffer, srcDesc,
\n+
1631 dstBuffer, dstDesc,
\n+
1632 numPatchCoords, patchCoords,
\n+
1633 patchTable, fvarChannel);
\n+
1634 delete instance;
\n+
1635 return r;
\n+
1636 }
\n+
1637 return false;
\n+
1638 }
\n+
1639 }
\n+
1640
\n+
1667 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1668 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1669 bool EvalPatchesFaceVarying(
\n+
1670 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1671 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1672 int numPatchCoords,
\n+
1673 PATCHCOORD_BUFFER *patchCoords,
\n+
1674 PATCH_TABLE *patchTable,
\n+
1675 int fvarChannel = 0) const {
\n+
1676
\n+
1677 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1678 dstBuffer->BindVBO(), dstDesc,
\n+
1679 0, BufferDescriptor(),
\n+
1680 0, BufferDescriptor(),
\n+
1681 numPatchCoords,
\n+
1682 patchCoords->BindVBO(),
\n+
1683 patchTable->GetFVarPatchArrays(fvarChannel),
\n+
1684 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1685 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
1686 }
\n+
1687
\n+
1734 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1735 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1736 static bool EvalPatchesFaceVarying(
\n+
1737 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1738 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1739 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1740 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1741 int numPatchCoords,
\n+
1742 PATCHCOORD_BUFFER *patchCoords,
\n+
1743 PATCH_TABLE *patchTable,
\n+
1744 int fvarChannel,
\n+
1745 GLComputeEvaluator const *instance,
\n+
1746 void * deviceContext = NULL) {
\n+
1747
\n+
1748 if (instance) {
\n+
1749 return instance->EvalPatchesFaceVarying(
\n+
1750 srcBuffer, srcDesc,
\n+
1751 dstBuffer, dstDesc,
\n+
1752 duBuffer, duDesc,
\n+
1753 dvBuffer, dvDesc,
\n+
1754 numPatchCoords, patchCoords,
\n+
1755 patchTable, fvarChannel);
\n+
1756 } else {
\n+
1757 // Create an instance on demand (slow)
\n+
1758 (void)deviceContext; // unused
\n+
1759 instance = Create(srcDesc, dstDesc,
\n+
1760 duDesc, dvDesc);
\n+
1761 if (instance) {
\n+
1762 bool r = instance->EvalPatchesFaceVarying(
\n+
1763 srcBuffer, srcDesc,
\n+
1764 dstBuffer, dstDesc,
\n+
1765 duBuffer, duDesc,
\n+
1766 dvBuffer, dvDesc,
\n+
1767 numPatchCoords, patchCoords,
\n+
1768 patchTable, fvarChannel);
\n+
1769 delete instance;
\n+
1770 return r;
\n+
1771 }
\n+
1772 return false;
\n+
1773 }
\n+
1774 }
\n+
1775
\n+
1814 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1815 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1816 bool EvalPatchesFaceVarying(
\n+
1817 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1818 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1819 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1820 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1821 int numPatchCoords,
\n+
1822 PATCHCOORD_BUFFER *patchCoords,
\n+
1823 PATCH_TABLE *patchTable,
\n+
1824 int fvarChannel = 0) const {
\n+
1825
\n+
1826 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
1827 dstBuffer->BindVBO(), dstDesc,
\n+
1828 duBuffer->BindVBO(), duDesc,
\n+
1829 dvBuffer->BindVBO(), dvDesc,
\n+
1830 numPatchCoords,
\n+
1831 patchCoords->BindVBO(),
\n+
1832 patchTable->GetFVarPatchArrays(fvarChannel),
\n+
1833 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1834 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
1835 }
\n+
1836
\n+
1901 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1902 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1903 static bool EvalPatchesFaceVarying(
\n+
1904 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1905 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1906 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1907 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1908 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1909 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1910 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1911 int numPatchCoords,
\n+
1912 PATCHCOORD_BUFFER *patchCoords,
\n+
1913 PATCH_TABLE *patchTable,
\n+
1914 int fvarChannel,
\n+
1915 GLComputeEvaluator const *instance,
\n+
1916 void * deviceContext = NULL) {
\n+
1917
\n+
1918 if (instance) {
\n+
1919 return instance->EvalPatchesFaceVarying(
\n+
1920 srcBuffer, srcDesc,
\n+
1921 dstBuffer, dstDesc,
\n+
1922 duBuffer, duDesc,
\n+
1923 dvBuffer, dvDesc,
\n+
1924 duuBuffer, duuDesc,
\n+
1925 duvBuffer, duvDesc,
\n+
1926 dvvBuffer, dvvDesc,
\n+
1927 numPatchCoords, patchCoords,
\n+
1928 patchTable, fvarChannel);
\n+
1929 } else {
\n+
1930 // Create an instance on demand (slow)
\n+
1931 (void)deviceContext; // unused
\n+
1932 instance = Create(srcDesc, dstDesc,
\n+
1933 duDesc, dvDesc,
\n+
1934 duuDesc, duvDesc, dvvDesc);
\n+
1935 if (instance) {
\n+
1936 bool r = instance->EvalPatchesFaceVarying(
\n+
1937 srcBuffer, srcDesc,
\n+
1938 dstBuffer, dstDesc,
\n+
1939 duBuffer, duDesc,
\n+
1940 dvBuffer, dvDesc,
\n+
1941 duuBuffer, duuDesc,
\n+
1942 duvBuffer, duvDesc,
\n+
1943 dvvBuffer, dvvDesc,
\n+
1944 numPatchCoords, patchCoords,
\n+
1945 patchTable, fvarChannel);
\n+
1946 delete instance;
\n+
1947 return r;
\n+
1948 }
\n+
1949 return false;
\n+
1950 }
\n+
1951 }
\n+
1952
\n+
2009 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
2010 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
2011 bool EvalPatchesFaceVarying(
\n+
2012 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
2013 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
2014 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
2015 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
2016 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
2017 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
2018 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
2019 int numPatchCoords,
\n+
2020 PATCHCOORD_BUFFER *patchCoords,
\n+
2021 PATCH_TABLE *patchTable,
\n+
2022 int fvarChannel = 0) const {
\n+
2023
\n+
2024 return EvalPatches(srcBuffer->BindVBO(), srcDesc,
\n+
2025 dstBuffer->BindVBO(), dstDesc,
\n+
2026 duBuffer->BindVBO(), duDesc,
\n+
2027 dvBuffer->BindVBO(), dvDesc,
\n+
2028 duuBuffer->BindVBO(), duuDesc,
\n+
2029 duvBuffer->BindVBO(), duvDesc,
\n+
2030 dvvBuffer->BindVBO(), dvvDesc,
\n+
2031 numPatchCoords,
\n+
2032 patchCoords->BindVBO(),
\n+
2033 patchTable->GetFVarPatchArrays(fvarChannel),
\n+
2034 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
2035 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
2036 }
\n+
2037
\n+
2043
\n+
2046 bool Compile(BufferDescriptor const &srcDesc,
\n+
2047 BufferDescriptor const &dstDesc,
\n+
2048 BufferDescriptor const &duDesc = BufferDescriptor(),
\n+
2049 BufferDescriptor const &dvDesc = BufferDescriptor(),
\n+
2050 BufferDescriptor const &duuDesc = BufferDescriptor(),
\n+
2051 BufferDescriptor const &duvDesc = BufferDescriptor(),
\n+
2052 BufferDescriptor const &dvvDesc = BufferDescriptor());
\n+
2053
\n+
2055 static void Synchronize(void *deviceContext);
\n+
2056
\n+
2057private:
\n+
2058 struct _StencilKernel {
\n+
2059 _StencilKernel();
\n+
2060 ~_StencilKernel();
\n+
2061 bool Compile(BufferDescriptor const &srcDesc,
\n+
2062 BufferDescriptor const &dstDesc,
\n+
2063 BufferDescriptor const &duDesc,
\n+
2064 BufferDescriptor const &dvDesc,
\n+
2065 BufferDescriptor const &duuDesc,
\n+
2066 BufferDescriptor const &duvDesc,
\n+
2067 BufferDescriptor const &dvvDesc,
\n+
2068 int workGroupSize);
\n+
2069 GLuint program;
\n+
2070 GLuint uniformStart;
\n+
2071 GLuint uniformEnd;
\n+
2072 GLuint uniformSrcOffset;
\n+
2073 GLuint uniformDstOffset;
\n+
2074 GLuint uniformDuDesc;
\n+
2075 GLuint uniformDvDesc;
\n+
2076 GLuint uniformDuuDesc;
\n+
2077 GLuint uniformDuvDesc;
\n+
2078 GLuint uniformDvvDesc;
\n+
2079 } _stencilKernel;
\n+
2080
\n+
2081 struct _PatchKernel {
\n+
2082 _PatchKernel();
\n+
2083 ~_PatchKernel();
\n+
2084 bool Compile(BufferDescriptor const &srcDesc,
\n+
2085 BufferDescriptor const &dstDesc,
\n+
2086 BufferDescriptor const &duDesc,
\n+
2087 BufferDescriptor const &dvDesc,
\n+
2088 BufferDescriptor const &duuDesc,
\n+
2089 BufferDescriptor const &duvDesc,
\n+
2090 BufferDescriptor const &dvvDesc,
\n+
2091 int workGroupSize);
\n+
2092 GLuint program;
\n+
2093 GLuint uniformSrcOffset;
\n+
2094 GLuint uniformDstOffset;
\n+
2095 GLuint uniformPatchArray;
\n+
2096 GLuint uniformDuDesc;
\n+
2097 GLuint uniformDvDesc;
\n+
2098 GLuint uniformDuuDesc;
\n+
2099 GLuint uniformDuvDesc;
\n+
2100 GLuint uniformDvvDesc;
\n+
2101 } _patchKernel;
\n+
2102
\n+
2103 int _workGroupSize;
\n+
2104 GLuint _patchArraysSSBO;
\n+
2105};
\n+
2106
\n+
2107} // end namespace Osd
\n+
2108
\n+
2109} // end namespace OPENSUBDIV_VERSION
\n+
2110using namespace OPENSUBDIV_VERSION;
\n+
2111
\n+
2112} // end namespace OpenSubdiv
\n+
2113
\n+
2114
\n+
2115#endif // OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::Create
static GLStencilTableTBO * Create(Far::StencilTable const *stencilTable, void *deviceContext=NULL)
Definition: glXFBEvaluator.h:54
\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GLStencilTableTBO
GLStencilTableTBO(Far::StencilTable const *stencilTable)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GLStencilTableTBO
GLStencilTableTBO(Far::LimitStencilTable const *limitStencilTable)
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::Create
static GLStencilTableTBO * Create(Far::LimitStencilTable const *limitStencilTable, void *deviceContext=NULL)
Definition: glXFBEvaluator.h:60
\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1287
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, bool interleavedDerivativeBuffers=false)
Definition: glXFBEvaluator.h:126
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glXFBEvaluator.h:506
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::GLXFBEvaluator
GLXFBEvaluator(bool interleavedDerivativeBuffers=false)
Constructor.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glXFBEvaluator.h:575
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays, GLuint patchIndexBuffer, GLuint patchParamsBuffer) const
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glXFBEvaluator.h:396
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: glXFBEvaluator.h:1144
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, bool interleavedDerivativeBuffers=false)
Definition: glXFBEvaluator.h:170
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, void *deviceContext)
Specialization to allow creation without a device context.
Definition: glXFBEvaluator.h:117
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glXFBEvaluator.h:306
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1351
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, GLuint sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer, GLuint duWeightsBuffer, GLuint dvWeightsBuffer, GLuint duuWeightsBuffer, GLuint duvWeightsBuffer, GLuint dvvWeightsBuffer, int start, int end) const
Dispatch the GLSL XFB kernel on on GPU asynchronously returns false if the kernel hasn't been compile...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:843
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, DEVICE_CONTEXT deviceContext)
Generic creator template.
Definition: glXFBEvaluator.h:140
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, DEVICE_CONTEXT deviceContext)
Generic creator template.
Definition: glXFBEvaluator.h:104
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Synchronize
static void Synchronize(void *kernel)
Wait the dispatched kernel finishes.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1227
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1617
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1889
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:765
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1809
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1016
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:943
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: glXFBEvaluator.h:1070
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glXFBEvaluator.h:457
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches
bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays, GLuint patchIndexBuffer, GLuint patchParamsBuffer) const
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create
static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, void *deviceContext)
Specialization to allow creation without a device context.
Definition: glXFBEvaluator.h:157
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:2084
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1512
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer, GLuint duWeightsBuffer, GLuint dvWeightsBuffer, int start, int end) const
Dispatch the GLSL XFB kernel on on GPU asynchronously returns false if the kernel hasn't been compile...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1679
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glXFBEvaluator.h:238
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::~GLXFBEvaluator
~GLXFBEvaluator()
Destructor. note that the GL context must be made current.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Compile
bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc=BufferDescriptor(), BufferDescriptor const &dvDesc=BufferDescriptor(), BufferDescriptor const &duuDesc=BufferDescriptor(), BufferDescriptor const &duvDesc=BufferDescriptor(), BufferDescriptor const &dvvDesc=BufferDescriptor())
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1428
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1742
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLXFBEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glXFBEvaluator.h:1976
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO
GL stencil table (Shader Storage buffer)
Definition: glComputeEvaluator.h:51
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GLStencilTableSSBO
GLStencilTableSSBO(Far::LimitStencilTable const *limitStencilTable)
\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GLStencilTableSSBO
GLStencilTableSSBO(Far::StencilTable const *stencilTable)
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::Create
static GLStencilTableSSBO * Create(Far::LimitStencilTable const *limitStencilTable, void *deviceContext=NULL)
Definition: glComputeEvaluator.h:58
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::Create
static GLStencilTableSSBO * Create(Far::StencilTable const *stencilTable, void *deviceContext=NULL)
Definition: glComputeEvaluator.h:53
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1214
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1736
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glComputeEvaluator.h:434
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1154
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1606
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glComputeEvaluator.h:502
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1278
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays, GLuint patchIndexBuffer, GLuint patchParamsBuffer) const
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::~GLComputeEvaluator
~GLComputeEvaluator()
Destructor. note that the GL context must be made current.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: glComputeEvaluator.h:1071
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glComputeEvaluator.h:236
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, GLuint sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer, GLuint duWeightsBuffer, GLuint dvWeightsBuffer, GLuint duuWeightsBuffer, GLuint duvWeightsBuffer, GLuint dvvWeightsBuffer, int start, int end) const
Dispatch the GLSL compute kernel on GPU asynchronously returns false if the kernel hasn't been compil...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glComputeEvaluator.h:168
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1439
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:769
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1903
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1544
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1816
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:943
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: glComputeEvaluator.h:997
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:691
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable) const
Generic stencil function.
Definition: glComputeEvaluator.h:386
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays, GLuint patchIndexBuffer, GLuint patchParamsBuffer) const
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:2011
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Create
static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, void *deviceContext=NULL)
Definition: glComputeEvaluator.h:99
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer, GLuint duWeightsBuffer, GLuint dvWeightsBuffer, int start, int end) const
Dispatch the GLSL compute kernel on GPU asynchronously returns false if the kernel hasn't been compil...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic static stencil function. This function has a same signature as other device kernels have so t...
Definition: glComputeEvaluator.h:326
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Synchronize
static void Synchronize(void *deviceContext)
Wait the dispatched kernel finishes.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Compile
bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc=BufferDescriptor(), BufferDescriptor const &dvDesc=BufferDescriptor(), BufferDescriptor const &duuDesc=BufferDescriptor(), BufferDescriptor const &duvDesc=BufferDescriptor(), BufferDescriptor const &dvvDesc=BufferDescriptor())
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying
bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1355
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Create
static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc, BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc, BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, void *deviceContext=NULL)
Definition: glComputeEvaluator.h:111
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying
bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:1669
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: glComputeEvaluator.h:870
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-glXFBEvaluator.h\n+glComputeEvaluator.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,16 +30,16 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H\n- 26#define OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H\n+ 25#ifndef OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H\n+ 26#define OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H\n 27\n 28#include \"../version.h\"\n 29\n 30#include \"../osd/opengl.h\"\n 31#include \"../osd/types.h\"\n 32#include \"../osd/bufferDescriptor.h\"\n 33\n@@ -50,1061 +50,999 @@\n 38 class PatchTable;\n 39 class StencilTable;\n 40 class LimitStencilTable;\n 41}\n 42\n 43namespace Osd {\n 44\n-52class GLStencilTableTBO {\n- 53public:\n-54 static GLStencilTableTBO *Create(\n- 55 Far::StencilTable const *stencilTable, void *deviceContext = NULL) {\n- 56 (void)deviceContext; // unused\n- 57 return new GLStencilTableTBO(stencilTable);\n- 58 }\n- 59\n-60 static GLStencilTableTBO *Create(\n- 61 Far::LimitStencilTable const *limitStencilTable,\n- 62 void *deviceContext = NULL) {\n- 63 (void)deviceContext; // unused\n- 64 return new GLStencilTableTBO(limitStencilTable);\n- 65 }\n- 66\n-67 explicit GLStencilTableTBO(Far::StencilTable const *stencilTable);\n-68 explicit GLStencilTableTBO(Far::LimitStencilTable const *limitStencilTable);\n-69 ~GLStencilTableTBO();\n- 70\n- 71 // interfaces needed for GLSLTransformFeedbackKernel\n-72 GLuint GetSizesTexture() const { return _sizes; }\n-73 GLuint GetOffsetsTexture() const { return _offsets; }\n-74 GLuint GetIndicesTexture() const { return _indices; }\n-75 GLuint GetWeightsTexture() const { return _weights; }\n-76 GLuint GetDuWeightsTexture() const { return _duWeights; }\n-77 GLuint GetDvWeightsTexture() const { return _dvWeights; }\n-78 GLuint GetDuuWeightsTexture() const { return _duuWeights; }\n-79 GLuint GetDuvWeightsTexture() const { return _duvWeights; }\n-80 GLuint GetDvvWeightsTexture() const { return _dvvWeights; }\n-81 int GetNumStencils() const { return _numStencils; }\n- 82\n- 83private:\n- 84 GLuint _sizes;\n- 85 GLuint _offsets;\n- 86 GLuint _indices;\n- 87 GLuint _weights;\n- 88 GLuint _duWeights;\n- 89 GLuint _dvWeights;\n- 90 GLuint _duuWeights;\n- 91 GLuint _duvWeights;\n- 92 GLuint _dvvWeights;\n- 93 int _numStencils;\n- 94};\n- 95\n- 96// -------------------------------------------------------------------------\n+51class GLStencilTableSSBO {\n+ 52public:\n+53 static GLStencilTableSSBO *Create(Far::StencilTable const *stencilTable,\n+ 54 void *deviceContext = NULL) {\n+ 55 (void)deviceContext; // unused\n+ 56 return new GLStencilTableSSBO(stencilTable);\n+ 57 }\n+58 static GLStencilTableSSBO *Create(\n+ 59 Far::LimitStencilTable const *limitStencilTable,\n+ 60 void *deviceContext = NULL) {\n+ 61 (void)deviceContext; // unused\n+ 62 return new GLStencilTableSSBO(limitStencilTable);\n+ 63 }\n+ 64\n+65 explicit GLStencilTableSSBO(Far::StencilTable const *stencilTable);\n+66 explicit GLStencilTableSSBO(Far::LimitStencilTable const\n+*limitStencilTable);\n+67 ~GLStencilTableSSBO();\n+ 68\n+ 69 // interfaces needed for GLSLComputeKernel\n+70 GLuint GetSizesBuffer() const { return _sizes; }\n+71 GLuint GetOffsetsBuffer() const { return _offsets; }\n+72 GLuint GetIndicesBuffer() const { return _indices; }\n+73 GLuint GetWeightsBuffer() const { return _weights; }\n+74 GLuint GetDuWeightsBuffer() const { return _duWeights; }\n+75 GLuint GetDvWeightsBuffer() const { return _dvWeights; }\n+76 GLuint GetDuuWeightsBuffer() const { return _duuWeights; }\n+77 GLuint GetDuvWeightsBuffer() const { return _duvWeights; }\n+78 GLuint GetDvvWeightsBuffer() const { return _dvvWeights; }\n+79 int GetNumStencils() const { return _numStencils; }\n+ 80\n+ 81private:\n+ 82 GLuint _sizes;\n+ 83 GLuint _offsets;\n+ 84 GLuint _indices;\n+ 85 GLuint _weights;\n+ 86 GLuint _duWeights;\n+ 87 GLuint _dvWeights;\n+ 88 GLuint _duuWeights;\n+ 89 GLuint _duvWeights;\n+ 90 GLuint _dvvWeights;\n+ 91 int _numStencils;\n+ 92};\n+ 93\n+ 94// -------------------------------------------------------------------------\n --\n- 97\n-98class GLXFBEvaluator {\n- 99public:\n-100 typedef bool Instantiatable;\n- 101\n- 103 template \n-104 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,\n- 105 BufferDescriptor const &dstDesc,\n- 106 BufferDescriptor const &duDesc,\n- 107 BufferDescriptor const &dvDesc,\n- 108 DEVICE_CONTEXT deviceContext) {\n- 109 bool interleavedDerivativeBuffers = deviceContext\n- 110 ? deviceContext->AreInterleavedDerivativeBuffers()\n- 111 : false;\n- 112 return Create(srcDesc, dstDesc, duDesc, dvDesc,\n- 113 interleavedDerivativeBuffers);\n- 114 }\n- 115\n-117 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,\n- 118 BufferDescriptor const &dstDesc,\n- 119 BufferDescriptor const &duDesc,\n- 120 BufferDescriptor const &dvDesc,\n- 121 void * deviceContext) {\n- 122 (void)deviceContext; // not used\n- 123 return Create(srcDesc, dstDesc, duDesc, dvDesc);\n- 124 }\n- 125\n-126 static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n- 127 BufferDescriptor const &dstDesc,\n- 128 BufferDescriptor const &duDesc,\n- 129 BufferDescriptor const &dvDesc,\n- 130 bool interleavedDerivativeBuffers = false) {\n- 131 GLXFBEvaluator *instance = new GLXFBEvaluator\n-(interleavedDerivativeBuffers);\n- 132 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc))\n- 133 return instance;\n- 134 delete instance;\n- 135 return NULL;\n- 136 }\n- 137\n- 139 template \n-140 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,\n- 141 BufferDescriptor const &dstDesc,\n- 142 BufferDescriptor const &duDesc,\n- 143 BufferDescriptor const &dvDesc,\n- 144 BufferDescriptor const &duuDesc,\n- 145 BufferDescriptor const &duvDesc,\n- 146 BufferDescriptor const &dvvDesc,\n- 147 DEVICE_CONTEXT deviceContext) {\n- 148 bool interleavedDerivativeBuffers = deviceContext\n- 149 ? deviceContext->AreInterleavedDerivativeBuffers()\n- 150 : false;\n- 151 return Create(srcDesc, dstDesc, duDesc, dvDesc,\n- 152 duuDesc, duvDesc, dvvDesc,\n- 153 interleavedDerivativeBuffers);\n- 154 }\n- 155\n-157 static GLXFBEvaluator *Create(BufferDescriptor const &srcDesc,\n- 158 BufferDescriptor const &dstDesc,\n- 159 BufferDescriptor const &duDesc,\n- 160 BufferDescriptor const &dvDesc,\n- 161 BufferDescriptor const &duuDesc,\n- 162 BufferDescriptor const &duvDesc,\n- 163 BufferDescriptor const &dvvDesc,\n- 164 void * deviceContext) {\n- 165 (void)deviceContext; // not used\n- 166 return Create(srcDesc, dstDesc, duDesc, dvDesc,\n- 167 duuDesc, duvDesc, dvvDesc);\n- 168 }\n- 169\n-170 static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n- 171 BufferDescriptor const &dstDesc,\n- 172 BufferDescriptor const &duDesc,\n- 173 BufferDescriptor const &dvDesc,\n- 174 BufferDescriptor const &duuDesc,\n- 175 BufferDescriptor const &duvDesc,\n- 176 BufferDescriptor const &dvvDesc,\n- 177 bool interleavedDerivativeBuffers = false) {\n- 178 GLXFBEvaluator *instance = new GLXFBEvaluator\n-(interleavedDerivativeBuffers);\n- 179 if (instance->Compile(srcDesc, dstDesc, duDesc, dvDesc,\n- 180 duuDesc, duvDesc, dvvDesc))\n- 181 return instance;\n- 182 delete instance;\n- 183 return NULL;\n- 184 }\n- 185\n-199 GLXFBEvaluator(bool interleavedDerivativeBuffers = false);\n- 200\n-202 ~GLXFBEvaluator();\n- 203\n- 209\n- 237 template Compile(srcDesc, dstDesc, duDesc, dvDesc,\n+ 122 duuDesc, duvDesc, dvvDesc))\n+ 123 return instance;\n+ 124 delete instance;\n+ 125 return NULL;\n+ 126 }\n+ 127\n+129 GLComputeEvaluator();\n+ 130\n+132 ~GLComputeEvaluator();\n+ 133\n+ 139\n+ 167 template \n+168 static bool EvalStencils(\n+ 169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 171 STENCIL_TABLE const *stencilTable,\n+ 172 GLComputeEvaluator const *instance,\n+ 173 void * deviceContext = NULL) {\n+ 174\n+ 175 if (instance) {\n+ 176 return instance->EvalStencils(srcBuffer, srcDesc,\n+ 177 dstBuffer, dstDesc,\n+ 178 stencilTable);\n+ 179 } else {\n+ 180 // Create an instance on demand (slow)\n+ 181 (void)deviceContext; // unused\n+ 182 instance = Create(srcDesc, dstDesc,\n+ 183 BufferDescriptor(),\n+ 184 BufferDescriptor());\n+ 185 if (instance) {\n+ 186 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 187 dstBuffer, dstDesc,\n+ 188 stencilTable);\n+ 189 delete instance;\n+ 190 return r;\n+ 191 }\n+ 192 return false;\n+ 193 }\n+ 194 }\n+ 195\n+ 235 template \n-238 static bool EvalStencils(\n- 239 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 240 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+236 static bool EvalStencils(\n+ 237 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 238 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 239 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 240 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n 241 STENCIL_TABLE const *stencilTable,\n- 242 GLXFBEvaluator const *instance,\n+ 242 GLComputeEvaluator const *instance,\n 243 void * deviceContext = NULL) {\n 244\n 245 if (instance) {\n 246 return instance->EvalStencils(srcBuffer, srcDesc,\n 247 dstBuffer, dstDesc,\n- 248 stencilTable);\n- 249 } else {\n- 250 // Create an instance on demand (slow)\n- 251 (void)deviceContext; // unused\n- 252 instance = Create(srcDesc, dstDesc,\n- 253 BufferDescriptor(),\n- 254 BufferDescriptor());\n+ 248 duBuffer, duDesc,\n+ 249 dvBuffer, dvDesc,\n+ 250 stencilTable);\n+ 251 } else {\n+ 252 // Create an instance on demand (slow)\n+ 253 (void)deviceContext; // unused\n+ 254 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);\n 255 if (instance) {\n 256 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n 257 dstBuffer, dstDesc,\n- 258 stencilTable);\n- 259 delete instance;\n- 260 return r;\n- 261 }\n- 262 return false;\n+ 258 duBuffer, duDesc,\n+ 259 dvBuffer, dvDesc,\n+ 260 stencilTable);\n+ 261 delete instance;\n+ 262 return r;\n 263 }\n- 264 }\n- 265\n- 305 template \n-306 static bool EvalStencils(\n- 307 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 308 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 309 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 310 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 311 STENCIL_TABLE const *stencilTable,\n- 312 GLXFBEvaluator const *instance,\n- 313 void * deviceContext = NULL) {\n- 314\n- 315 if (instance) {\n- 316 return instance->EvalStencils(srcBuffer, srcDesc,\n- 317 dstBuffer, dstDesc,\n- 318 duBuffer, duDesc,\n- 319 dvBuffer, dvDesc,\n- 320 stencilTable);\n- 321 } else {\n- 322 // Create an instance on demand (slow)\n- 323 (void)deviceContext; // unused\n- 324 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);\n- 325 if (instance) {\n- 326 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 327 dstBuffer, dstDesc,\n- 328 duBuffer, duDesc,\n- 329 dvBuffer, dvDesc,\n- 330 stencilTable);\n- 331 delete instance;\n- 332 return r;\n- 333 }\n- 334 return false;\n- 335 }\n- 336 }\n+326 static bool EvalStencils(\n+ 327 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 328 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 329 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 330 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 331 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 332 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 333 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 334 STENCIL_TABLE const *stencilTable,\n+ 335 GLComputeEvaluator const *instance,\n+ 336 void * deviceContext = NULL) {\n 337\n- 395 template EvalStencils(srcBuffer, srcDesc,\n+ 340 dstBuffer, dstDesc,\n+ 341 duBuffer, duDesc,\n+ 342 dvBuffer, dvDesc,\n+ 343 duuBuffer, duuDesc,\n+ 344 duvBuffer, duvDesc,\n+ 345 dvvBuffer, dvvDesc,\n+ 346 stencilTable);\n+ 347 } else {\n+ 348 // Create an instance on demand (slow)\n+ 349 (void)deviceContext; // unused\n+ 350 instance = Create(srcDesc, dstDesc, duDesc, dvDesc,\n+ 351 duuDesc, duvDesc, dvvDesc);\n+ 352 if (instance) {\n+ 353 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n+ 354 dstBuffer, dstDesc,\n+ 355 duBuffer, duDesc,\n+ 356 dvBuffer, dvDesc,\n+ 357 duuBuffer, duuDesc,\n+ 358 duvBuffer, duvDesc,\n+ 359 dvvBuffer, dvvDesc,\n+ 360 stencilTable);\n+ 361 delete instance;\n+ 362 return r;\n+ 363 }\n+ 364 return false;\n+ 365 }\n+ 366 }\n+ 367\n+ 385 template \n-396 static bool EvalStencils(\n- 397 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 398 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 399 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 400 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 401 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 402 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 403 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 404 STENCIL_TABLE const *stencilTable,\n- 405 GLXFBEvaluator const *instance,\n- 406 void * deviceContext = NULL) {\n- 407\n- 408 if (instance) {\n- 409 return instance->EvalStencils(srcBuffer, srcDesc,\n- 410 dstBuffer, dstDesc,\n- 411 duBuffer, duDesc,\n- 412 dvBuffer, dvDesc,\n- 413 duuBuffer, duuDesc,\n- 414 duvBuffer, duvDesc,\n- 415 dvvBuffer, dvvDesc,\n- 416 stencilTable);\n- 417 } else {\n- 418 // Create an instance on demand (slow)\n- 419 (void)deviceContext; // unused\n- 420 instance = Create(srcDesc, dstDesc,\n- 421 duDesc, dvDesc,\n- 422 duuDesc, duvDesc, dvvDesc);\n- 423 if (instance) {\n- 424 bool r = instance->EvalStencils(srcBuffer, srcDesc,\n- 425 dstBuffer, dstDesc,\n- 426 duBuffer, duDesc,\n- 427 dvBuffer, dvDesc,\n- 428 duuBuffer, duuDesc,\n- 429 duvBuffer, duvDesc,\n- 430 dvvBuffer, dvvDesc,\n- 431 stencilTable);\n- 432 delete instance;\n- 433 return r;\n- 434 }\n- 435 return false;\n- 436 }\n- 437 }\n- 438\n- 456 template BindVBO(), srcDesc,\n+ 391 dstBuffer->BindVBO(), dstDesc,\n+ 392 0, BufferDescriptor(),\n+ 393 0, BufferDescriptor(),\n+ 394 stencilTable->GetSizesBuffer(),\n+ 395 stencilTable->GetOffsetsBuffer(),\n+ 396 stencilTable->GetIndicesBuffer(),\n+ 397 stencilTable->GetWeightsBuffer(),\n+ 398 0,\n+ 399 0,\n+ 400 /* start = */ 0,\n+ 401 /* end = */ stencilTable->GetNumStencils());\n+ 402 }\n+ 403\n+ 433 template \n-457 bool EvalStencils(\n- 458 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 459 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 460 STENCIL_TABLE const *stencilTable) const {\n- 461\n- 462 return EvalStencils(srcBuffer->BindVBO(), srcDesc,\n- 463 dstBuffer->BindVBO(), dstDesc,\n- 464 0, BufferDescriptor(),\n- 465 0, BufferDescriptor(),\n- 466 stencilTable->GetSizesTexture(),\n- 467 stencilTable->GetOffsetsTexture(),\n- 468 stencilTable->GetIndicesTexture(),\n- 469 stencilTable->GetWeightsTexture(),\n- 470 0,\n- 471 0,\n- 472 /* start = */ 0,\n- 473 /* end = */ stencilTable->GetNumStencils());\n- 474 }\n- 475\n- 505 template BindVBO(), srcDesc,\n+ 441 dstBuffer->BindVBO(), dstDesc,\n+ 442 duBuffer->BindVBO(), duDesc,\n+ 443 dvBuffer->BindVBO(), dvDesc,\n+ 444 stencilTable->GetSizesBuffer(),\n+ 445 stencilTable->GetOffsetsBuffer(),\n+ 446 stencilTable->GetIndicesBuffer(),\n+ 447 stencilTable->GetWeightsBuffer(),\n+ 448 stencilTable->GetDuWeightsBuffer(),\n+ 449 stencilTable->GetDvWeightsBuffer(),\n+ 450 /* start = */ 0,\n+ 451 /* end = */ stencilTable->GetNumStencils());\n+ 452 }\n+ 453\n+ 501 template \n-506 bool EvalStencils(\n- 507 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 508 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 509 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 510 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 511 STENCIL_TABLE const *stencilTable) const {\n- 512\n- 513 return EvalStencils(srcBuffer->BindVBO(), srcDesc,\n- 514 dstBuffer->BindVBO(), dstDesc,\n- 515 duBuffer->BindVBO(), duDesc,\n- 516 dvBuffer->BindVBO(), dvDesc,\n- 517 stencilTable->GetSizesTexture(),\n- 518 stencilTable->GetOffsetsTexture(),\n- 519 stencilTable->GetIndicesTexture(),\n- 520 stencilTable->GetWeightsTexture(),\n- 521 stencilTable->GetDuWeightsTexture(),\n- 522 stencilTable->GetDvWeightsTexture(),\n- 523 /* start = */ 0,\n- 524 /* end = */ stencilTable->GetNumStencils());\n- 525 }\n- 526\n- 574 template \n-575 bool EvalStencils(\n- 576 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 577 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 578 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 579 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 580 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 581 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 582 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 583 STENCIL_TABLE const *stencilTable) const {\n- 584\n- 585 return EvalStencils(srcBuffer->BindVBO(), srcDesc,\n- 586 dstBuffer->BindVBO(), dstDesc,\n- 587 duBuffer->BindVBO(), duDesc,\n- 588 dvBuffer->BindVBO(), dvDesc,\n- 589 duuBuffer->BindVBO(), duuDesc,\n- 590 duvBuffer->BindVBO(), duvDesc,\n- 591 dvvBuffer->BindVBO(), dvvDesc,\n- 592 stencilTable->GetSizesTexture(),\n- 593 stencilTable->GetOffsetsTexture(),\n- 594 stencilTable->GetIndicesTexture(),\n- 595 stencilTable->GetWeightsTexture(),\n- 596 stencilTable->GetDuWeightsTexture(),\n- 597 stencilTable->GetDvWeightsTexture(),\n- 598 stencilTable->GetDuuWeightsTexture(),\n- 599 stencilTable->GetDuvWeightsTexture(),\n- 600 stencilTable->GetDvvWeightsTexture(),\n- 601 /* start = */ 0,\n- 602 /* end = */ stencilTable->GetNumStencils());\n- 603 }\n- 604\n-640 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n- 641 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n- 642 GLuint duBuffer, BufferDescriptor const &duDesc,\n- 643 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n- 644 GLuint sizesBuffer,\n- 645 GLuint offsetsBuffer,\n- 646 GLuint indicesBuffer,\n- 647 GLuint weightsBuffer,\n- 648 GLuint duWeightsBuffer,\n- 649 GLuint dvWeightsBuffer,\n- 650 int start,\n- 651 int end) const;\n- 652\n-706 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n- 707 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n- 708 GLuint duBuffer, BufferDescriptor const &duDesc,\n- 709 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n- 710 GLuint duuBuffer, BufferDescriptor const &duuDesc,\n- 711 GLuint duvBuffer, BufferDescriptor const &duvDesc,\n- 712 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,\n- 713 GLuint sizesBuffer,\n- 714 GLuint offsetsBuffer,\n- 715 GLuint indicesBuffer,\n- 716 GLuint weightsBuffer,\n- 717 GLuint duWeightsBuffer,\n- 718 GLuint dvWeightsBuffer,\n- 719 GLuint duuWeightsBuffer,\n- 720 GLuint duvWeightsBuffer,\n- 721 GLuint dvvWeightsBuffer,\n- 722 int start,\n- 723 int end) const;\n- 724\n- 730\n- 763 template \n-765 static bool EvalPatches(\n- 766 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 767 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 768 int numPatchCoords,\n- 769 PATCHCOORD_BUFFER *patchCoords,\n- 770 PATCH_TABLE *patchTable,\n- 771 GLXFBEvaluator const *instance,\n- 772 void * deviceContext = NULL) {\n- 773\n- 774 if (instance) {\n- 775 return instance->EvalPatches(srcBuffer, srcDesc,\n- 776 dstBuffer, dstDesc,\n- 777 numPatchCoords, patchCoords,\n- 778 patchTable);\n- 779 } else {\n- 780 // Create an instance on demand (slow)\n- 781 (void)deviceContext; // unused\n- 782 instance = Create(srcDesc, dstDesc,\n- 783 BufferDescriptor(),\n- 784 BufferDescriptor());\n- 785 if (instance) {\n- 786 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 787 dstBuffer, dstDesc,\n- 788 numPatchCoords, patchCoords,\n- 789 patchTable);\n- 790 delete instance;\n- 791 return r;\n- 792 }\n- 793 return false;\n- 794 }\n- 795 }\n- 796\n- 841 template \n-843 static bool EvalPatches(\n- 844 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 845 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 846 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 847 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 848 int numPatchCoords,\n- 849 PATCHCOORD_BUFFER *patchCoords,\n- 850 PATCH_TABLE *patchTable,\n- 851 GLXFBEvaluator const *instance,\n- 852 void * deviceContext = NULL) {\n- 853\n- 854 if (instance) {\n- 855 return instance->EvalPatches(srcBuffer, srcDesc,\n- 856 dstBuffer, dstDesc,\n- 857 duBuffer, duDesc,\n- 858 dvBuffer, dvDesc,\n- 859 numPatchCoords, patchCoords,\n- 860 patchTable);\n- 861 } else {\n- 862 // Create an instance on demand (slow)\n- 863 (void)deviceContext; // unused\n- 864 instance = Create(srcDesc, dstDesc, duDesc, dvDesc);\n- 865 if (instance) {\n- 866 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 867 dstBuffer, dstDesc,\n- 868 duBuffer, duDesc,\n- 869 dvBuffer, dvDesc,\n- 870 numPatchCoords, patchCoords,\n- 871 patchTable);\n- 872 delete instance;\n- 873 return r;\n- 874 }\n- 875 return false;\n- 876 }\n- 877 }\n- 878\n+502 bool EvalStencils(\n+ 503 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 504 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 505 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 506 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 507 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 508 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 509 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 510 STENCIL_TABLE const *stencilTable) const {\n+ 511 return EvalStencils(srcBuffer->BindVBO(), srcDesc,\n+ 512 dstBuffer->BindVBO(), dstDesc,\n+ 513 duBuffer->BindVBO(), duDesc,\n+ 514 dvBuffer->BindVBO(), dvDesc,\n+ 515 duuBuffer->BindVBO(), duuDesc,\n+ 516 duvBuffer->BindVBO(), duvDesc,\n+ 517 dvvBuffer->BindVBO(), dvvDesc,\n+ 518 stencilTable->GetSizesBuffer(),\n+ 519 stencilTable->GetOffsetsBuffer(),\n+ 520 stencilTable->GetIndicesBuffer(),\n+ 521 stencilTable->GetWeightsBuffer(),\n+ 522 stencilTable->GetDuWeightsBuffer(),\n+ 523 stencilTable->GetDvWeightsBuffer(),\n+ 524 stencilTable->GetDuuWeightsBuffer(),\n+ 525 stencilTable->GetDuvWeightsBuffer(),\n+ 526 stencilTable->GetDvvWeightsBuffer(),\n+ 527 /* start = */ 0,\n+ 528 /* end = */ stencilTable->GetNumStencils());\n+ 529 }\n+ 530\n+566 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n+ 567 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n+ 568 GLuint duBuffer, BufferDescriptor const &duDesc,\n+ 569 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n+ 570 GLuint sizesBuffer,\n+ 571 GLuint offsetsBuffer,\n+ 572 GLuint indicesBuffer,\n+ 573 GLuint weightsBuffer,\n+ 574 GLuint duWeightsBuffer,\n+ 575 GLuint dvWeightsBuffer,\n+ 576 int start,\n+ 577 int end) const;\n+ 578\n+632 bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n+ 633 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n+ 634 GLuint duBuffer, BufferDescriptor const &duDesc,\n+ 635 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n+ 636 GLuint duuBuffer, BufferDescriptor const &duuDesc,\n+ 637 GLuint duvBuffer, BufferDescriptor const &duvDesc,\n+ 638 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 639 GLuint sizesBuffer,\n+ 640 GLuint offsetsBuffer,\n+ 641 GLuint indicesBuffer,\n+ 642 GLuint weightsBuffer,\n+ 643 GLuint duWeightsBuffer,\n+ 644 GLuint dvWeightsBuffer,\n+ 645 GLuint duuWeightsBuffer,\n+ 646 GLuint duvWeightsBuffer,\n+ 647 GLuint dvvWeightsBuffer,\n+ 648 int start,\n+ 649 int end) const;\n+ 650\n+ 656\n+ 689 template \n+691 static bool EvalPatches(\n+ 692 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 693 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 694 int numPatchCoords,\n+ 695 PATCHCOORD_BUFFER *patchCoords,\n+ 696 PATCH_TABLE *patchTable,\n+ 697 GLComputeEvaluator const *instance,\n+ 698 void * deviceContext = NULL) {\n+ 699\n+ 700 if (instance) {\n+ 701 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 702 dstBuffer, dstDesc,\n+ 703 numPatchCoords, patchCoords,\n+ 704 patchTable);\n+ 705 } else {\n+ 706 // Create an instance on demand (slow)\n+ 707 (void)deviceContext; // unused\n+ 708 instance = Create(srcDesc, dstDesc,\n+ 709 BufferDescriptor(),\n+ 710 BufferDescriptor());\n+ 711 if (instance) {\n+ 712 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 713 dstBuffer, dstDesc,\n+ 714 numPatchCoords, patchCoords,\n+ 715 patchTable);\n+ 716 delete instance;\n+ 717 return r;\n+ 718 }\n+ 719 return false;\n+ 720 }\n+ 721 }\n+ 722\n+ 767 template \n+769 static bool EvalPatches(\n+ 770 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 771 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 772 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 773 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 774 int numPatchCoords,\n+ 775 PATCHCOORD_BUFFER *patchCoords,\n+ 776 PATCH_TABLE *patchTable,\n+ 777 GLComputeEvaluator const *instance,\n+ 778 void * deviceContext = NULL) {\n+ 779\n+ 780 if (instance) {\n+ 781 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 782 dstBuffer, dstDesc,\n+ 783 duBuffer, duDesc,\n+ 784 dvBuffer, dvDesc,\n+ 785 numPatchCoords, patchCoords,\n+ 786 patchTable);\n+ 787 } else {\n+ 788 // Create an instance on demand (slow)\n+ 789 (void)deviceContext; // unused\n+ 790 instance = Create(srcDesc, dstDesc,\n+ 791 duDesc, dvDesc);\n+ 792 if (instance) {\n+ 793 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 794 dstBuffer, dstDesc,\n+ 795 duBuffer, duDesc,\n+ 796 dvBuffer, dvDesc,\n+ 797 numPatchCoords, patchCoords,\n+ 798 patchTable);\n+ 799 delete instance;\n+ 800 return r;\n+ 801 }\n+ 802 return false;\n+ 803 }\n+ 804 }\n+ 805\n+ 868 template \n+870 static bool EvalPatches(\n+ 871 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 872 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 873 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 874 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 875 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 876 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 877 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 878 int numPatchCoords,\n+ 879 PATCHCOORD_BUFFER *patchCoords,\n+ 880 PATCH_TABLE *patchTable,\n+ 881 GLComputeEvaluator const *instance,\n+ 882 void * deviceContext = NULL) {\n+ 883\n+ 884 if (instance) {\n+ 885 return instance->EvalPatches(srcBuffer, srcDesc,\n+ 886 dstBuffer, dstDesc,\n+ 887 duBuffer, duDesc,\n+ 888 dvBuffer, dvDesc,\n+ 889 duuBuffer, duuDesc,\n+ 890 duvBuffer, duvDesc,\n+ 891 dvvBuffer, dvvDesc,\n+ 892 numPatchCoords, patchCoords,\n+ 893 patchTable);\n+ 894 } else {\n+ 895 // Create an instance on demand (slow)\n+ 896 (void)deviceContext; // unused\n+ 897 instance = Create(srcDesc, dstDesc,\n+ 898 duDesc, dvDesc,\n+ 899 duuDesc, duvDesc, dvvDesc);\n+ 900 if (instance) {\n+ 901 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n+ 902 dstBuffer, dstDesc,\n+ 903 duBuffer, duDesc,\n+ 904 dvBuffer, dvDesc,\n+ 905 duuBuffer, duuDesc,\n+ 906 duvBuffer, duvDesc,\n+ 907 dvvBuffer, dvvDesc,\n+ 908 numPatchCoords, patchCoords,\n+ 909 patchTable);\n+ 910 delete instance;\n+ 911 return r;\n+ 912 }\n+ 913 return false;\n+ 914 }\n+ 915 }\n+ 916\n 941 template \n-943 static bool EvalPatches(\n+943 bool EvalPatches(\n 944 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n 945 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 946 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 947 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 948 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 949 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 950 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 951 int numPatchCoords,\n- 952 PATCHCOORD_BUFFER *patchCoords,\n- 953 PATCH_TABLE *patchTable,\n- 954 GLXFBEvaluator const *instance,\n- 955 void * deviceContext = NULL) {\n- 956\n- 957 if (instance) {\n- 958 return instance->EvalPatches(srcBuffer, srcDesc,\n- 959 dstBuffer, dstDesc,\n- 960 duBuffer, duDesc,\n- 961 dvBuffer, dvDesc,\n- 962 duuBuffer, duuDesc,\n- 963 duvBuffer, duvDesc,\n- 964 dvvBuffer, dvvDesc,\n- 965 numPatchCoords, patchCoords,\n- 966 patchTable);\n- 967 } else {\n- 968 // Create an instance on demand (slow)\n- 969 (void)deviceContext; // unused\n- 970 instance = Create(srcDesc, dstDesc,\n- 971 duDesc, dvDesc,\n- 972 duuDesc, duvDesc, dvvDesc);\n- 973 if (instance) {\n- 974 bool r = instance->EvalPatches(srcBuffer, srcDesc,\n- 975 dstBuffer, dstDesc,\n- 976 duBuffer, duDesc,\n- 977 dvBuffer, dvDesc,\n- 978 duuBuffer, duuDesc,\n- 979 duvBuffer, duvDesc,\n- 980 dvvBuffer, dvvDesc,\n- 981 numPatchCoords, patchCoords,\n- 982 patchTable);\n- 983 delete instance;\n- 984 return r;\n- 985 }\n- 986 return false;\n- 987 }\n- 988 }\n- 989\n- 1014 template \n-1016 bool EvalPatches(\n- 1017 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1018 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1019 int numPatchCoords,\n- 1020 PATCHCOORD_BUFFER *patchCoords,\n- 1021 PATCH_TABLE *patchTable) const {\n- 1022\n- 1023 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1024 dstBuffer->BindVBO(), dstDesc,\n- 1025 0, BufferDescriptor(),\n- 1026 0, BufferDescriptor(),\n- 1027 numPatchCoords,\n- 1028 patchCoords->BindVBO(),\n- 1029 patchTable->GetPatchArrays(),\n- 1030 patchTable->GetPatchIndexTextureBuffer(),\n- 1031 patchTable->GetPatchParamTextureBuffer());\n- 1032 }\n- 1033\n- 1068 template \n-1070 bool EvalPatches(\n- 1071 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1072 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1073 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1074 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1075 int numPatchCoords,\n- 1076 PATCHCOORD_BUFFER *patchCoords,\n- 1077 PATCH_TABLE *patchTable) const {\n- 1078\n- 1079 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1080 dstBuffer->BindVBO(), dstDesc,\n- 1081 duBuffer->BindVBO(), duDesc,\n- 1082 dvBuffer->BindVBO(), dvDesc,\n- 1083 numPatchCoords,\n- 1084 patchCoords->BindVBO(),\n- 1085 patchTable->GetPatchArrays(),\n- 1086 patchTable->GetPatchIndexTextureBuffer(),\n- 1087 patchTable->GetPatchParamTextureBuffer());\n- 1088 }\n- 1089\n- 1142 template \n-1144 bool EvalPatches(\n- 1145 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1146 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1147 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1148 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1149 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1150 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1151 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1152 int numPatchCoords,\n- 1153 PATCHCOORD_BUFFER *patchCoords,\n- 1154 PATCH_TABLE *patchTable) const {\n- 1155\n- 1156 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1157 dstBuffer->BindVBO(), dstDesc,\n- 1158 duBuffer->BindVBO(), duDesc,\n- 1159 dvBuffer->BindVBO(), dvDesc,\n- 1160 duuBuffer->BindVBO(), duuDesc,\n- 1161 duvBuffer->BindVBO(), duvDesc,\n- 1162 dvvBuffer->BindVBO(), dvvDesc,\n- 1163 numPatchCoords,\n- 1164 patchCoords->BindVBO(),\n- 1165 patchTable->GetPatchArrays(),\n- 1166 patchTable->GetPatchIndexTextureBuffer(),\n- 1167 patchTable->GetPatchParamTextureBuffer());\n- 1168 }\n- 1169\n-1170 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n- 1171 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n- 1172 GLuint duBuffer, BufferDescriptor const &duDesc,\n- 1173 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n- 1174 int numPatchCoords,\n- 1175 GLuint patchCoordsBuffer,\n- 1176 const PatchArrayVector &patchArrays,\n- 1177 GLuint patchIndexBuffer,\n- 1178 GLuint patchParamsBuffer) const;\n- 1179\n-1180 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n- 1181 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n- 1182 GLuint duBuffer, BufferDescriptor const &duDesc,\n- 1183 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n- 1184 GLuint duuBuffer, BufferDescriptor const &duuDesc,\n- 1185 GLuint duvBuffer, BufferDescriptor const &duvDesc,\n- 1186 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1187 int numPatchCoords,\n- 1188 GLuint patchCoordsBuffer,\n- 1189 const PatchArrayVector &patchArrays,\n- 1190 GLuint patchIndexBuffer,\n- 1191 GLuint patchParamsBuffer) const;\n- 1192\n- 1225 template \n-1227 static bool EvalPatchesVarying(\n- 1228 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1229 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1230 int numPatchCoords,\n- 1231 PATCHCOORD_BUFFER *patchCoords,\n- 1232 PATCH_TABLE *patchTable,\n- 1233 GLXFBEvaluator const *instance,\n- 1234 void * deviceContext = NULL) {\n- 1235\n- 1236 if (instance) {\n- 1237 return instance->EvalPatchesVarying(\n- 1238 srcBuffer, srcDesc,\n- 1239 dstBuffer, dstDesc,\n- 1240 numPatchCoords, patchCoords,\n- 1241 patchTable);\n- 1242 } else {\n- 1243 // Create an instance on demand (slow)\n- 1244 (void)deviceContext; // unused\n- 1245 instance = Create(srcDesc, dstDesc,\n- 1246 BufferDescriptor(),\n- 1247 BufferDescriptor());\n- 1248 if (instance) {\n- 1249 bool r = instance->EvalPatchesVarying(\n- 1250 srcBuffer, srcDesc,\n- 1251 dstBuffer, dstDesc,\n- 1252 numPatchCoords, patchCoords,\n- 1253 patchTable);\n- 1254 delete instance;\n- 1255 return r;\n- 1256 }\n- 1257 return false;\n- 1258 }\n- 1259 }\n- 1260\n- 1285 template \n-1287 bool EvalPatchesVarying(\n- 1288 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1289 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1290 int numPatchCoords,\n- 1291 PATCHCOORD_BUFFER *patchCoords,\n- 1292 PATCH_TABLE *patchTable) const {\n- 1293\n- 1294 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1295 dstBuffer->BindVBO(), dstDesc,\n- 1296 0, BufferDescriptor(),\n- 1297 0, BufferDescriptor(),\n- 1298 numPatchCoords,\n- 1299 patchCoords->BindVBO(),\n- 1300 patchTable->GetVaryingPatchArrays(),\n- 1301 patchTable->GetVaryingPatchIndexTextureBuffer(),\n- 1302 patchTable->GetPatchParamTextureBuffer());\n- 1303 }\n- 1304\n- 1349 template \n-1351 static bool EvalPatchesVarying(\n- 1352 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1353 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1354 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1355 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1356 int numPatchCoords,\n- 1357 PATCHCOORD_BUFFER *patchCoords,\n- 1358 PATCH_TABLE *patchTable,\n- 1359 GLXFBEvaluator const *instance,\n- 1360 void * deviceContext = NULL) {\n- 1361\n- 1362 if (instance) {\n- 1363 return instance->EvalPatchesVarying(\n- 1364 srcBuffer, srcDesc,\n- 1365 dstBuffer, dstDesc,\n- 1366 duBuffer, duDesc,\n- 1367 dvBuffer, dvDesc,\n- 1368 numPatchCoords, patchCoords,\n- 1369 patchTable);\n- 1370 } else {\n- 1371 // Create an instance on demand (slow)\n- 1372 (void)deviceContext; // unused\n- 1373 instance = Create(srcDesc, dstDesc,\n- 1374 duDesc, dvDesc);\n- 1375 if (instance) {\n- 1376 bool r = instance->EvalPatchesVarying(\n- 1377 srcBuffer, srcDesc,\n- 1378 dstBuffer, dstDesc,\n- 1379 duBuffer, duDesc,\n- 1380 dvBuffer, dvDesc,\n- 1381 numPatchCoords, patchCoords,\n- 1382 patchTable);\n- 1383 delete instance;\n- 1384 return r;\n- 1385 }\n- 1386 return false;\n- 1387 }\n- 1388 }\n- 1389\n- 1426 template \n-1428 bool EvalPatchesVarying(\n- 1429 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1430 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1431 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1432 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1433 int numPatchCoords,\n- 1434 PATCHCOORD_BUFFER *patchCoords,\n- 1435 PATCH_TABLE *patchTable) const {\n- 1436\n- 1437 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1438 dstBuffer->BindVBO(), dstDesc,\n- 1439 duBuffer->BindVBO(), duDesc,\n- 1440 dvBuffer->BindVBO(), dvDesc,\n- 1441 numPatchCoords,\n- 1442 patchCoords->BindVBO(),\n- 1443 patchTable->GetVaryingPatchArrays(),\n- 1444 patchTable->GetVaryingPatchIndexTextureBuffer(),\n- 1445 patchTable->GetPatchParamTextureBuffer());\n- 1446 }\n- 1447\n- 1510 template \n-1512 static bool EvalPatchesVarying(\n- 1513 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1514 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1515 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1516 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1517 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1518 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1519 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1520 int numPatchCoords,\n- 1521 PATCHCOORD_BUFFER *patchCoords,\n- 1522 PATCH_TABLE *patchTable,\n- 1523 GLXFBEvaluator const *instance,\n- 1524 void * deviceContext = NULL) {\n- 1525\n- 1526 if (instance) {\n- 1527 return instance->EvalPatchesVarying(\n- 1528 srcBuffer, srcDesc,\n- 1529 dstBuffer, dstDesc,\n- 1530 duBuffer, duDesc,\n- 1531 dvBuffer, dvDesc,\n- 1532 duuBuffer, duuDesc,\n- 1533 duvBuffer, duvDesc,\n- 1534 dvvBuffer, dvvDesc,\n- 1535 numPatchCoords, patchCoords,\n- 1536 patchTable);\n- 1537 } else {\n- 1538 // Create an instance on demand (slow)\n- 1539 (void)deviceContext; // unused\n- 1540 instance = Create(srcDesc, dstDesc,\n- 1541 duDesc, dvDesc,\n- 1542 duuDesc, duvDesc, dvvDesc);\n- 1543 if (instance) {\n- 1544 bool r = instance->EvalPatchesVarying(\n- 1545 srcBuffer, srcDesc,\n- 1546 dstBuffer, dstDesc,\n- 1547 duBuffer, duDesc,\n- 1548 dvBuffer, dvDesc,\n- 1549 duuBuffer, duuDesc,\n- 1550 duvBuffer, duvDesc,\n- 1551 dvvBuffer, dvvDesc,\n- 1552 numPatchCoords, patchCoords,\n- 1553 patchTable);\n- 1554 delete instance;\n- 1555 return r;\n- 1556 }\n- 1557 return false;\n- 1558 }\n- 1559 }\n- 1560\n- 1615 template \n-1617 bool EvalPatchesVarying(\n- 1618 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1619 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1620 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1621 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1622 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1623 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1624 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1625 int numPatchCoords,\n- 1626 PATCHCOORD_BUFFER *patchCoords,\n- 1627 PATCH_TABLE *patchTable) const {\n- 1628\n- 1629 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1630 dstBuffer->BindVBO(), dstDesc,\n- 1631 duBuffer->BindVBO(), duDesc,\n- 1632 dvBuffer->BindVBO(), dvDesc,\n- 1633 duuBuffer->BindVBO(), duuDesc,\n- 1634 duvBuffer->BindVBO(), duvDesc,\n- 1635 dvvBuffer->BindVBO(), dvvDesc,\n- 1636 numPatchCoords,\n- 1637 patchCoords->BindVBO(),\n- 1638 patchTable->GetVaryingPatchArrays(),\n- 1639 patchTable->GetVaryingPatchIndexTextureBuffer(),\n- 1640 patchTable->GetPatchParamTextureBuffer());\n- 1641 }\n- 1642\n- 1677 template \n-1679 static bool EvalPatchesFaceVarying(\n- 1680 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1681 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1682 int numPatchCoords,\n- 1683 PATCHCOORD_BUFFER *patchCoords,\n- 1684 PATCH_TABLE *patchTable,\n- 1685 int fvarChannel,\n- 1686 GLXFBEvaluator const *instance,\n- 1687 void * deviceContext = NULL) {\n- 1688\n- 1689 if (instance) {\n- 1690 return instance->EvalPatchesFaceVarying(\n- 1691 srcBuffer, srcDesc,\n- 1692 dstBuffer, dstDesc,\n- 1693 numPatchCoords, patchCoords,\n- 1694 patchTable, fvarChannel);\n- 1695 } else {\n- 1696 // Create an instance on demand (slow)\n- 1697 (void)deviceContext; // unused\n- 1698 instance = Create(srcDesc, dstDesc,\n- 1699 BufferDescriptor(),\n- 1700 BufferDescriptor());\n- 1701 if (instance) {\n- 1702 bool r = instance->EvalPatchesFaceVarying(\n- 1703 srcBuffer, srcDesc,\n- 1704 dstBuffer, dstDesc,\n- 1705 numPatchCoords, patchCoords,\n- 1706 patchTable, fvarChannel);\n- 1707 delete instance;\n- 1708 return r;\n- 1709 }\n- 1710 return false;\n- 1711 }\n- 1712 }\n- 1713\n- 1740 template \n-1742 bool EvalPatchesFaceVarying(\n- 1743 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1744 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1745 int numPatchCoords,\n- 1746 PATCHCOORD_BUFFER *patchCoords,\n- 1747 PATCH_TABLE *patchTable,\n- 1748 int fvarChannel = 0) const {\n- 1749\n- 1750 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1751 dstBuffer->BindVBO(), dstDesc,\n- 1752 0, BufferDescriptor(),\n- 1753 0, BufferDescriptor(),\n- 1754 numPatchCoords,\n- 1755 patchCoords->BindVBO(),\n- 1756 patchTable->GetFVarPatchArrays(fvarChannel),\n- 1757 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),\n- 1758 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));\n- 1759 }\n- 1760\n- 1807 template \n-1809 static bool EvalPatchesFaceVarying(\n- 1810 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1811 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1812 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1813 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1814 int numPatchCoords,\n- 1815 PATCHCOORD_BUFFER *patchCoords,\n- 1816 PATCH_TABLE *patchTable,\n- 1817 int fvarChannel,\n- 1818 GLXFBEvaluator const *instance,\n- 1819 void * deviceContext = NULL) {\n- 1820\n- 1821 if (instance) {\n- 1822 return instance->EvalPatchesFaceVarying(\n- 1823 srcBuffer, srcDesc,\n- 1824 dstBuffer, dstDesc,\n- 1825 duBuffer, duDesc,\n- 1826 dvBuffer, dvDesc,\n- 1827 numPatchCoords, patchCoords,\n- 1828 patchTable, fvarChannel);\n- 1829 } else {\n- 1830 // Create an instance on demand (slow)\n- 1831 (void)deviceContext; // unused\n- 1832 instance = Create(srcDesc, dstDesc,\n- 1833 duDesc, dvDesc);\n- 1834 if (instance) {\n- 1835 bool r = instance->EvalPatchesFaceVarying(\n- 1836 srcBuffer, srcDesc,\n- 1837 dstBuffer, dstDesc,\n- 1838 duBuffer, duDesc,\n- 1839 dvBuffer, dvDesc,\n- 1840 numPatchCoords, patchCoords,\n- 1841 patchTable, fvarChannel);\n- 1842 delete instance;\n- 1843 return r;\n- 1844 }\n- 1845 return false;\n- 1846 }\n- 1847 }\n- 1848\n- 1887 template \n-1889 bool EvalPatchesFaceVarying(\n- 1890 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1891 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1892 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1893 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1894 int numPatchCoords,\n- 1895 PATCHCOORD_BUFFER *patchCoords,\n- 1896 PATCH_TABLE *patchTable,\n- 1897 int fvarChannel = 0) const {\n- 1898\n- 1899 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 1900 dstBuffer->BindVBO(), dstDesc,\n- 1901 duBuffer->BindVBO(), duDesc,\n- 1902 dvBuffer->BindVBO(), dvDesc,\n- 1903 numPatchCoords,\n- 1904 patchCoords->BindVBO(),\n- 1905 patchTable->GetFVarPatchArrays(fvarChannel),\n- 1906 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),\n- 1907 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));\n- 1908 }\n- 1909\n- 1974 template \n-1976 static bool EvalPatchesFaceVarying(\n- 1977 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1978 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1979 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1980 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1981 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1982 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1983 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1984 int numPatchCoords,\n- 1985 PATCHCOORD_BUFFER *patchCoords,\n- 1986 PATCH_TABLE *patchTable,\n- 1987 int fvarChannel,\n- 1988 GLXFBEvaluator const *instance,\n- 1989 void * deviceContext = NULL) {\n- 1990\n- 1991 if (instance) {\n- 1992 return instance->EvalPatchesFaceVarying(\n- 1993 srcBuffer, srcDesc,\n- 1994 dstBuffer, dstDesc,\n- 1995 duBuffer, duDesc,\n- 1996 dvBuffer, dvDesc,\n- 1997 duuBuffer, duuDesc,\n- 1998 duvBuffer, duvDesc,\n- 1999 dvvBuffer, dvvDesc,\n- 2000 numPatchCoords, patchCoords,\n- 2001 patchTable, fvarChannel);\n- 2002 } else {\n- 2003 // Create an instance on demand (slow)\n- 2004 (void)deviceContext; // unused\n- 2005 instance = Create(srcDesc, dstDesc,\n- 2006 duDesc, dvDesc,\n- 2007 duuDesc, duvDesc, dvvDesc);\n- 2008 if (instance) {\n- 2009 bool r = instance->EvalPatchesFaceVarying(\n- 2010 srcBuffer, srcDesc,\n- 2011 dstBuffer, dstDesc,\n- 2012 duBuffer, duDesc,\n- 2013 dvBuffer, dvDesc,\n- 2014 duuBuffer, duuDesc,\n- 2015 duvBuffer, duvDesc,\n- 2016 dvvBuffer, dvvDesc,\n- 2017 numPatchCoords, patchCoords,\n- 2018 patchTable, fvarChannel);\n- 2019 delete instance;\n- 2020 return r;\n- 2021 }\n- 2022 return false;\n- 2023 }\n- 2024 }\n- 2025\n- 2082 template \n-2084 bool EvalPatchesFaceVarying(\n- 2085 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 2086 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 2087 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 2088 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 2089 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 2090 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 2091 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 2092 int numPatchCoords,\n- 2093 PATCHCOORD_BUFFER *patchCoords,\n- 2094 PATCH_TABLE *patchTable,\n- 2095 int fvarChannel = 0) const {\n- 2096\n- 2097 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n- 2098 dstBuffer->BindVBO(), dstDesc,\n- 2099 duBuffer->BindVBO(), duDesc,\n- 2100 dvBuffer->BindVBO(), dvDesc,\n- 2101 duuBuffer->BindVBO(), duuDesc,\n- 2102 duvBuffer->BindVBO(), duvDesc,\n- 2103 dvvBuffer->BindVBO(), dvvDesc,\n- 2104 numPatchCoords,\n- 2105 patchCoords->BindVBO(),\n- 2106 patchTable->GetFVarPatchArrays(fvarChannel),\n- 2107 patchTable->GetFVarPatchIndexTextureBuffer(fvarChannel),\n- 2108 patchTable->GetFVarPatchParamTextureBuffer(fvarChannel));\n- 2109 }\n- 2110\n- 2116\n-2119 bool Compile(BufferDescriptor const &srcDesc,\n- 2120 BufferDescriptor const &dstDesc,\n- 2121 BufferDescriptor const &duDesc = BufferDescriptor(),\n- 2122 BufferDescriptor const &dvDesc = BufferDescriptor(),\n- 2123 BufferDescriptor const &duuDesc = BufferDescriptor(),\n- 2124 BufferDescriptor const &duvDesc = BufferDescriptor(),\n- 2125 BufferDescriptor const &dvvDesc = BufferDescriptor());\n- 2126\n-2128 static void Synchronize(void *kernel);\n- 2129\n- 2130private:\n- 2131 GLuint _srcBufferTexture;\n- 2132 GLuint _patchArraysUBO;\n- 2133 bool _interleavedDerivativeBuffers;\n- 2134\n- 2135 struct _StencilKernel {\n- 2136 _StencilKernel();\n- 2137 ~_StencilKernel();\n- 2138 bool Compile(BufferDescriptor const &srcDesc,\n- 2139 BufferDescriptor const &dstDesc,\n- 2140 BufferDescriptor const &duDesc,\n- 2141 BufferDescriptor const &dvDesc,\n- 2142 BufferDescriptor const &duuDesc,\n- 2143 BufferDescriptor const &duvDesc,\n- 2144 BufferDescriptor const &dvvDesc,\n- 2145 bool interleavedDerivativeBuffers);\n- 2146 GLuint program;\n- 2147 GLint uniformSrcBufferTexture;\n- 2148 GLint uniformSrcOffset; // src buffer offset (in elements)\n- 2149\n- 2150 GLint uniformSizesTexture;\n- 2151 GLint uniformOffsetsTexture;\n- 2152 GLint uniformIndicesTexture;\n- 2153 GLint uniformWeightsTexture;\n- 2154 GLint uniformDuWeightsTexture;\n- 2155 GLint uniformDvWeightsTexture;\n- 2156 GLint uniformDuuWeightsTexture;\n- 2157 GLint uniformDuvWeightsTexture;\n- 2158 GLint uniformDvvWeightsTexture;\n- 2159 GLint uniformStart; // range\n- 2160 GLint uniformEnd;\n- 2161 } _stencilKernel;\n- 2162\n- 2163 struct _PatchKernel {\n- 2164 _PatchKernel();\n- 2165 ~_PatchKernel();\n- 2166 bool Compile(BufferDescriptor const &srcDesc,\n- 2167 BufferDescriptor const &dstDesc,\n- 2168 BufferDescriptor const &duDesc,\n- 2169 BufferDescriptor const &dvDesc,\n- 2170 BufferDescriptor const &duuDesc,\n- 2171 BufferDescriptor const &duvDesc,\n- 2172 BufferDescriptor const &dvvDesc,\n- 2173 bool interleavedDerivativeBuffers);\n- 2174 GLuint program;\n- 2175 GLint uniformSrcBufferTexture;\n- 2176 GLint uniformSrcOffset; // src buffer offset (in elements)\n- 2177\n- 2178 GLint uniformPatchArraysUBOBinding;\n- 2179 GLint uniformPatchParamTexture;\n- 2180 GLint uniformPatchIndexTexture;\n- 2181 } _patchKernel;\n- 2182\n- 2183};\n- 2184\n- 2185} // end namespace Osd\n- 2186\n- 2187} // end namespace OPENSUBDIV_VERSION\n- 2188using namespace OPENSUBDIV_VERSION;\n- 2189\n- 2190} // end namespace OpenSubdiv\n- 2191\n- 2192\n- 2193#endif // OPENSUBDIV3_OSD_GL_XFB_EVALUATOR_H\n+ 946 int numPatchCoords,\n+ 947 PATCHCOORD_BUFFER *patchCoords,\n+ 948 PATCH_TABLE *patchTable) const {\n+ 949\n+ 950 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 951 dstBuffer->BindVBO(), dstDesc,\n+ 952 0, BufferDescriptor(),\n+ 953 0, BufferDescriptor(),\n+ 954 numPatchCoords,\n+ 955 patchCoords->BindVBO(),\n+ 956 patchTable->GetPatchArrays(),\n+ 957 patchTable->GetPatchIndexBuffer(),\n+ 958 patchTable->GetPatchParamBuffer());\n+ 959 }\n+ 960\n+ 995 template \n+997 bool EvalPatches(\n+ 998 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 999 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1000 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1001 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1002 int numPatchCoords,\n+ 1003 PATCHCOORD_BUFFER *patchCoords,\n+ 1004 PATCH_TABLE *patchTable) const {\n+ 1005\n+ 1006 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1007 dstBuffer->BindVBO(), dstDesc,\n+ 1008 duBuffer->BindVBO(), duDesc,\n+ 1009 dvBuffer->BindVBO(), dvDesc,\n+ 1010 numPatchCoords,\n+ 1011 patchCoords->BindVBO(),\n+ 1012 patchTable->GetPatchArrays(),\n+ 1013 patchTable->GetPatchIndexBuffer(),\n+ 1014 patchTable->GetPatchParamBuffer());\n+ 1015 }\n+ 1016\n+ 1069 template \n+1071 bool EvalPatches(\n+ 1072 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1073 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1074 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1075 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1076 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1077 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1078 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1079 int numPatchCoords,\n+ 1080 PATCHCOORD_BUFFER *patchCoords,\n+ 1081 PATCH_TABLE *patchTable) const {\n+ 1082\n+ 1083 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1084 dstBuffer->BindVBO(), dstDesc,\n+ 1085 duBuffer->BindVBO(), duDesc,\n+ 1086 dvBuffer->BindVBO(), dvDesc,\n+ 1087 duuBuffer->BindVBO(), duuDesc,\n+ 1088 duvBuffer->BindVBO(), duvDesc,\n+ 1089 dvvBuffer->BindVBO(), dvvDesc,\n+ 1090 numPatchCoords,\n+ 1091 patchCoords->BindVBO(),\n+ 1092 patchTable->GetPatchArrays(),\n+ 1093 patchTable->GetPatchIndexBuffer(),\n+ 1094 patchTable->GetPatchParamBuffer());\n+ 1095 }\n+ 1096\n+1097 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n+ 1098 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n+ 1099 GLuint duBuffer, BufferDescriptor const &duDesc,\n+ 1100 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n+ 1101 int numPatchCoords,\n+ 1102 GLuint patchCoordsBuffer,\n+ 1103 const PatchArrayVector &patchArrays,\n+ 1104 GLuint patchIndexBuffer,\n+ 1105 GLuint patchParamsBuffer) const;\n+ 1106\n+1107 bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc,\n+ 1108 GLuint dstBuffer, BufferDescriptor const &dstDesc,\n+ 1109 GLuint duBuffer, BufferDescriptor const &duDesc,\n+ 1110 GLuint dvBuffer, BufferDescriptor const &dvDesc,\n+ 1111 GLuint duuBuffer, BufferDescriptor const &duuDesc,\n+ 1112 GLuint duvBuffer, BufferDescriptor const &duvDesc,\n+ 1113 GLuint dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1114 int numPatchCoords,\n+ 1115 GLuint patchCoordsBuffer,\n+ 1116 const PatchArrayVector &patchArrays,\n+ 1117 GLuint patchIndexBuffer,\n+ 1118 GLuint patchParamsBuffer) const;\n+ 1119\n+ 1152 template \n+1154 static bool EvalPatchesVarying(\n+ 1155 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1156 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1157 int numPatchCoords,\n+ 1158 PATCHCOORD_BUFFER *patchCoords,\n+ 1159 PATCH_TABLE *patchTable,\n+ 1160 GLComputeEvaluator const *instance,\n+ 1161 void * deviceContext = NULL) {\n+ 1162\n+ 1163 if (instance) {\n+ 1164 return instance->EvalPatchesVarying(\n+ 1165 srcBuffer, srcDesc,\n+ 1166 dstBuffer, dstDesc,\n+ 1167 numPatchCoords, patchCoords,\n+ 1168 patchTable);\n+ 1169 } else {\n+ 1170 // Create an instance on demand (slow)\n+ 1171 (void)deviceContext; // unused\n+ 1172 instance = Create(srcDesc, dstDesc,\n+ 1173 BufferDescriptor(),\n+ 1174 BufferDescriptor());\n+ 1175 if (instance) {\n+ 1176 bool r = instance->EvalPatchesVarying(\n+ 1177 srcBuffer, srcDesc,\n+ 1178 dstBuffer, dstDesc,\n+ 1179 numPatchCoords, patchCoords,\n+ 1180 patchTable);\n+ 1181 delete instance;\n+ 1182 return r;\n+ 1183 }\n+ 1184 return false;\n+ 1185 }\n+ 1186 }\n+ 1187\n+ 1212 template \n+1214 bool EvalPatchesVarying(\n+ 1215 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1216 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1217 int numPatchCoords,\n+ 1218 PATCHCOORD_BUFFER *patchCoords,\n+ 1219 PATCH_TABLE *patchTable) const {\n+ 1220\n+ 1221 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1222 dstBuffer->BindVBO(), dstDesc,\n+ 1223 0, BufferDescriptor(),\n+ 1224 0, BufferDescriptor(),\n+ 1225 numPatchCoords,\n+ 1226 patchCoords->BindVBO(),\n+ 1227 patchTable->GetVaryingPatchArrays(),\n+ 1228 patchTable->GetVaryingPatchIndexBuffer(),\n+ 1229 patchTable->GetPatchParamBuffer());\n+ 1230 }\n+ 1231\n+ 1276 template \n+1278 static bool EvalPatchesVarying(\n+ 1279 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1280 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1281 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1282 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1283 int numPatchCoords,\n+ 1284 PATCHCOORD_BUFFER *patchCoords,\n+ 1285 PATCH_TABLE *patchTable,\n+ 1286 GLComputeEvaluator const *instance,\n+ 1287 void * deviceContext = NULL) {\n+ 1288\n+ 1289 if (instance) {\n+ 1290 return instance->EvalPatchesVarying(\n+ 1291 srcBuffer, srcDesc,\n+ 1292 dstBuffer, dstDesc,\n+ 1293 duBuffer, duDesc,\n+ 1294 dvBuffer, dvDesc,\n+ 1295 numPatchCoords, patchCoords,\n+ 1296 patchTable);\n+ 1297 } else {\n+ 1298 // Create an instance on demand (slow)\n+ 1299 (void)deviceContext; // unused\n+ 1300 instance = Create(srcDesc, dstDesc,\n+ 1301 duDesc, dvDesc);\n+ 1302 if (instance) {\n+ 1303 bool r = instance->EvalPatchesVarying(\n+ 1304 srcBuffer, srcDesc,\n+ 1305 dstBuffer, dstDesc,\n+ 1306 duBuffer, duDesc,\n+ 1307 dvBuffer, dvDesc,\n+ 1308 numPatchCoords, patchCoords,\n+ 1309 patchTable);\n+ 1310 delete instance;\n+ 1311 return r;\n+ 1312 }\n+ 1313 return false;\n+ 1314 }\n+ 1315 }\n+ 1316\n+ 1353 template \n+1355 bool EvalPatchesVarying(\n+ 1356 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1357 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1358 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1359 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1360 int numPatchCoords,\n+ 1361 PATCHCOORD_BUFFER *patchCoords,\n+ 1362 PATCH_TABLE *patchTable) const {\n+ 1363\n+ 1364 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1365 dstBuffer->BindVBO(), dstDesc,\n+ 1366 duBuffer->BindVBO(), duDesc,\n+ 1367 dvBuffer->BindVBO(), dvDesc,\n+ 1368 numPatchCoords,\n+ 1369 patchCoords->BindVBO(),\n+ 1370 patchTable->GetVaryingPatchArrays(),\n+ 1371 patchTable->GetVaryingPatchIndexBuffer(),\n+ 1372 patchTable->GetPatchParamBuffer());\n+ 1373 }\n+ 1374\n+ 1437 template \n+1439 static bool EvalPatchesVarying(\n+ 1440 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1441 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1442 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1443 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1444 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1445 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1446 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1447 int numPatchCoords,\n+ 1448 PATCHCOORD_BUFFER *patchCoords,\n+ 1449 PATCH_TABLE *patchTable,\n+ 1450 GLComputeEvaluator const *instance,\n+ 1451 void * deviceContext = NULL) {\n+ 1452\n+ 1453 if (instance) {\n+ 1454 return instance->EvalPatchesVarying(\n+ 1455 srcBuffer, srcDesc,\n+ 1456 dstBuffer, dstDesc,\n+ 1457 duBuffer, duDesc,\n+ 1458 dvBuffer, dvDesc,\n+ 1459 duuBuffer, duuDesc,\n+ 1460 duvBuffer, duvDesc,\n+ 1461 dvvBuffer, dvvDesc,\n+ 1462 numPatchCoords, patchCoords,\n+ 1463 patchTable);\n+ 1464 } else {\n+ 1465 // Create an instance on demand (slow)\n+ 1466 (void)deviceContext; // unused\n+ 1467 instance = Create(srcDesc, dstDesc,\n+ 1468 duDesc, dvDesc,\n+ 1469 duuDesc, duvDesc, dvvDesc);\n+ 1470 if (instance) {\n+ 1471 bool r = instance->EvalPatchesVarying(\n+ 1472 srcBuffer, srcDesc,\n+ 1473 dstBuffer, dstDesc,\n+ 1474 duBuffer, duDesc,\n+ 1475 dvBuffer, dvDesc,\n+ 1476 duuBuffer, duuDesc,\n+ 1477 duvBuffer, duvDesc,\n+ 1478 dvvBuffer, dvvDesc,\n+ 1479 numPatchCoords, patchCoords,\n+ 1480 patchTable);\n+ 1481 delete instance;\n+ 1482 return r;\n+ 1483 }\n+ 1484 return false;\n+ 1485 }\n+ 1486 }\n+ 1487\n+ 1542 template \n+1544 bool EvalPatchesVarying(\n+ 1545 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1546 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1547 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1548 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1549 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1550 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1551 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1552 int numPatchCoords,\n+ 1553 PATCHCOORD_BUFFER *patchCoords,\n+ 1554 PATCH_TABLE *patchTable) const {\n+ 1555\n+ 1556 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1557 dstBuffer->BindVBO(), dstDesc,\n+ 1558 duBuffer->BindVBO(), duDesc,\n+ 1559 dvBuffer->BindVBO(), dvDesc,\n+ 1560 duuBuffer->BindVBO(), duuDesc,\n+ 1561 duvBuffer->BindVBO(), duvDesc,\n+ 1562 dvvBuffer->BindVBO(), dvvDesc,\n+ 1563 numPatchCoords,\n+ 1564 patchCoords->BindVBO(),\n+ 1565 patchTable->GetVaryingPatchArrays(),\n+ 1566 patchTable->GetVaryingPatchIndexBuffer(),\n+ 1567 patchTable->GetPatchParamBuffer());\n+ 1568 }\n+ 1569\n+ 1604 template \n+1606 static bool EvalPatchesFaceVarying(\n+ 1607 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1608 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1609 int numPatchCoords,\n+ 1610 PATCHCOORD_BUFFER *patchCoords,\n+ 1611 PATCH_TABLE *patchTable,\n+ 1612 int fvarChannel,\n+ 1613 GLComputeEvaluator const *instance,\n+ 1614 void * deviceContext = NULL) {\n+ 1615\n+ 1616 if (instance) {\n+ 1617 return instance->EvalPatchesFaceVarying(\n+ 1618 srcBuffer, srcDesc,\n+ 1619 dstBuffer, dstDesc,\n+ 1620 numPatchCoords, patchCoords,\n+ 1621 patchTable, fvarChannel);\n+ 1622 } else {\n+ 1623 // Create an instance on demand (slow)\n+ 1624 (void)deviceContext; // unused\n+ 1625 instance = Create(srcDesc, dstDesc,\n+ 1626 BufferDescriptor(),\n+ 1627 BufferDescriptor());\n+ 1628 if (instance) {\n+ 1629 bool r = instance->EvalPatchesFaceVarying(\n+ 1630 srcBuffer, srcDesc,\n+ 1631 dstBuffer, dstDesc,\n+ 1632 numPatchCoords, patchCoords,\n+ 1633 patchTable, fvarChannel);\n+ 1634 delete instance;\n+ 1635 return r;\n+ 1636 }\n+ 1637 return false;\n+ 1638 }\n+ 1639 }\n+ 1640\n+ 1667 template \n+1669 bool EvalPatchesFaceVarying(\n+ 1670 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1671 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1672 int numPatchCoords,\n+ 1673 PATCHCOORD_BUFFER *patchCoords,\n+ 1674 PATCH_TABLE *patchTable,\n+ 1675 int fvarChannel = 0) const {\n+ 1676\n+ 1677 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1678 dstBuffer->BindVBO(), dstDesc,\n+ 1679 0, BufferDescriptor(),\n+ 1680 0, BufferDescriptor(),\n+ 1681 numPatchCoords,\n+ 1682 patchCoords->BindVBO(),\n+ 1683 patchTable->GetFVarPatchArrays(fvarChannel),\n+ 1684 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1685 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 1686 }\n+ 1687\n+ 1734 template \n+1736 static bool EvalPatchesFaceVarying(\n+ 1737 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1738 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1739 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1740 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1741 int numPatchCoords,\n+ 1742 PATCHCOORD_BUFFER *patchCoords,\n+ 1743 PATCH_TABLE *patchTable,\n+ 1744 int fvarChannel,\n+ 1745 GLComputeEvaluator const *instance,\n+ 1746 void * deviceContext = NULL) {\n+ 1747\n+ 1748 if (instance) {\n+ 1749 return instance->EvalPatchesFaceVarying(\n+ 1750 srcBuffer, srcDesc,\n+ 1751 dstBuffer, dstDesc,\n+ 1752 duBuffer, duDesc,\n+ 1753 dvBuffer, dvDesc,\n+ 1754 numPatchCoords, patchCoords,\n+ 1755 patchTable, fvarChannel);\n+ 1756 } else {\n+ 1757 // Create an instance on demand (slow)\n+ 1758 (void)deviceContext; // unused\n+ 1759 instance = Create(srcDesc, dstDesc,\n+ 1760 duDesc, dvDesc);\n+ 1761 if (instance) {\n+ 1762 bool r = instance->EvalPatchesFaceVarying(\n+ 1763 srcBuffer, srcDesc,\n+ 1764 dstBuffer, dstDesc,\n+ 1765 duBuffer, duDesc,\n+ 1766 dvBuffer, dvDesc,\n+ 1767 numPatchCoords, patchCoords,\n+ 1768 patchTable, fvarChannel);\n+ 1769 delete instance;\n+ 1770 return r;\n+ 1771 }\n+ 1772 return false;\n+ 1773 }\n+ 1774 }\n+ 1775\n+ 1814 template \n+1816 bool EvalPatchesFaceVarying(\n+ 1817 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1818 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1819 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1820 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1821 int numPatchCoords,\n+ 1822 PATCHCOORD_BUFFER *patchCoords,\n+ 1823 PATCH_TABLE *patchTable,\n+ 1824 int fvarChannel = 0) const {\n+ 1825\n+ 1826 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 1827 dstBuffer->BindVBO(), dstDesc,\n+ 1828 duBuffer->BindVBO(), duDesc,\n+ 1829 dvBuffer->BindVBO(), dvDesc,\n+ 1830 numPatchCoords,\n+ 1831 patchCoords->BindVBO(),\n+ 1832 patchTable->GetFVarPatchArrays(fvarChannel),\n+ 1833 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1834 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 1835 }\n+ 1836\n+ 1901 template \n+1903 static bool EvalPatchesFaceVarying(\n+ 1904 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1905 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1906 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1907 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1908 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1909 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1910 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1911 int numPatchCoords,\n+ 1912 PATCHCOORD_BUFFER *patchCoords,\n+ 1913 PATCH_TABLE *patchTable,\n+ 1914 int fvarChannel,\n+ 1915 GLComputeEvaluator const *instance,\n+ 1916 void * deviceContext = NULL) {\n+ 1917\n+ 1918 if (instance) {\n+ 1919 return instance->EvalPatchesFaceVarying(\n+ 1920 srcBuffer, srcDesc,\n+ 1921 dstBuffer, dstDesc,\n+ 1922 duBuffer, duDesc,\n+ 1923 dvBuffer, dvDesc,\n+ 1924 duuBuffer, duuDesc,\n+ 1925 duvBuffer, duvDesc,\n+ 1926 dvvBuffer, dvvDesc,\n+ 1927 numPatchCoords, patchCoords,\n+ 1928 patchTable, fvarChannel);\n+ 1929 } else {\n+ 1930 // Create an instance on demand (slow)\n+ 1931 (void)deviceContext; // unused\n+ 1932 instance = Create(srcDesc, dstDesc,\n+ 1933 duDesc, dvDesc,\n+ 1934 duuDesc, duvDesc, dvvDesc);\n+ 1935 if (instance) {\n+ 1936 bool r = instance->EvalPatchesFaceVarying(\n+ 1937 srcBuffer, srcDesc,\n+ 1938 dstBuffer, dstDesc,\n+ 1939 duBuffer, duDesc,\n+ 1940 dvBuffer, dvDesc,\n+ 1941 duuBuffer, duuDesc,\n+ 1942 duvBuffer, duvDesc,\n+ 1943 dvvBuffer, dvvDesc,\n+ 1944 numPatchCoords, patchCoords,\n+ 1945 patchTable, fvarChannel);\n+ 1946 delete instance;\n+ 1947 return r;\n+ 1948 }\n+ 1949 return false;\n+ 1950 }\n+ 1951 }\n+ 1952\n+ 2009 template \n+2011 bool EvalPatchesFaceVarying(\n+ 2012 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 2013 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 2014 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 2015 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 2016 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 2017 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 2018 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 2019 int numPatchCoords,\n+ 2020 PATCHCOORD_BUFFER *patchCoords,\n+ 2021 PATCH_TABLE *patchTable,\n+ 2022 int fvarChannel = 0) const {\n+ 2023\n+ 2024 return EvalPatches(srcBuffer->BindVBO(), srcDesc,\n+ 2025 dstBuffer->BindVBO(), dstDesc,\n+ 2026 duBuffer->BindVBO(), duDesc,\n+ 2027 dvBuffer->BindVBO(), dvDesc,\n+ 2028 duuBuffer->BindVBO(), duuDesc,\n+ 2029 duvBuffer->BindVBO(), duvDesc,\n+ 2030 dvvBuffer->BindVBO(), dvvDesc,\n+ 2031 numPatchCoords,\n+ 2032 patchCoords->BindVBO(),\n+ 2033 patchTable->GetFVarPatchArrays(fvarChannel),\n+ 2034 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 2035 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 2036 }\n+ 2037\n+ 2043\n+2046 bool Compile(BufferDescriptor const &srcDesc,\n+ 2047 BufferDescriptor const &dstDesc,\n+ 2048 BufferDescriptor const &duDesc = BufferDescriptor(),\n+ 2049 BufferDescriptor const &dvDesc = BufferDescriptor(),\n+ 2050 BufferDescriptor const &duuDesc = BufferDescriptor(),\n+ 2051 BufferDescriptor const &duvDesc = BufferDescriptor(),\n+ 2052 BufferDescriptor const &dvvDesc = BufferDescriptor());\n+ 2053\n+2055 static void Synchronize(void *deviceContext);\n+ 2056\n+ 2057private:\n+ 2058 struct _StencilKernel {\n+ 2059 _StencilKernel();\n+ 2060 ~_StencilKernel();\n+ 2061 bool Compile(BufferDescriptor const &srcDesc,\n+ 2062 BufferDescriptor const &dstDesc,\n+ 2063 BufferDescriptor const &duDesc,\n+ 2064 BufferDescriptor const &dvDesc,\n+ 2065 BufferDescriptor const &duuDesc,\n+ 2066 BufferDescriptor const &duvDesc,\n+ 2067 BufferDescriptor const &dvvDesc,\n+ 2068 int workGroupSize);\n+ 2069 GLuint program;\n+ 2070 GLuint uniformStart;\n+ 2071 GLuint uniformEnd;\n+ 2072 GLuint uniformSrcOffset;\n+ 2073 GLuint uniformDstOffset;\n+ 2074 GLuint uniformDuDesc;\n+ 2075 GLuint uniformDvDesc;\n+ 2076 GLuint uniformDuuDesc;\n+ 2077 GLuint uniformDuvDesc;\n+ 2078 GLuint uniformDvvDesc;\n+ 2079 } _stencilKernel;\n+ 2080\n+ 2081 struct _PatchKernel {\n+ 2082 _PatchKernel();\n+ 2083 ~_PatchKernel();\n+ 2084 bool Compile(BufferDescriptor const &srcDesc,\n+ 2085 BufferDescriptor const &dstDesc,\n+ 2086 BufferDescriptor const &duDesc,\n+ 2087 BufferDescriptor const &dvDesc,\n+ 2088 BufferDescriptor const &duuDesc,\n+ 2089 BufferDescriptor const &duvDesc,\n+ 2090 BufferDescriptor const &dvvDesc,\n+ 2091 int workGroupSize);\n+ 2092 GLuint program;\n+ 2093 GLuint uniformSrcOffset;\n+ 2094 GLuint uniformDstOffset;\n+ 2095 GLuint uniformPatchArray;\n+ 2096 GLuint uniformDuDesc;\n+ 2097 GLuint uniformDvDesc;\n+ 2098 GLuint uniformDuuDesc;\n+ 2099 GLuint uniformDuvDesc;\n+ 2100 GLuint uniformDvvDesc;\n+ 2101 } _patchKernel;\n+ 2102\n+ 2103 int _workGroupSize;\n+ 2104 GLuint _patchArraysSSBO;\n+ 2105};\n+ 2106\n+ 2107} // end namespace Osd\n+ 2108\n+ 2109} // end namespace OPENSUBDIV_VERSION\n+ 2110using namespace OPENSUBDIV_VERSION;\n+ 2111\n+ 2112} // end namespace OpenSubdiv\n+ 2113\n+ 2114\n+ 2115#endif // OPENSUBDIV3_OSD_GL_COMPUTE_EVALUATOR_H\n OpenSubdiv\n Definition: limits.h:32\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n std::vector< PatchArray > PatchArrayVector\n Definition: types.h:132\n OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable\n Stencil table class wrapping the template for compatibility.\n@@ -1112,372 +1050,345 @@\n OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable\n Limit stencil table class wrapping the template for compatibility.\n Definition: stencilTable.h:583\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n BufferDescriptor is a struct which describes buffer elements in interleaved\n data buffers....\n Definition: bufferDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO\n-GL TextureBuffer stencil table.\n-Definition: glXFBEvaluator.h:52\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetSizesTexture\n-GLuint GetSizesTexture() const\n-Definition: glXFBEvaluator.h:72\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::Create\n-static GLStencilTableTBO * Create(Far::StencilTable const *stencilTable, void\n-*deviceContext=NULL)\n-Definition: glXFBEvaluator.h:54\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetDvvWeightsTexture\n-GLuint GetDvvWeightsTexture() const\n-Definition: glXFBEvaluator.h:80\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetNumStencils\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO\n+GL stencil table (Shader Storage buffer)\n+Definition: glComputeEvaluator.h:51\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetDuvWeightsBuffer\n+GLuint GetDuvWeightsBuffer() const\n+Definition: glComputeEvaluator.h:77\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetWeightsBuffer\n+GLuint GetWeightsBuffer() const\n+Definition: glComputeEvaluator.h:73\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GLStencilTableSSBO\n+GLStencilTableSSBO(Far::LimitStencilTable const *limitStencilTable)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetDuWeightsBuffer\n+GLuint GetDuWeightsBuffer() const\n+Definition: glComputeEvaluator.h:74\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::~GLStencilTableSSBO\n+~GLStencilTableSSBO()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetNumStencils\n int GetNumStencils() const\n-Definition: glXFBEvaluator.h:81\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetDuvWeightsTexture\n-GLuint GetDuvWeightsTexture() const\n-Definition: glXFBEvaluator.h:79\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetDuuWeightsTexture\n-GLuint GetDuuWeightsTexture() const\n-Definition: glXFBEvaluator.h:78\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetIndicesTexture\n-GLuint GetIndicesTexture() const\n-Definition: glXFBEvaluator.h:74\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::~GLStencilTableTBO\n-~GLStencilTableTBO()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GLStencilTableTBO\n-GLStencilTableTBO(Far::StencilTable const *stencilTable)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GLStencilTableTBO\n-GLStencilTableTBO(Far::LimitStencilTable const *limitStencilTable)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetWeightsTexture\n-GLuint GetWeightsTexture() const\n-Definition: glXFBEvaluator.h:75\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::Create\n-static GLStencilTableTBO * Create(Far::LimitStencilTable const\n+Definition: glComputeEvaluator.h:79\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetIndicesBuffer\n+GLuint GetIndicesBuffer() const\n+Definition: glComputeEvaluator.h:72\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetOffsetsBuffer\n+GLuint GetOffsetsBuffer() const\n+Definition: glComputeEvaluator.h:71\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GLStencilTableSSBO\n+GLStencilTableSSBO(Far::StencilTable const *stencilTable)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetDvvWeightsBuffer\n+GLuint GetDvvWeightsBuffer() const\n+Definition: glComputeEvaluator.h:78\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetSizesBuffer\n+GLuint GetSizesBuffer() const\n+Definition: glComputeEvaluator.h:70\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::Create\n+static GLStencilTableSSBO * Create(Far::LimitStencilTable const\n *limitStencilTable, void *deviceContext=NULL)\n-Definition: glXFBEvaluator.h:60\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetDvWeightsTexture\n-GLuint GetDvWeightsTexture() const\n-Definition: glXFBEvaluator.h:77\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetDuWeightsTexture\n-GLuint GetDuWeightsTexture() const\n-Definition: glXFBEvaluator.h:76\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GetOffsetsTexture\n-GLuint GetOffsetsTexture() const\n-Definition: glXFBEvaluator.h:73\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator\n-Definition: glXFBEvaluator.h:98\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n+Definition: glComputeEvaluator.h:58\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetDuuWeightsBuffer\n+GLuint GetDuuWeightsBuffer() const\n+Definition: glComputeEvaluator.h:76\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GetDvWeightsBuffer\n+GLuint GetDvWeightsBuffer() const\n+Definition: glComputeEvaluator.h:75\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::Create\n+static GLStencilTableSSBO * Create(Far::StencilTable const *stencilTable, void\n+*deviceContext=NULL)\n+Definition: glComputeEvaluator.h:53\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator\n+Definition: glComputeEvaluator.h:96\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const\n Generic limit eval function. This function has a same signature as other device\n kernels have so that ...\n-Definition: glXFBEvaluator.h:1287\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n-static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, bool interleavedDerivativeBuffers=false)\n-Definition: glXFBEvaluator.h:126\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+Definition: glComputeEvaluator.h:1214\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLComputeEvaluator\n+const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glComputeEvaluator.h:1736\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n &dvDesc, STENCIL_TABLE const *stencilTable) const\n Generic stencil function.\n-Definition: glXFBEvaluator.h:506\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::GLXFBEvaluator\n-GLXFBEvaluator(bool interleavedDerivativeBuffers=false)\n-Constructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+Definition: glComputeEvaluator.h:434\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n+GLComputeEvaluator const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glComputeEvaluator.h:1154\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n+fvarChannel, GLComputeEvaluator const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glComputeEvaluator.h:1606\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable) const\n Generic stencil function.\n-Definition: glXFBEvaluator.h:575\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Instantiatable\n+Definition: glComputeEvaluator.h:502\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void\n+*deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glComputeEvaluator.h:1278\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Instantiatable\n bool Instantiatable\n-Definition: glXFBEvaluator.h:100\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+Definition: glComputeEvaluator.h:98\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint\n duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor\n const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, int\n numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays,\n GLuint patchIndexBuffer, GLuint patchParamsBuffer) const\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n-GLXFBEvaluator const *instance, void *deviceContext=NULL)\n-Generic static stencil function. This function has a same signature as other\n-device kernels have so t...\n-Definition: glXFBEvaluator.h:396\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::~GLComputeEvaluator\n+~GLComputeEvaluator()\n+Destructor. note that the GL context must be made current.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n *patchCoords, PATCH_TABLE *patchTable) const\n Generic limit eval function with derivatives. This function has a same\n signature as other device kern...\n-Definition: glXFBEvaluator.h:1144\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n-static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n-BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, bool\n-interleavedDerivativeBuffers=false)\n-Definition: glXFBEvaluator.h:170\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n-static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, void *deviceContext)\n-Specialization to allow creation without a device context.\n-Definition: glXFBEvaluator.h:117\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+Definition: glComputeEvaluator.h:1071\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable,\n-GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+GLComputeEvaluator const *instance, void *deviceContext=NULL)\n Generic static stencil function. This function has a same signature as other\n device kernels have so t...\n-Definition: glXFBEvaluator.h:306\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void\n-*deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glXFBEvaluator.h:1351\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+Definition: glComputeEvaluator.h:236\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint\n duuBuffer, BufferDescriptor const &duuDesc, GLuint duvBuffer, BufferDescriptor\n const &duvDesc, GLuint dvvBuffer, BufferDescriptor const &dvvDesc, GLuint\n sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer,\n GLuint duWeightsBuffer, GLuint dvWeightsBuffer, GLuint duuWeightsBuffer, GLuint\n duvWeightsBuffer, GLuint dvvWeightsBuffer, int start, int end) const\n-Dispatch the GLSL XFB kernel on on GPU asynchronously returns false if the\n-kernel hasn't been compile...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+Dispatch the GLSL compute kernel on GPU asynchronously returns false if the\n+kernel hasn't been compil...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n+const *stencilTable, GLComputeEvaluator const *instance, void\n+*deviceContext=NULL)\n+Generic static stencil function. This function has a same signature as other\n+device kernels have so t...\n+Definition: glComputeEvaluator.h:168\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator\n+const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glComputeEvaluator.h:1439\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+*patchTable, GLComputeEvaluator const *instance, void *deviceContext=NULL)\n Generic limit eval function. This function has a same signature as other device\n kernels have so that ...\n-Definition: glXFBEvaluator.h:843\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n-static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n-BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc,\n-DEVICE_CONTEXT deviceContext)\n-Generic creator template.\n-Definition: glXFBEvaluator.h:140\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n-static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, DEVICE_CONTEXT deviceContext)\n-Generic creator template.\n-Definition: glXFBEvaluator.h:104\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Synchronize\n-static void Synchronize(void *kernel)\n-Wait the dispatched kernel finishes.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n-GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+Definition: glComputeEvaluator.h:769\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n+GLComputeEvaluator const *instance, void *deviceContext=NULL)\n Generic limit eval function. This function has a same signature as other device\n kernels have so that ...\n-Definition: glXFBEvaluator.h:1227\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n+Definition: glComputeEvaluator.h:1903\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n *patchCoords, PATCH_TABLE *patchTable) const\n Generic limit eval function. This function has a same signature as other device\n kernels have so that ...\n-Definition: glXFBEvaluator.h:1617\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n+Definition: glComputeEvaluator.h:1544\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0) const\n Generic limit eval function. This function has a same signature as other device\n kernels have so that ...\n-Definition: glXFBEvaluator.h:1889\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const\n-*instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glXFBEvaluator.h:765\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, GLXFBEvaluator const\n-*instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glXFBEvaluator.h:1809\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+Definition: glComputeEvaluator.h:1816\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable) const\n Generic limit eval function. This function has a same signature as other device\n kernels have so that ...\n-Definition: glXFBEvaluator.h:1016\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const *instance, void\n-*deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glXFBEvaluator.h:943\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+Definition: glComputeEvaluator.h:943\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::GLComputeEvaluator\n+GLComputeEvaluator()\n+Constructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n *patchTable) const\n Generic limit eval function with derivatives. This function has a same\n signature as other device kern...\n-Definition: glXFBEvaluator.h:1070\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+Definition: glComputeEvaluator.h:997\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator\n+const *instance, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: glComputeEvaluator.h:691\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const\n *stencilTable) const\n Generic stencil function.\n-Definition: glXFBEvaluator.h:457\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatches\n+Definition: glComputeEvaluator.h:386\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n bool EvalPatches(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, int\n numPatchCoords, GLuint patchCoordsBuffer, const PatchArrayVector &patchArrays,\n GLuint patchIndexBuffer, GLuint patchParamsBuffer) const\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Create\n-static GLXFBEvaluator * Create(BufferDescriptor const &srcDesc,\n-BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n-BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n-BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, void\n-*deviceContext)\n-Specialization to allow creation without a device context.\n-Definition: glXFBEvaluator.h:157\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel=0)\n const\n Generic limit eval function. This function has a same signature as other device\n kernels have so that ...\n-Definition: glXFBEvaluator.h:2084\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, GLXFBEvaluator const\n-*instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glXFBEvaluator.h:1512\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+Definition: glComputeEvaluator.h:2011\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Create\n+static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, void *deviceContext=NULL)\n+Definition: glComputeEvaluator.h:99\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n bool EvalStencils(GLuint srcBuffer, BufferDescriptor const &srcDesc, GLuint\n dstBuffer, BufferDescriptor const &dstDesc, GLuint duBuffer, BufferDescriptor\n const &duDesc, GLuint dvBuffer, BufferDescriptor const &dvDesc, GLuint\n sizesBuffer, GLuint offsetsBuffer, GLuint indicesBuffer, GLuint weightsBuffer,\n GLuint duWeightsBuffer, GLuint dvWeightsBuffer, int start, int end) const\n-Dispatch the GLSL XFB kernel on on GPU asynchronously returns false if the\n-kernel hasn't been compile...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n-fvarChannel, GLXFBEvaluator const *instance, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: glXFBEvaluator.h:1679\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalStencils\n+Dispatch the GLSL compute kernel on GPU asynchronously returns false if the\n+kernel hasn't been compil...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalStencils\n static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n-const *stencilTable, GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n+GLComputeEvaluator const *instance, void *deviceContext=NULL)\n Generic static stencil function. This function has a same signature as other\n device kernels have so t...\n-Definition: glXFBEvaluator.h:238\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::~GLXFBEvaluator\n-~GLXFBEvaluator()\n-Destructor. note that the GL context must be made current.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::Compile\n+Definition: glComputeEvaluator.h:326\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Synchronize\n+static void Synchronize(void *deviceContext)\n+Wait the dispatched kernel finishes.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Compile\n bool Compile(BufferDescriptor const &srcDesc, BufferDescriptor const &dstDesc,\n BufferDescriptor const &duDesc=BufferDescriptor(), BufferDescriptor const\n &dvDesc=BufferDescriptor(), BufferDescriptor const &duuDesc=BufferDescriptor(),\n BufferDescriptor const &duvDesc=BufferDescriptor(), BufferDescriptor const\n &dvvDesc=BufferDescriptor())\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesVarying\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesVarying\n bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n *patchTable) const\n Generic limit eval function. This function has a same signature as other device\n kernels have so that ...\n-Definition: glXFBEvaluator.h:1428\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n+Definition: glComputeEvaluator.h:1355\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::Create\n+static GLComputeEvaluator * Create(BufferDescriptor const &srcDesc,\n+BufferDescriptor const &dstDesc, BufferDescriptor const &duDesc,\n+BufferDescriptor const &dvDesc, BufferDescriptor const &duuDesc,\n+BufferDescriptor const &duvDesc, BufferDescriptor const &dvvDesc, void\n+*deviceContext=NULL)\n+Definition: glComputeEvaluator.h:111\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatchesFaceVarying\n bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n fvarChannel=0) const\n Generic limit eval function. This function has a same signature as other device\n kernels have so that ...\n-Definition: glXFBEvaluator.h:1742\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n-GLXFBEvaluator const *instance, void *deviceContext=NULL)\n+Definition: glComputeEvaluator.h:1669\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, GLComputeEvaluator const *instance, void\n+*deviceContext=NULL)\n Generic limit eval function. This function has a same signature as other device\n kernels have so that ...\n-Definition: glXFBEvaluator.h:1976\n+Definition: glComputeEvaluator.h:870\n * opensubdiv\n * osd\n- * glXFBEvaluator.h\n+ * glComputeEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00854.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00854.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glMesh.h File Reference\n+OpenSubdiv: opensubdiv/osd/glslPatchShaderSource.h File Reference\n \n \n \n \n \n \n \n@@ -81,43 +81,43 @@\n \n \n \n
\n
\n-Namespaces |\n-Typedefs
\n-
glMesh.h File Reference
\n+Classes |\n+Namespaces
\n+
glslPatchShaderSource.h File Reference
\n \n
\n
#include "../version.h"
\n-#include "../osd/mesh.h"
\n-#include "../osd/glPatchTable.h"
\n+#include <string>
\n+#include "../far/patchDescriptor.h"
\n
\n

Go to the source code of this file.

\n \n+\n+\n+\n+

\n+Classes

class  GLSLPatchShaderSource
 
\n \n \n \n \n \n \n \n-

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n-\n-\n-\n

\n-Typedefs

typedef MeshInterface< GLPatchTable > GLMeshInterface
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,27 +4,27 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-Namespaces | Typedefs\n-glMesh.h File Reference\n+Classes | Namespaces\n+glslPatchShaderSource.h File Reference\n #include \"../version.h\"\n-#include \"../osd/mesh.h\"\n-#include \"../osd/glPatchTable.h\"\n+#include \n+#include \"../far/patchDescriptor.h\"\n Go_to_the_source_code_of_this_file.\n+ Classes\n+class \u00a0GLSLPatchShaderSource\n+\u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n- Typedefs\n-typedef MeshInterface< GLPatchTable >\u00a0GLMeshInterface\n-\u00a0\n * opensubdiv\n * osd\n- * glMesh.h\n+ * glslPatchShaderSource.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00854.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00854.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00854 = [\n- [\"GLMeshInterface\", \"a00854.html#a7f9cf2b729178bdbb847fc4afd91af9e\", null]\n+ [\"GLSLPatchShaderSource\", \"a01233.html\", null]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00854_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00854_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glMesh.h Source File\n+OpenSubdiv: opensubdiv/osd/glslPatchShaderSource.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
glMesh.h
\n+
glslPatchShaderSource.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2013 Pixar
\n+
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,45 +107,62 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_GL_MESH_H
\n-
26#define OPENSUBDIV3_OSD_GL_MESH_H
\n+
25#ifndef OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE_H
\n+
26#define OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE_H
\n
27
\n
28#include "../version.h"
\n-
29
\n-
30#include "../osd/mesh.h"
\n-
31#include "../osd/glPatchTable.h"
\n-
32
\n-
33namespace OpenSubdiv {
\n-
34namespace OPENSUBDIV_VERSION {
\n-
35
\n-
36namespace Osd {
\n-
37
\n-
38typedef MeshInterface<GLPatchTable> GLMeshInterface;
\n-
39
\n+
29#include <string>
\n+
30#include "../far/patchDescriptor.h"
\n+
31
\n+
32namespace OpenSubdiv {
\n+
33namespace OPENSUBDIV_VERSION {
\n+
34
\n+
35namespace Osd {
\n+
36
\n+
37class GLSLPatchShaderSource {
\n+
38public:
\n+
39 static std::string GetCommonShaderSource();
\n
40
\n-
41} // end namespace Osd
\n+
41 static std::string GetPatchBasisShaderSource();
\n
42
\n-
43} // end namespace OPENSUBDIV_VERSION
\n-
44using namespace OPENSUBDIV_VERSION;
\n+
43 static std::string GetVertexShaderSource(
\n+
44 Far::PatchDescriptor::Type type);
\n
45
\n-
46} // end namespace OpenSubdiv
\n-
47
\n-
48#endif // OPENSUBDIV3_OSD_GL_MESH_H
\n+
46 static std::string GetTessControlShaderSource(
\n+
47 Far::PatchDescriptor::Type type);
\n+
48
\n+
49 static std::string GetTessEvalShaderSource(
\n+
50 Far::PatchDescriptor::Type type);
\n+
51};
\n+
52
\n+
53} // end namespace Osd
\n+
54
\n+
55} // end namespace OPENSUBDIV_VERSION
\n+
56using namespace OPENSUBDIV_VERSION;
\n+
57
\n+
58} // end namespace OpenSubdiv
\n+
59
\n+
60#endif // OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLMeshInterface
MeshInterface< GLPatchTable > GLMeshInterface
Definition: glMesh.h:38
\n-\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::GetVertexShaderSource
static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::GetTessEvalShaderSource
static std::string GetTessEvalShaderSource(Far::PatchDescriptor::Type type)
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::GetTessControlShaderSource
static std::string GetTessControlShaderSource(Far::PatchDescriptor::Type type)
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-glMesh.h\n+glslPatchShaderSource.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2013 Pixar\n+ 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,42 +30,69 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_GL_MESH_H\n- 26#define OPENSUBDIV3_OSD_GL_MESH_H\n+ 25#ifndef OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE_H\n+ 26#define OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE_H\n 27\n 28#include \"../version.h\"\n- 29\n- 30#include \"../osd/mesh.h\"\n- 31#include \"../osd/glPatchTable.h\"\n- 32\n- 33namespace OpenSubdiv {\n- 34namespace OPENSUBDIV_VERSION {\n- 35\n- 36namespace Osd {\n- 37\n-38typedef MeshInterface GLMeshInterface;\n- 39\n+ 29#include \n+ 30#include \"../far/patchDescriptor.h\"\n+ 31\n+ 32namespace OpenSubdiv {\n+ 33namespace OPENSUBDIV_VERSION {\n+ 34\n+ 35namespace Osd {\n+ 36\n+37class GLSLPatchShaderSource {\n+ 38public:\n+39 static std::string GetCommonShaderSource();\n 40\n- 41} // end namespace Osd\n+41 static std::string GetPatchBasisShaderSource();\n 42\n- 43} // end namespace OPENSUBDIV_VERSION\n- 44using namespace OPENSUBDIV_VERSION;\n+43 static std::string GetVertexShaderSource(\n+ 44 Far::PatchDescriptor::Type type);\n 45\n- 46} // end namespace OpenSubdiv\n- 47\n- 48#endif // OPENSUBDIV3_OSD_GL_MESH_H\n+46 static std::string GetTessControlShaderSource(\n+ 47 Far::PatchDescriptor::Type type);\n+ 48\n+49 static std::string GetTessEvalShaderSource(\n+ 50 Far::PatchDescriptor::Type type);\n+ 51};\n+ 52\n+ 53} // end namespace Osd\n+ 54\n+ 55} // end namespace OPENSUBDIV_VERSION\n+ 56using namespace OPENSUBDIV_VERSION;\n+ 57\n+ 58} // end namespace OpenSubdiv\n+ 59\n+ 60#endif // OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLMeshInterface\n-MeshInterface< GLPatchTable > GLMeshInterface\n-Definition: glMesh.h:38\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface\n-Definition: mesh.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n+Type\n+Definition: patchDescriptor.h:48\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource\n+Definition: glslPatchShaderSource.h:37\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::\n+GetVertexShaderSource\n+static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::\n+GetTessEvalShaderSource\n+static std::string GetTessEvalShaderSource(Far::PatchDescriptor::Type type)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::\n+GetPatchBasisShaderSource\n+static std::string GetPatchBasisShaderSource()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::\n+GetCommonShaderSource\n+static std::string GetCommonShaderSource()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::\n+GetTessControlShaderSource\n+static std::string GetTessControlShaderSource(Far::PatchDescriptor::Type type)\n * opensubdiv\n * osd\n- * glMesh.h\n+ * glslPatchShaderSource.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00857.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00857.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glVertexBuffer.h File Reference\n+OpenSubdiv: opensubdiv/osd/glMesh.h File Reference\n \n \n \n \n \n \n \n@@ -81,44 +81,43 @@\n \n \n \n
\n
\n-Classes |\n-Namespaces
\n-
glVertexBuffer.h File Reference
\n+Namespaces |\n+Typedefs
\n+
glMesh.h File Reference
\n \n
\n
#include "../version.h"
\n-#include "../osd/opengl.h"
\n-#include <cstddef>
\n+#include "../osd/mesh.h"
\n+#include "../osd/glPatchTable.h"
\n
\n

Go to the source code of this file.

\n \n-\n-\n-\n-\n-

\n-Classes

class  GLVertexBuffer
 Concrete vertex buffer class for GLSL subdivision and OpenGL drawing. More...
 
\n \n \n \n \n \n \n \n+

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n+\n+\n+\n

\n+Typedefs

typedef MeshInterface< GLPatchTable > GLMeshInterface
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,29 +4,27 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-Classes | Namespaces\n-glVertexBuffer.h File Reference\n+Namespaces | Typedefs\n+glMesh.h File Reference\n #include \"../version.h\"\n-#include \"../osd/opengl.h\"\n-#include \n+#include \"../osd/mesh.h\"\n+#include \"../osd/glPatchTable.h\"\n Go_to_the_source_code_of_this_file.\n- Classes\n-class \u00a0GLVertexBuffer\n-\u00a0 Concrete vertex buffer class for GLSL subdivision and OpenGL drawing.\n- More...\n-\u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n+ Typedefs\n+typedef MeshInterface< GLPatchTable >\u00a0GLMeshInterface\n+\u00a0\n * opensubdiv\n * osd\n- * glVertexBuffer.h\n+ * glMesh.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00857.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00857.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00857 = [\n- [\"GLVertexBuffer\", \"a01237.html\", \"a01237\"]\n+ [\"GLMeshInterface\", \"a00857.html#a7f9cf2b729178bdbb847fc4afd91af9e\", null]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00857_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00857_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glVertexBuffer.h Source File\n+OpenSubdiv: opensubdiv/osd/glMesh.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
glVertexBuffer.h
\n+
glMesh.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,76 +107,45 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H
\n-
26#define OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H
\n+
25#ifndef OPENSUBDIV3_OSD_GL_MESH_H
\n+
26#define OPENSUBDIV3_OSD_GL_MESH_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30#include "../osd/opengl.h"
\n-
31#include <cstddef>
\n+
30#include "../osd/mesh.h"
\n+
31#include "../osd/glPatchTable.h"
\n
32
\n
33namespace OpenSubdiv {
\n
34namespace OPENSUBDIV_VERSION {
\n
35
\n
36namespace Osd {
\n
37
\n-
44class GLVertexBuffer {
\n-
45public:
\n-
47 static GLVertexBuffer * Create(int numElements, int numVertices,
\n-
48 void *deviceContext = NULL);
\n-
49
\n-
51 ~GLVertexBuffer();
\n-
52
\n-
55 void UpdateData(const float *src, int startVertex, int numVertices,
\n-
56 void *deviceContext = NULL);
\n-
57
\n-
59 int GetNumElements() const;
\n-
60
\n-
62 int GetNumVertices() const;
\n-
63
\n-
65 GLuint BindVBO(void *deviceContext = NULL);
\n-
66
\n-
67protected:
\n-
69 GLVertexBuffer(int numElements, int numVertices);
\n-
70
\n-
73 bool allocate();
\n-
74
\n-
75private:
\n-
76 int _numElements;
\n-
77 int _numVertices;
\n-
78 GLuint _vbo;
\n-
79};
\n-
80
\n-
81} // end namespace Osd
\n-
82
\n-
83} // end namespace OPENSUBDIV_VERSION
\n-
84using namespace OPENSUBDIV_VERSION;
\n-
85
\n-
86} // end namespace OpenSubdiv
\n-
87
\n-
88#endif // OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H
\n+
38typedef MeshInterface<GLPatchTable> GLMeshInterface;
\n+
39
\n+
40
\n+
41} // end namespace Osd
\n+
42
\n+
43} // end namespace OPENSUBDIV_VERSION
\n+
44using namespace OPENSUBDIV_VERSION;
\n+
45
\n+
46} // end namespace OpenSubdiv
\n+
47
\n+
48#endif // OPENSUBDIV3_OSD_GL_MESH_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer
Concrete vertex buffer class for GLSL subdivision and OpenGL drawing.
Definition: glVertexBuffer.h:44
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::BindVBO
GLuint BindVBO(void *deviceContext=NULL)
Returns the GL buffer object.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GetNumElements
int GetNumElements() const
Returns how many elements defined in this vertex buffer.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::UpdateData
void UpdateData(const float *src, int startVertex, int numVertices, void *deviceContext=NULL)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GLVertexBuffer
GLVertexBuffer(int numElements, int numVertices)
Constructor.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GetNumVertices
int GetNumVertices() const
Returns how many vertices allocated in this vertex buffer.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::Create
static GLVertexBuffer * Create(int numElements, int numVertices, void *deviceContext=NULL)
Creator. Returns NULL if error.
\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLMeshInterface
MeshInterface< GLPatchTable > GLMeshInterface
Definition: glMesh.h:38
\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-glVertexBuffer.h\n+glMesh.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,88 +30,42 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H\n- 26#define OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H\n+ 25#ifndef OPENSUBDIV3_OSD_GL_MESH_H\n+ 26#define OPENSUBDIV3_OSD_GL_MESH_H\n 27\n 28#include \"../version.h\"\n 29\n- 30#include \"../osd/opengl.h\"\n- 31#include \n+ 30#include \"../osd/mesh.h\"\n+ 31#include \"../osd/glPatchTable.h\"\n 32\n 33namespace OpenSubdiv {\n 34namespace OPENSUBDIV_VERSION {\n 35\n 36namespace Osd {\n 37\n-44class GLVertexBuffer {\n- 45public:\n-47 static GLVertexBuffer * Create(int numElements, int numVertices,\n- 48 void *deviceContext = NULL);\n- 49\n-51 ~GLVertexBuffer();\n- 52\n-55 void UpdateData(const float *src, int startVertex, int numVertices,\n- 56 void *deviceContext = NULL);\n- 57\n-59 int GetNumElements() const;\n- 60\n-62 int GetNumVertices() const;\n- 63\n-65 GLuint BindVBO(void *deviceContext = NULL);\n- 66\n- 67protected:\n-69 GLVertexBuffer(int numElements, int numVertices);\n- 70\n-73 bool allocate();\n- 74\n- 75private:\n- 76 int _numElements;\n- 77 int _numVertices;\n- 78 GLuint _vbo;\n- 79};\n- 80\n- 81} // end namespace Osd\n- 82\n- 83} // end namespace OPENSUBDIV_VERSION\n- 84using namespace OPENSUBDIV_VERSION;\n- 85\n- 86} // end namespace OpenSubdiv\n- 87\n- 88#endif // OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H\n+38typedef MeshInterface GLMeshInterface;\n+ 39\n+ 40\n+ 41} // end namespace Osd\n+ 42\n+ 43} // end namespace OPENSUBDIV_VERSION\n+ 44using namespace OPENSUBDIV_VERSION;\n+ 45\n+ 46} // end namespace OpenSubdiv\n+ 47\n+ 48#endif // OPENSUBDIV3_OSD_GL_MESH_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer\n-Concrete vertex buffer class for GLSL subdivision and OpenGL drawing.\n-Definition: glVertexBuffer.h:44\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::~GLVertexBuffer\n-~GLVertexBuffer()\n-Destructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::BindVBO\n-GLuint BindVBO(void *deviceContext=NULL)\n-Returns the GL buffer object.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GetNumElements\n-int GetNumElements() const\n-Returns how many elements defined in this vertex buffer.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::UpdateData\n-void UpdateData(const float *src, int startVertex, int numVertices, void\n-*deviceContext=NULL)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GLVertexBuffer\n-GLVertexBuffer(int numElements, int numVertices)\n-Constructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GetNumVertices\n-int GetNumVertices() const\n-Returns how many vertices allocated in this vertex buffer.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::Create\n-static GLVertexBuffer * Create(int numElements, int numVertices, void\n-*deviceContext=NULL)\n-Creator. Returns NULL if error.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::allocate\n-bool allocate()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLMeshInterface\n+MeshInterface< GLPatchTable > GLMeshInterface\n+Definition: glMesh.h:38\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface\n+Definition: mesh.h:68\n * opensubdiv\n * osd\n- * glVertexBuffer.h\n+ * glMesh.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00860.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00860.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glslPatchShaderSource.h File Reference\n+OpenSubdiv: opensubdiv/osd/glPatchTable.h File Reference\n \n \n \n \n \n \n \n@@ -83,41 +83,45 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
glslPatchShaderSource.h File Reference
\n+
glPatchTable.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include <string>
\n-#include "../far/patchDescriptor.h"
\n+#include "../osd/nonCopyable.h"
\n+#include "../osd/opengl.h"
\n+#include "../osd/types.h"
\n+#include <vector>
\n
\n

Go to the source code of this file.

\n \n \n-\n+\n \n

\n Classes

class  GLSLPatchShaderSource
class  GLPatchTable
 
\n \n \n \n \n \n+\n+\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,26 +5,30 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-glslPatchShaderSource.h File Reference\n+glPatchTable.h File Reference\n #include \"../version.h\"\n-#include \n-#include \"../far/patchDescriptor.h\"\n+#include \"../osd/nonCopyable.h\"\n+#include \"../osd/opengl.h\"\n+#include \"../osd/types.h\"\n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0GLSLPatchShaderSource\n+class \u00a0GLPatchTable\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n+\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * glslPatchShaderSource.h\n+ * glPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00860.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00860.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00860 = [\n- [\"GLSLPatchShaderSource\", \"a01233.html\", null]\n+ [\"GLPatchTable\", \"a01229.html\", \"a01229\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00860_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00860_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glslPatchShaderSource.h Source File\n+OpenSubdiv: opensubdiv/osd/glPatchTable.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
glslPatchShaderSource.h
\n+
glPatchTable.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,62 +107,172 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE_H
\n-
26#define OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE_H
\n+
25#ifndef OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n+
26#define OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n
27
\n
28#include "../version.h"
\n-
29#include <string>
\n-
30#include "../far/patchDescriptor.h"
\n-
31
\n-
32namespace OpenSubdiv {
\n-
33namespace OPENSUBDIV_VERSION {
\n-
34
\n-
35namespace Osd {
\n-
36
\n-
37class GLSLPatchShaderSource {
\n-
38public:
\n-
39 static std::string GetCommonShaderSource();
\n-
40
\n-
41 static std::string GetPatchBasisShaderSource();
\n+
29
\n+
30#include "../osd/nonCopyable.h"
\n+
31#include "../osd/opengl.h"
\n+
32#include "../osd/types.h"
\n+
33
\n+
34#include <vector>
\n+
35
\n+
36namespace OpenSubdiv {
\n+
37namespace OPENSUBDIV_VERSION {
\n+
38
\n+
39namespace Far{
\n+
40 class PatchTable;
\n+
41};
\n
42
\n-
43 static std::string GetVertexShaderSource(
\n-
44 Far::PatchDescriptor::Type type);
\n-
45
\n-
46 static std::string GetTessControlShaderSource(
\n-
47 Far::PatchDescriptor::Type type);
\n+
43namespace Osd {
\n+
44
\n+
45class GLPatchTable : private NonCopyable<GLPatchTable> {
\n+
46public:
\n+
47 typedef GLuint VertexBufferBinding;
\n
48
\n-
49 static std::string GetTessEvalShaderSource(
\n-
50 Far::PatchDescriptor::Type type);
\n-
51};
\n-
52
\n-
53} // end namespace Osd
\n-
54
\n-
55} // end namespace OPENSUBDIV_VERSION
\n-
56using namespace OPENSUBDIV_VERSION;
\n-
57
\n-
58} // end namespace OpenSubdiv
\n-
59
\n-
60#endif // OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE
\n+
49 ~GLPatchTable();
\n+
50
\n+
51 static GLPatchTable *Create(Far::PatchTable const *farPatchTable,
\n+
52 void *deviceContext = NULL);
\n+
53
\n+
55 PatchArrayVector const &GetPatchArrays() const {
\n+
56 return _patchArrays;
\n+
57 }
\n+
58
\n+
60 GLuint GetPatchIndexBuffer() const {
\n+
61 return _patchIndexBuffer;
\n+
62 }
\n+
63
\n+
65 GLuint GetPatchParamBuffer() const {
\n+
66 return _patchParamBuffer;
\n+
67 }
\n+
68
\n+
70 GLuint GetPatchIndexTextureBuffer() const {
\n+
71 return _patchIndexTexture;
\n+
72 }
\n+
73
\n+
75 GLuint GetPatchParamTextureBuffer() const {
\n+
76 return _patchParamTexture;
\n+
77 }
\n+
78
\n+
80 PatchArrayVector const &GetVaryingPatchArrays() const {
\n+
81 return _varyingPatchArrays;
\n+
82 }
\n+
83
\n+
85 GLuint GetVaryingPatchIndexBuffer() const {
\n+
86 return _varyingIndexBuffer;
\n+
87 }
\n+
88
\n+
90 GLuint GetVaryingPatchIndexTextureBuffer() const {
\n+
91 return _varyingIndexTexture;
\n+
92 }
\n+
93
\n+
95 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }
\n+
96
\n+
98 PatchArrayVector const &GetFVarPatchArrays(int fvarChannel = 0) const {
\n+
99 return _fvarPatchArrays[fvarChannel];
\n+
100 }
\n+
101
\n+
103 GLuint GetFVarPatchIndexBuffer(int fvarChannel = 0) const {
\n+
104 return _fvarIndexBuffers[fvarChannel];
\n+
105 }
\n+
106
\n+
108 GLuint GetFVarPatchIndexTextureBuffer(int fvarChannel = 0) const {
\n+
109 return _fvarIndexTextures[fvarChannel];
\n+
110 }
\n+
111
\n+
113 GLuint GetFVarPatchParamBuffer(int fvarChannel = 0) const {
\n+
114 return _fvarParamBuffers[fvarChannel];
\n+
115 }
\n+
116
\n+
118 GLuint GetFVarPatchParamTextureBuffer(int fvarChannel = 0) const {
\n+
119 return _fvarParamTextures[fvarChannel];
\n+
120 }
\n+
121
\n+
122protected:
\n+
123 GLPatchTable();
\n+
124
\n+
125 // allocate buffers from patchTable
\n+
126 bool allocate(Far::PatchTable const *farPatchTable);
\n+
127
\n+
128 PatchArrayVector _patchArrays;
\n+
129
\n+
130 GLuint _patchIndexBuffer;
\n+
131 GLuint _patchParamBuffer;
\n+
132
\n+
133 GLuint _patchIndexTexture;
\n+
134 GLuint _patchParamTexture;
\n+
135
\n+
136 PatchArrayVector _varyingPatchArrays;
\n+
137 GLuint _varyingIndexBuffer;
\n+
138 GLuint _varyingIndexTexture;
\n+
139
\n+
140 std::vector<PatchArrayVector> _fvarPatchArrays;
\n+
141 std::vector<GLuint> _fvarIndexBuffers;
\n+
142 std::vector<GLuint> _fvarIndexTextures;
\n+
143
\n+
144 std::vector<GLuint> _fvarParamBuffers;
\n+
145 std::vector<GLuint> _fvarParamTextures;
\n+
146};
\n+
147
\n+
148
\n+
149} // end namespace Osd
\n+
150
\n+
151} // end namespace OPENSUBDIV_VERSION
\n+
152using namespace OPENSUBDIV_VERSION;
\n+
153
\n+
154} // end namespace OpenSubdiv
\n+
155
\n+
156#endif // OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n \n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::GetVertexShaderSource
static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::GetTessEvalShaderSource
static std::string GetTessEvalShaderSource(Far::PatchDescriptor::Type type)
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::GetTessControlShaderSource
static std::string GetTessControlShaderSource(Far::PatchDescriptor::Type type)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchArrays
PatchArrayVector const & GetPatchArrays() const
Returns the patch arrays for vertex index buffer data.
Definition: glPatchTable.h:55
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchIndexBuffer
GLuint GetFVarPatchIndexBuffer(int fvarChannel=0) const
Returns the GL index buffer containing face-varying control vertices.
Definition: glPatchTable.h:103
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchParamTextureBuffer
GLuint GetPatchParamTextureBuffer() const
Returns the GL texture buffer containing the patch parameter.
Definition: glPatchTable.h:75
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::allocate
bool allocate(Far::PatchTable const *farPatchTable)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::Create
static GLPatchTable * Create(Far::PatchTable const *farPatchTable, void *deviceContext=NULL)
\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchParamBuffer
GLuint GetPatchParamBuffer() const
Returns the GL index buffer containing the patch parameter.
Definition: glPatchTable.h:65
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchParamBuffer
GLuint GetFVarPatchParamBuffer(int fvarChannel=0) const
Returns the GL index buffer containing face-varying patch params.
Definition: glPatchTable.h:113
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchIndexTextureBuffer
GLuint GetFVarPatchIndexTextureBuffer(int fvarChannel=0) const
Returns the GL texture buffer containing face-varying control vertices.
Definition: glPatchTable.h:108
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchParamTextureBuffer
GLuint GetFVarPatchParamTextureBuffer(int fvarChannel=0) const
Returns the GL texture buffer containing face-varying patch params.
Definition: glPatchTable.h:118
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchIndexBuffer
GLuint GetPatchIndexBuffer() const
Returns the GL index buffer containing the patch control vertices.
Definition: glPatchTable.h:60
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetNumFVarChannels
int GetNumFVarChannels() const
Returns the number of face-varying channel buffers.
Definition: glPatchTable.h:95
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchArrays
PatchArrayVector const & GetFVarPatchArrays(int fvarChannel=0) const
Returns the patch arrays for face-varying index buffer data.
Definition: glPatchTable.h:98
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarPatchArrays
std::vector< PatchArrayVector > _fvarPatchArrays
Definition: glPatchTable.h:140
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchIndexTextureBuffer
GLuint GetPatchIndexTextureBuffer() const
Returns the GL texture buffer containing the patch control vertices.
Definition: glPatchTable.h:70
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetVaryingPatchIndexTextureBuffer
GLuint GetVaryingPatchIndexTextureBuffer() const
Returns the GL texture buffer containing the varying control vertices.
Definition: glPatchTable.h:90
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetVaryingPatchIndexBuffer
GLuint GetVaryingPatchIndexBuffer() const
Returns the GL index buffer containing the varying control vertices.
Definition: glPatchTable.h:85
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetVaryingPatchArrays
PatchArrayVector const & GetVaryingPatchArrays() const
Returns the patch arrays for varying index buffer data.
Definition: glPatchTable.h:80
\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-glslPatchShaderSource.h\n+glPatchTable.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,69 +30,249 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE_H\n- 26#define OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE_H\n+ 25#ifndef OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n+ 26#define OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n 27\n 28#include \"../version.h\"\n- 29#include \n- 30#include \"../far/patchDescriptor.h\"\n- 31\n- 32namespace OpenSubdiv {\n- 33namespace OPENSUBDIV_VERSION {\n- 34\n- 35namespace Osd {\n- 36\n-37class GLSLPatchShaderSource {\n- 38public:\n-39 static std::string GetCommonShaderSource();\n- 40\n-41 static std::string GetPatchBasisShaderSource();\n+ 29\n+ 30#include \"../osd/nonCopyable.h\"\n+ 31#include \"../osd/opengl.h\"\n+ 32#include \"../osd/types.h\"\n+ 33\n+ 34#include \n+ 35\n+ 36namespace OpenSubdiv {\n+ 37namespace OPENSUBDIV_VERSION {\n+ 38\n+ 39namespace Far{\n+ 40 class PatchTable;\n+ 41};\n 42\n-43 static std::string GetVertexShaderSource(\n- 44 Far::PatchDescriptor::Type type);\n- 45\n-46 static std::string GetTessControlShaderSource(\n- 47 Far::PatchDescriptor::Type type);\n+ 43namespace Osd {\n+ 44\n+45class GLPatchTable : private NonCopyable {\n+ 46public:\n+47 typedef GLuint VertexBufferBinding;\n 48\n-49 static std::string GetTessEvalShaderSource(\n- 50 Far::PatchDescriptor::Type type);\n- 51};\n- 52\n- 53} // end namespace Osd\n- 54\n- 55} // end namespace OPENSUBDIV_VERSION\n- 56using namespace OPENSUBDIV_VERSION;\n- 57\n- 58} // end namespace OpenSubdiv\n- 59\n- 60#endif // OPENSUBDIV3_OSD_GLSL_PATCH_SHADER_SOURCE\n+49 ~GLPatchTable();\n+ 50\n+51 static GLPatchTable *Create(Far::PatchTable const *farPatchTable,\n+ 52 void *deviceContext = NULL);\n+ 53\n+55 PatchArrayVector const &GetPatchArrays() const {\n+ 56 return _patchArrays;\n+ 57 }\n+ 58\n+60 GLuint GetPatchIndexBuffer() const {\n+ 61 return _patchIndexBuffer;\n+ 62 }\n+ 63\n+65 GLuint GetPatchParamBuffer() const {\n+ 66 return _patchParamBuffer;\n+ 67 }\n+ 68\n+70 GLuint GetPatchIndexTextureBuffer() const {\n+ 71 return _patchIndexTexture;\n+ 72 }\n+ 73\n+75 GLuint GetPatchParamTextureBuffer() const {\n+ 76 return _patchParamTexture;\n+ 77 }\n+ 78\n+80 PatchArrayVector const &GetVaryingPatchArrays() const {\n+ 81 return _varyingPatchArrays;\n+ 82 }\n+ 83\n+85 GLuint GetVaryingPatchIndexBuffer() const {\n+ 86 return _varyingIndexBuffer;\n+ 87 }\n+ 88\n+90 GLuint GetVaryingPatchIndexTextureBuffer() const {\n+ 91 return _varyingIndexTexture;\n+ 92 }\n+ 93\n+95 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }\n+ 96\n+98 PatchArrayVector const &GetFVarPatchArrays(int fvarChannel = 0) const {\n+ 99 return _fvarPatchArrays[fvarChannel];\n+ 100 }\n+ 101\n+103 GLuint GetFVarPatchIndexBuffer(int fvarChannel = 0) const {\n+ 104 return _fvarIndexBuffers[fvarChannel];\n+ 105 }\n+ 106\n+108 GLuint GetFVarPatchIndexTextureBuffer(int fvarChannel = 0) const {\n+ 109 return _fvarIndexTextures[fvarChannel];\n+ 110 }\n+ 111\n+113 GLuint GetFVarPatchParamBuffer(int fvarChannel = 0) const {\n+ 114 return _fvarParamBuffers[fvarChannel];\n+ 115 }\n+ 116\n+118 GLuint GetFVarPatchParamTextureBuffer(int fvarChannel = 0) const {\n+ 119 return _fvarParamTextures[fvarChannel];\n+ 120 }\n+ 121\n+ 122protected:\n+123 GLPatchTable();\n+ 124\n+ 125 // allocate buffers from patchTable\n+126 bool allocate(Far::PatchTable const *farPatchTable);\n+ 127\n+128 PatchArrayVector _patchArrays;\n+ 129\n+130 GLuint _patchIndexBuffer;\n+131 GLuint _patchParamBuffer;\n+ 132\n+133 GLuint _patchIndexTexture;\n+134 GLuint _patchParamTexture;\n+ 135\n+136 PatchArrayVector _varyingPatchArrays;\n+137 GLuint _varyingIndexBuffer;\n+138 GLuint _varyingIndexTexture;\n+ 139\n+140 std::vector _fvarPatchArrays;\n+141 std::vector _fvarIndexBuffers;\n+142 std::vector _fvarIndexTextures;\n+ 143\n+144 std::vector _fvarParamBuffers;\n+145 std::vector _fvarParamTextures;\n+ 146};\n+ 147\n+ 148\n+ 149} // end namespace Osd\n+ 150\n+ 151} // end namespace OPENSUBDIV_VERSION\n+ 152using namespace OPENSUBDIV_VERSION;\n+ 153\n+ 154} // end namespace OpenSubdiv\n+ 155\n+ 156#endif // OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n-Type\n-Definition: patchDescriptor.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource\n-Definition: glslPatchShaderSource.h:37\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::\n-GetVertexShaderSource\n-static std::string GetVertexShaderSource(Far::PatchDescriptor::Type type)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::\n-GetTessEvalShaderSource\n-static std::string GetTessEvalShaderSource(Far::PatchDescriptor::Type type)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::\n-GetPatchBasisShaderSource\n-static std::string GetPatchBasisShaderSource()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::\n-GetCommonShaderSource\n-static std::string GetCommonShaderSource()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLSLPatchShaderSource::\n-GetTessControlShaderSource\n-static std::string GetTessControlShaderSource(Far::PatchDescriptor::Type type)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n+std::vector< PatchArray > PatchArrayVector\n+Definition: types.h:132\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n+Container for arrays of parametric patches.\n+Definition: patchTable.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable\n+Definition: glPatchTable.h:45\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::VertexBufferBinding\n+GLuint VertexBufferBinding\n+Definition: glPatchTable.h:47\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarIndexBuffers\n+std::vector< GLuint > _fvarIndexBuffers\n+Definition: glPatchTable.h:141\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchArrays\n+PatchArrayVector const & GetPatchArrays() const\n+Returns the patch arrays for vertex index buffer data.\n+Definition: glPatchTable.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchIndexBuffer\n+GLuint GetFVarPatchIndexBuffer(int fvarChannel=0) const\n+Returns the GL index buffer containing face-varying control vertices.\n+Definition: glPatchTable.h:103\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::~GLPatchTable\n+~GLPatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarParamBuffers\n+std::vector< GLuint > _fvarParamBuffers\n+Definition: glPatchTable.h:144\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchParamTextureBuffer\n+GLuint GetPatchParamTextureBuffer() const\n+Returns the GL texture buffer containing the patch parameter.\n+Definition: glPatchTable.h:75\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_varyingIndexBuffer\n+GLuint _varyingIndexBuffer\n+Definition: glPatchTable.h:137\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::allocate\n+bool allocate(Far::PatchTable const *farPatchTable)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::Create\n+static GLPatchTable * Create(Far::PatchTable const *farPatchTable, void\n+*deviceContext=NULL)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_varyingIndexTexture\n+GLuint _varyingIndexTexture\n+Definition: glPatchTable.h:138\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarParamTextures\n+std::vector< GLuint > _fvarParamTextures\n+Definition: glPatchTable.h:145\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_patchArrays\n+PatchArrayVector _patchArrays\n+Definition: glPatchTable.h:128\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_patchIndexBuffer\n+GLuint _patchIndexBuffer\n+Definition: glPatchTable.h:130\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchParamBuffer\n+GLuint GetPatchParamBuffer() const\n+Returns the GL index buffer containing the patch parameter.\n+Definition: glPatchTable.h:65\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchParamBuffer\n+GLuint GetFVarPatchParamBuffer(int fvarChannel=0) const\n+Returns the GL index buffer containing face-varying patch params.\n+Definition: glPatchTable.h:113\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_patchParamTexture\n+GLuint _patchParamTexture\n+Definition: glPatchTable.h:134\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GLPatchTable\n+GLPatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::\n+GetFVarPatchIndexTextureBuffer\n+GLuint GetFVarPatchIndexTextureBuffer(int fvarChannel=0) const\n+Returns the GL texture buffer containing face-varying control vertices.\n+Definition: glPatchTable.h:108\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::\n+GetFVarPatchParamTextureBuffer\n+GLuint GetFVarPatchParamTextureBuffer(int fvarChannel=0) const\n+Returns the GL texture buffer containing face-varying patch params.\n+Definition: glPatchTable.h:118\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchIndexBuffer\n+GLuint GetPatchIndexBuffer() const\n+Returns the GL index buffer containing the patch control vertices.\n+Definition: glPatchTable.h:60\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetNumFVarChannels\n+int GetNumFVarChannels() const\n+Returns the number of face-varying channel buffers.\n+Definition: glPatchTable.h:95\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchArrays\n+PatchArrayVector const & GetFVarPatchArrays(int fvarChannel=0) const\n+Returns the patch arrays for face-varying index buffer data.\n+Definition: glPatchTable.h:98\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarIndexTextures\n+std::vector< GLuint > _fvarIndexTextures\n+Definition: glPatchTable.h:142\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_patchIndexTexture\n+GLuint _patchIndexTexture\n+Definition: glPatchTable.h:133\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarPatchArrays\n+std::vector< PatchArrayVector > _fvarPatchArrays\n+Definition: glPatchTable.h:140\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchIndexTextureBuffer\n+GLuint GetPatchIndexTextureBuffer() const\n+Returns the GL texture buffer containing the patch control vertices.\n+Definition: glPatchTable.h:70\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::\n+GetVaryingPatchIndexTextureBuffer\n+GLuint GetVaryingPatchIndexTextureBuffer() const\n+Returns the GL texture buffer containing the varying control vertices.\n+Definition: glPatchTable.h:90\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_varyingPatchArrays\n+PatchArrayVector _varyingPatchArrays\n+Definition: glPatchTable.h:136\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetVaryingPatchIndexBuffer\n+GLuint GetVaryingPatchIndexBuffer() const\n+Returns the GL index buffer containing the varying control vertices.\n+Definition: glPatchTable.h:85\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetVaryingPatchArrays\n+PatchArrayVector const & GetVaryingPatchArrays() const\n+Returns the patch arrays for varying index buffer data.\n+Definition: glPatchTable.h:80\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_patchParamBuffer\n+GLuint _patchParamBuffer\n+Definition: glPatchTable.h:131\n * opensubdiv\n * osd\n- * glslPatchShaderSource.h\n+ * glPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00863.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00863.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glPatchTable.h File Reference\n+OpenSubdiv: opensubdiv/osd/glLegacyGregoryPatchTable.h File Reference\n \n \n \n \n \n \n \n@@ -83,45 +83,42 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
glPatchTable.h File Reference
\n+
glLegacyGregoryPatchTable.h File Reference
\n
\n
\n
#include "../version.h"
\n+#include "../far/patchTable.h"
\n #include "../osd/nonCopyable.h"
\n-#include "../osd/opengl.h"
\n-#include "../osd/types.h"
\n-#include <vector>
\n+#include "../osd/opengl.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n+\n \n

\n Classes

class  GLPatchTable
class  GLLegacyGregoryPatchTable
 
\n \n \n \n \n \n-\n-\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,30 +5,27 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-glPatchTable.h File Reference\n+glLegacyGregoryPatchTable.h File Reference\n #include \"../version.h\"\n+#include \"../far/patchTable.h\"\n #include \"../osd/nonCopyable.h\"\n #include \"../osd/opengl.h\"\n-#include \"../osd/types.h\"\n-#include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0GLPatchTable\n+class \u00a0GLLegacyGregoryPatchTable\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n-\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * glPatchTable.h\n+ * glLegacyGregoryPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00863.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00863.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00863 = [\n- [\"GLPatchTable\", \"a01229.html\", \"a01229\"]\n+ [\"GLLegacyGregoryPatchTable\", \"a01225.html\", \"a01225\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00863_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00863_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glPatchTable.h Source File\n+OpenSubdiv: opensubdiv/osd/glLegacyGregoryPatchTable.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
glPatchTable.h
\n+
glLegacyGregoryPatchTable.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,172 +107,93 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n-
26#define OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n+
25#ifndef OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H
\n+
26#define OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30#include "../osd/nonCopyable.h"
\n-
31#include "../osd/opengl.h"
\n-
32#include "../osd/types.h"
\n+
30#include "../far/patchTable.h"
\n+
31#include "../osd/nonCopyable.h"
\n+
32#include "../osd/opengl.h"
\n
33
\n-
34#include <vector>
\n-
35
\n-
36namespace OpenSubdiv {
\n-
37namespace OPENSUBDIV_VERSION {
\n+
34namespace OpenSubdiv {
\n+
35namespace OPENSUBDIV_VERSION {
\n+
36
\n+
37namespace Osd {
\n
38
\n-
39namespace Far{
\n-
40 class PatchTable;
\n-
41};
\n-
42
\n-
43namespace Osd {
\n-
44
\n-
45class GLPatchTable : private NonCopyable<GLPatchTable> {
\n-
46public:
\n-
47 typedef GLuint VertexBufferBinding;
\n-
48
\n-
49 ~GLPatchTable();
\n-
50
\n-
51 static GLPatchTable *Create(Far::PatchTable const *farPatchTable,
\n-
52 void *deviceContext = NULL);
\n-
53
\n-
55 PatchArrayVector const &GetPatchArrays() const {
\n-
56 return _patchArrays;
\n-
57 }
\n-
58
\n-
60 GLuint GetPatchIndexBuffer() const {
\n-
61 return _patchIndexBuffer;
\n-
62 }
\n-
63
\n-
65 GLuint GetPatchParamBuffer() const {
\n-
66 return _patchParamBuffer;
\n-
67 }
\n-
68
\n-
70 GLuint GetPatchIndexTextureBuffer() const {
\n-
71 return _patchIndexTexture;
\n-
72 }
\n-
73
\n-
75 GLuint GetPatchParamTextureBuffer() const {
\n-
76 return _patchParamTexture;
\n-
77 }
\n+
39class GLLegacyGregoryPatchTable
\n+
40 : private NonCopyable<GLLegacyGregoryPatchTable> {
\n+
41public:
\n+
42 ~GLLegacyGregoryPatchTable();
\n+
43
\n+
44 static GLLegacyGregoryPatchTable *Create(Far::PatchTable const *patchTable);
\n+
45
\n+
46 void UpdateVertexBuffer(GLuint vbo);
\n+
47
\n+
48 GLuint GetVertexTextureBuffer() const {
\n+
49 return _vertexTextureBuffer;
\n+
50 }
\n+
51
\n+
52 GLuint GetVertexValenceTextureBuffer() const {
\n+
53 return _vertexValenceTextureBuffer;
\n+
54 }
\n+
55
\n+
56 GLuint GetQuadOffsetsTextureBuffer() const {
\n+
57 return _quadOffsetsTextureBuffer;
\n+
58 }
\n+
59
\n+
60 GLuint GetQuadOffsetsBase(Far::PatchDescriptor::Type type) {
\n+
61 if (type == Far::PatchDescriptor::GREGORY_BOUNDARY) {
\n+
62 return _quadOffsetsBase[1];
\n+
63 }
\n+
64 return _quadOffsetsBase[0];
\n+
65 }
\n+
66
\n+
67protected:
\n+
68 GLLegacyGregoryPatchTable();
\n+
69
\n+
70private:
\n+
71 GLuint _vertexTextureBuffer;
\n+
72 GLuint _vertexValenceTextureBuffer;
\n+
73 GLuint _quadOffsetsTextureBuffer;
\n+
74 GLuint _quadOffsetsBase[2]; // gregory, boundaryGregory
\n+
75};
\n+
76
\n+
77
\n
78
\n-
80 PatchArrayVector const &GetVaryingPatchArrays() const {
\n-
81 return _varyingPatchArrays;
\n-
82 }
\n+
79} // end namespace Osd
\n+
80
\n+
81} // end namespace OPENSUBDIV_VERSION
\n+
82using namespace OPENSUBDIV_VERSION;
\n
83
\n-
85 GLuint GetVaryingPatchIndexBuffer() const {
\n-
86 return _varyingIndexBuffer;
\n-
87 }
\n-
88
\n-
90 GLuint GetVaryingPatchIndexTextureBuffer() const {
\n-
91 return _varyingIndexTexture;
\n-
92 }
\n-
93
\n-
95 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }
\n-
96
\n-
98 PatchArrayVector const &GetFVarPatchArrays(int fvarChannel = 0) const {
\n-
99 return _fvarPatchArrays[fvarChannel];
\n-
100 }
\n-
101
\n-
103 GLuint GetFVarPatchIndexBuffer(int fvarChannel = 0) const {
\n-
104 return _fvarIndexBuffers[fvarChannel];
\n-
105 }
\n-
106
\n-
108 GLuint GetFVarPatchIndexTextureBuffer(int fvarChannel = 0) const {
\n-
109 return _fvarIndexTextures[fvarChannel];
\n-
110 }
\n-
111
\n-
113 GLuint GetFVarPatchParamBuffer(int fvarChannel = 0) const {
\n-
114 return _fvarParamBuffers[fvarChannel];
\n-
115 }
\n-
116
\n-
118 GLuint GetFVarPatchParamTextureBuffer(int fvarChannel = 0) const {
\n-
119 return _fvarParamTextures[fvarChannel];
\n-
120 }
\n-
121
\n-
122protected:
\n-
123 GLPatchTable();
\n-
124
\n-
125 // allocate buffers from patchTable
\n-
126 bool allocate(Far::PatchTable const *farPatchTable);
\n-
127
\n-
128 PatchArrayVector _patchArrays;
\n-
129
\n-
130 GLuint _patchIndexBuffer;
\n-
131 GLuint _patchParamBuffer;
\n-
132
\n-
133 GLuint _patchIndexTexture;
\n-
134 GLuint _patchParamTexture;
\n-
135
\n-
136 PatchArrayVector _varyingPatchArrays;
\n-
137 GLuint _varyingIndexBuffer;
\n-
138 GLuint _varyingIndexTexture;
\n-
139
\n-
140 std::vector<PatchArrayVector> _fvarPatchArrays;
\n-
141 std::vector<GLuint> _fvarIndexBuffers;
\n-
142 std::vector<GLuint> _fvarIndexTextures;
\n-
143
\n-
144 std::vector<GLuint> _fvarParamBuffers;
\n-
145 std::vector<GLuint> _fvarParamTextures;
\n-
146};
\n-
147
\n-
148
\n-
149} // end namespace Osd
\n-
150
\n-
151} // end namespace OPENSUBDIV_VERSION
\n-
152using namespace OPENSUBDIV_VERSION;
\n-
153
\n-
154} // end namespace OpenSubdiv
\n-
155
\n-
156#endif // OPENSUBDIV3_OSD_GL_PATCH_TABLE_H
\n+
84} // end namespace OpenSubdiv
\n+
85
\n+
86#endif // OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n+\n+\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchArrays
PatchArrayVector const & GetPatchArrays() const
Returns the patch arrays for vertex index buffer data.
Definition: glPatchTable.h:55
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchIndexBuffer
GLuint GetFVarPatchIndexBuffer(int fvarChannel=0) const
Returns the GL index buffer containing face-varying control vertices.
Definition: glPatchTable.h:103
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchParamTextureBuffer
GLuint GetPatchParamTextureBuffer() const
Returns the GL texture buffer containing the patch parameter.
Definition: glPatchTable.h:75
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::allocate
bool allocate(Far::PatchTable const *farPatchTable)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::Create
static GLPatchTable * Create(Far::PatchTable const *farPatchTable, void *deviceContext=NULL)
\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchParamBuffer
GLuint GetPatchParamBuffer() const
Returns the GL index buffer containing the patch parameter.
Definition: glPatchTable.h:65
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchParamBuffer
GLuint GetFVarPatchParamBuffer(int fvarChannel=0) const
Returns the GL index buffer containing face-varying patch params.
Definition: glPatchTable.h:113
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchIndexTextureBuffer
GLuint GetFVarPatchIndexTextureBuffer(int fvarChannel=0) const
Returns the GL texture buffer containing face-varying control vertices.
Definition: glPatchTable.h:108
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchParamTextureBuffer
GLuint GetFVarPatchParamTextureBuffer(int fvarChannel=0) const
Returns the GL texture buffer containing face-varying patch params.
Definition: glPatchTable.h:118
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchIndexBuffer
GLuint GetPatchIndexBuffer() const
Returns the GL index buffer containing the patch control vertices.
Definition: glPatchTable.h:60
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetNumFVarChannels
int GetNumFVarChannels() const
Returns the number of face-varying channel buffers.
Definition: glPatchTable.h:95
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchArrays
PatchArrayVector const & GetFVarPatchArrays(int fvarChannel=0) const
Returns the patch arrays for face-varying index buffer data.
Definition: glPatchTable.h:98
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarPatchArrays
std::vector< PatchArrayVector > _fvarPatchArrays
Definition: glPatchTable.h:140
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchIndexTextureBuffer
GLuint GetPatchIndexTextureBuffer() const
Returns the GL texture buffer containing the patch control vertices.
Definition: glPatchTable.h:70
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetVaryingPatchIndexTextureBuffer
GLuint GetVaryingPatchIndexTextureBuffer() const
Returns the GL texture buffer containing the varying control vertices.
Definition: glPatchTable.h:90
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetVaryingPatchIndexBuffer
GLuint GetVaryingPatchIndexBuffer() const
Returns the GL index buffer containing the varying control vertices.
Definition: glPatchTable.h:85
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetVaryingPatchArrays
PatchArrayVector const & GetVaryingPatchArrays() const
Returns the patch arrays for varying index buffer data.
Definition: glPatchTable.h:80
\n-\n+\n+\n+\n+\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::Create
static GLLegacyGregoryPatchTable * Create(Far::PatchTable const *patchTable)
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-glPatchTable.h\n+glLegacyGregoryPatchTable.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,249 +30,113 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n- 26#define OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n+ 25#ifndef OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H\n+ 26#define OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H\n 27\n 28#include \"../version.h\"\n 29\n- 30#include \"../osd/nonCopyable.h\"\n- 31#include \"../osd/opengl.h\"\n- 32#include \"../osd/types.h\"\n+ 30#include \"../far/patchTable.h\"\n+ 31#include \"../osd/nonCopyable.h\"\n+ 32#include \"../osd/opengl.h\"\n 33\n- 34#include \n- 35\n- 36namespace OpenSubdiv {\n- 37namespace OPENSUBDIV_VERSION {\n+ 34namespace OpenSubdiv {\n+ 35namespace OPENSUBDIV_VERSION {\n+ 36\n+ 37namespace Osd {\n 38\n- 39namespace Far{\n- 40 class PatchTable;\n- 41};\n- 42\n- 43namespace Osd {\n- 44\n-45class GLPatchTable : private NonCopyable {\n- 46public:\n-47 typedef GLuint VertexBufferBinding;\n- 48\n-49 ~GLPatchTable();\n- 50\n-51 static GLPatchTable *Create(Far::PatchTable const *farPatchTable,\n- 52 void *deviceContext = NULL);\n- 53\n-55 PatchArrayVector const &GetPatchArrays() const {\n- 56 return _patchArrays;\n- 57 }\n- 58\n-60 GLuint GetPatchIndexBuffer() const {\n- 61 return _patchIndexBuffer;\n- 62 }\n- 63\n-65 GLuint GetPatchParamBuffer() const {\n- 66 return _patchParamBuffer;\n- 67 }\n- 68\n-70 GLuint GetPatchIndexTextureBuffer() const {\n- 71 return _patchIndexTexture;\n- 72 }\n- 73\n-75 GLuint GetPatchParamTextureBuffer() const {\n- 76 return _patchParamTexture;\n- 77 }\n+39class GLLegacyGregoryPatchTable\n+ 40 : private NonCopyable {\n+ 41public:\n+42 ~GLLegacyGregoryPatchTable();\n+ 43\n+44 static GLLegacyGregoryPatchTable *Create(Far::PatchTable const *patchTable);\n+ 45\n+46 void UpdateVertexBuffer(GLuint vbo);\n+ 47\n+48 GLuint GetVertexTextureBuffer() const {\n+ 49 return _vertexTextureBuffer;\n+ 50 }\n+ 51\n+52 GLuint GetVertexValenceTextureBuffer() const {\n+ 53 return _vertexValenceTextureBuffer;\n+ 54 }\n+ 55\n+56 GLuint GetQuadOffsetsTextureBuffer() const {\n+ 57 return _quadOffsetsTextureBuffer;\n+ 58 }\n+ 59\n+60 GLuint GetQuadOffsetsBase(Far::PatchDescriptor::Type type) {\n+ 61 if (type == Far::PatchDescriptor::GREGORY_BOUNDARY) {\n+ 62 return _quadOffsetsBase[1];\n+ 63 }\n+ 64 return _quadOffsetsBase[0];\n+ 65 }\n+ 66\n+ 67protected:\n+68 GLLegacyGregoryPatchTable();\n+ 69\n+ 70private:\n+ 71 GLuint _vertexTextureBuffer;\n+ 72 GLuint _vertexValenceTextureBuffer;\n+ 73 GLuint _quadOffsetsTextureBuffer;\n+ 74 GLuint _quadOffsetsBase[2]; // gregory, boundaryGregory\n+ 75};\n+ 76\n+ 77\n 78\n-80 PatchArrayVector const &GetVaryingPatchArrays() const {\n- 81 return _varyingPatchArrays;\n- 82 }\n+ 79} // end namespace Osd\n+ 80\n+ 81} // end namespace OPENSUBDIV_VERSION\n+ 82using namespace OPENSUBDIV_VERSION;\n 83\n-85 GLuint GetVaryingPatchIndexBuffer() const {\n- 86 return _varyingIndexBuffer;\n- 87 }\n- 88\n-90 GLuint GetVaryingPatchIndexTextureBuffer() const {\n- 91 return _varyingIndexTexture;\n- 92 }\n- 93\n-95 int GetNumFVarChannels() const { return (int)_fvarPatchArrays.size(); }\n- 96\n-98 PatchArrayVector const &GetFVarPatchArrays(int fvarChannel = 0) const {\n- 99 return _fvarPatchArrays[fvarChannel];\n- 100 }\n- 101\n-103 GLuint GetFVarPatchIndexBuffer(int fvarChannel = 0) const {\n- 104 return _fvarIndexBuffers[fvarChannel];\n- 105 }\n- 106\n-108 GLuint GetFVarPatchIndexTextureBuffer(int fvarChannel = 0) const {\n- 109 return _fvarIndexTextures[fvarChannel];\n- 110 }\n- 111\n-113 GLuint GetFVarPatchParamBuffer(int fvarChannel = 0) const {\n- 114 return _fvarParamBuffers[fvarChannel];\n- 115 }\n- 116\n-118 GLuint GetFVarPatchParamTextureBuffer(int fvarChannel = 0) const {\n- 119 return _fvarParamTextures[fvarChannel];\n- 120 }\n- 121\n- 122protected:\n-123 GLPatchTable();\n- 124\n- 125 // allocate buffers from patchTable\n-126 bool allocate(Far::PatchTable const *farPatchTable);\n- 127\n-128 PatchArrayVector _patchArrays;\n- 129\n-130 GLuint _patchIndexBuffer;\n-131 GLuint _patchParamBuffer;\n- 132\n-133 GLuint _patchIndexTexture;\n-134 GLuint _patchParamTexture;\n- 135\n-136 PatchArrayVector _varyingPatchArrays;\n-137 GLuint _varyingIndexBuffer;\n-138 GLuint _varyingIndexTexture;\n- 139\n-140 std::vector _fvarPatchArrays;\n-141 std::vector _fvarIndexBuffers;\n-142 std::vector _fvarIndexTextures;\n- 143\n-144 std::vector _fvarParamBuffers;\n-145 std::vector _fvarParamTextures;\n- 146};\n- 147\n- 148\n- 149} // end namespace Osd\n- 150\n- 151} // end namespace OPENSUBDIV_VERSION\n- 152using namespace OPENSUBDIV_VERSION;\n- 153\n- 154} // end namespace OpenSubdiv\n- 155\n- 156#endif // OPENSUBDIV3_OSD_GL_PATCH_TABLE_H\n+ 84} // end namespace OpenSubdiv\n+ 85\n+ 86#endif // OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n-std::vector< PatchArray > PatchArrayVector\n-Definition: types.h:132\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n+Type\n+Definition: patchDescriptor.h:48\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_BOUNDARY\n+@ GREGORY_BOUNDARY\n+Definition: patchDescriptor.h:61\n OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n Container for arrays of parametric patches.\n Definition: patchTable.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable\n-Definition: glPatchTable.h:45\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::VertexBufferBinding\n-GLuint VertexBufferBinding\n-Definition: glPatchTable.h:47\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarIndexBuffers\n-std::vector< GLuint > _fvarIndexBuffers\n-Definition: glPatchTable.h:141\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchArrays\n-PatchArrayVector const & GetPatchArrays() const\n-Returns the patch arrays for vertex index buffer data.\n-Definition: glPatchTable.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchIndexBuffer\n-GLuint GetFVarPatchIndexBuffer(int fvarChannel=0) const\n-Returns the GL index buffer containing face-varying control vertices.\n-Definition: glPatchTable.h:103\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::~GLPatchTable\n-~GLPatchTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarParamBuffers\n-std::vector< GLuint > _fvarParamBuffers\n-Definition: glPatchTable.h:144\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchParamTextureBuffer\n-GLuint GetPatchParamTextureBuffer() const\n-Returns the GL texture buffer containing the patch parameter.\n-Definition: glPatchTable.h:75\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_varyingIndexBuffer\n-GLuint _varyingIndexBuffer\n-Definition: glPatchTable.h:137\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::allocate\n-bool allocate(Far::PatchTable const *farPatchTable)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::Create\n-static GLPatchTable * Create(Far::PatchTable const *farPatchTable, void\n-*deviceContext=NULL)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_varyingIndexTexture\n-GLuint _varyingIndexTexture\n-Definition: glPatchTable.h:138\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarParamTextures\n-std::vector< GLuint > _fvarParamTextures\n-Definition: glPatchTable.h:145\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_patchArrays\n-PatchArrayVector _patchArrays\n-Definition: glPatchTable.h:128\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_patchIndexBuffer\n-GLuint _patchIndexBuffer\n-Definition: glPatchTable.h:130\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchParamBuffer\n-GLuint GetPatchParamBuffer() const\n-Returns the GL index buffer containing the patch parameter.\n-Definition: glPatchTable.h:65\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchParamBuffer\n-GLuint GetFVarPatchParamBuffer(int fvarChannel=0) const\n-Returns the GL index buffer containing face-varying patch params.\n-Definition: glPatchTable.h:113\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_patchParamTexture\n-GLuint _patchParamTexture\n-Definition: glPatchTable.h:134\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GLPatchTable\n-GLPatchTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::\n-GetFVarPatchIndexTextureBuffer\n-GLuint GetFVarPatchIndexTextureBuffer(int fvarChannel=0) const\n-Returns the GL texture buffer containing face-varying control vertices.\n-Definition: glPatchTable.h:108\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::\n-GetFVarPatchParamTextureBuffer\n-GLuint GetFVarPatchParamTextureBuffer(int fvarChannel=0) const\n-Returns the GL texture buffer containing face-varying patch params.\n-Definition: glPatchTable.h:118\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchIndexBuffer\n-GLuint GetPatchIndexBuffer() const\n-Returns the GL index buffer containing the patch control vertices.\n-Definition: glPatchTable.h:60\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetNumFVarChannels\n-int GetNumFVarChannels() const\n-Returns the number of face-varying channel buffers.\n-Definition: glPatchTable.h:95\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetFVarPatchArrays\n-PatchArrayVector const & GetFVarPatchArrays(int fvarChannel=0) const\n-Returns the patch arrays for face-varying index buffer data.\n-Definition: glPatchTable.h:98\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarIndexTextures\n-std::vector< GLuint > _fvarIndexTextures\n-Definition: glPatchTable.h:142\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_patchIndexTexture\n-GLuint _patchIndexTexture\n-Definition: glPatchTable.h:133\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_fvarPatchArrays\n-std::vector< PatchArrayVector > _fvarPatchArrays\n-Definition: glPatchTable.h:140\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetPatchIndexTextureBuffer\n-GLuint GetPatchIndexTextureBuffer() const\n-Returns the GL texture buffer containing the patch control vertices.\n-Definition: glPatchTable.h:70\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::\n-GetVaryingPatchIndexTextureBuffer\n-GLuint GetVaryingPatchIndexTextureBuffer() const\n-Returns the GL texture buffer containing the varying control vertices.\n-Definition: glPatchTable.h:90\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_varyingPatchArrays\n-PatchArrayVector _varyingPatchArrays\n-Definition: glPatchTable.h:136\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetVaryingPatchIndexBuffer\n-GLuint GetVaryingPatchIndexBuffer() const\n-Returns the GL index buffer containing the varying control vertices.\n-Definition: glPatchTable.h:85\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GetVaryingPatchArrays\n-PatchArrayVector const & GetVaryingPatchArrays() const\n-Returns the patch arrays for varying index buffer data.\n-Definition: glPatchTable.h:80\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::_patchParamBuffer\n-GLuint _patchParamBuffer\n-Definition: glPatchTable.h:131\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable\n+Definition: glLegacyGregoryPatchTable.h:40\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n+GetVertexValenceTextureBuffer\n+GLuint GetVertexValenceTextureBuffer() const\n+Definition: glLegacyGregoryPatchTable.h:52\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n+GetQuadOffsetsTextureBuffer\n+GLuint GetQuadOffsetsTextureBuffer() const\n+Definition: glLegacyGregoryPatchTable.h:56\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n+~GLLegacyGregoryPatchTable\n+~GLLegacyGregoryPatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n+GLLegacyGregoryPatchTable\n+GLLegacyGregoryPatchTable()\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n+UpdateVertexBuffer\n+void UpdateVertexBuffer(GLuint vbo)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n+GetQuadOffsetsBase\n+GLuint GetQuadOffsetsBase(Far::PatchDescriptor::Type type)\n+Definition: glLegacyGregoryPatchTable.h:60\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n+GetVertexTextureBuffer\n+GLuint GetVertexTextureBuffer() const\n+Definition: glLegacyGregoryPatchTable.h:48\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::Create\n+static GLLegacyGregoryPatchTable * Create(Far::PatchTable const *patchTable)\n * opensubdiv\n * osd\n- * glPatchTable.h\n+ * glLegacyGregoryPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00866.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00866.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glLegacyGregoryPatchTable.h File Reference\n+OpenSubdiv: opensubdiv/osd/glVertexBuffer.h File Reference\n \n \n \n \n \n \n \n@@ -83,27 +83,27 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
glLegacyGregoryPatchTable.h File Reference
\n+
glVertexBuffer.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../far/patchTable.h"
\n-#include "../osd/nonCopyable.h"
\n-#include "../osd/opengl.h"
\n+#include "../osd/opengl.h"
\n+#include <cstddef>
\n
\n

Go to the source code of this file.

\n \n \n-\n+\n+\n \n

\n Classes

class  GLLegacyGregoryPatchTable
class  GLVertexBuffer
 Concrete vertex buffer class for GLSL subdivision and OpenGL drawing. More...
 
\n \n \n \n \n@@ -112,13 +112,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,27 +5,28 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-glLegacyGregoryPatchTable.h File Reference\n+glVertexBuffer.h File Reference\n #include \"../version.h\"\n-#include \"../far/patchTable.h\"\n-#include \"../osd/nonCopyable.h\"\n #include \"../osd/opengl.h\"\n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0GLLegacyGregoryPatchTable\n+class \u00a0GLVertexBuffer\n+\u00a0 Concrete vertex buffer class for GLSL subdivision and OpenGL drawing.\n+ More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * glLegacyGregoryPatchTable.h\n+ * glVertexBuffer.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00866.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00866.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00866 = [\n- [\"GLLegacyGregoryPatchTable\", \"a01225.html\", \"a01225\"]\n+ [\"GLVertexBuffer\", \"a01237.html\", \"a01237\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00866_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00866_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/glLegacyGregoryPatchTable.h Source File\n+OpenSubdiv: opensubdiv/osd/glVertexBuffer.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
glLegacyGregoryPatchTable.h
\n+
glVertexBuffer.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2015 Pixar
\n+
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,93 +107,76 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H
\n-
26#define OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H
\n+
25#ifndef OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H
\n+
26#define OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30#include "../far/patchTable.h"
\n-
31#include "../osd/nonCopyable.h"
\n-
32#include "../osd/opengl.h"
\n-
33
\n-
34namespace OpenSubdiv {
\n-
35namespace OPENSUBDIV_VERSION {
\n-
36
\n-
37namespace Osd {
\n-
38
\n-
39class GLLegacyGregoryPatchTable
\n-
40 : private NonCopyable<GLLegacyGregoryPatchTable> {
\n-
41public:
\n-
42 ~GLLegacyGregoryPatchTable();
\n-
43
\n-
44 static GLLegacyGregoryPatchTable *Create(Far::PatchTable const *patchTable);
\n-
45
\n-
46 void UpdateVertexBuffer(GLuint vbo);
\n-
47
\n-
48 GLuint GetVertexTextureBuffer() const {
\n-
49 return _vertexTextureBuffer;
\n-
50 }
\n-
51
\n-
52 GLuint GetVertexValenceTextureBuffer() const {
\n-
53 return _vertexValenceTextureBuffer;
\n-
54 }
\n-
55
\n-
56 GLuint GetQuadOffsetsTextureBuffer() const {
\n-
57 return _quadOffsetsTextureBuffer;
\n-
58 }
\n-
59
\n-
60 GLuint GetQuadOffsetsBase(Far::PatchDescriptor::Type type) {
\n-
61 if (type == Far::PatchDescriptor::GREGORY_BOUNDARY) {
\n-
62 return _quadOffsetsBase[1];
\n-
63 }
\n-
64 return _quadOffsetsBase[0];
\n-
65 }
\n+
30#include "../osd/opengl.h"
\n+
31#include <cstddef>
\n+
32
\n+
33namespace OpenSubdiv {
\n+
34namespace OPENSUBDIV_VERSION {
\n+
35
\n+
36namespace Osd {
\n+
37
\n+
44class GLVertexBuffer {
\n+
45public:
\n+
47 static GLVertexBuffer * Create(int numElements, int numVertices,
\n+
48 void *deviceContext = NULL);
\n+
49
\n+
51 ~GLVertexBuffer();
\n+
52
\n+
55 void UpdateData(const float *src, int startVertex, int numVertices,
\n+
56 void *deviceContext = NULL);
\n+
57
\n+
59 int GetNumElements() const;
\n+
60
\n+
62 int GetNumVertices() const;
\n+
63
\n+
65 GLuint BindVBO(void *deviceContext = NULL);
\n
66
\n
67protected:
\n-
68 GLLegacyGregoryPatchTable();
\n-
69
\n-
70private:
\n-
71 GLuint _vertexTextureBuffer;
\n-
72 GLuint _vertexValenceTextureBuffer;
\n-
73 GLuint _quadOffsetsTextureBuffer;
\n-
74 GLuint _quadOffsetsBase[2]; // gregory, boundaryGregory
\n-
75};
\n-
76
\n-
77
\n-
78
\n-
79} // end namespace Osd
\n+
69 GLVertexBuffer(int numElements, int numVertices);
\n+
70
\n+
73 bool allocate();
\n+
74
\n+
75private:
\n+
76 int _numElements;
\n+
77 int _numVertices;
\n+
78 GLuint _vbo;
\n+
79};
\n
80
\n-
81} // end namespace OPENSUBDIV_VERSION
\n-
82using namespace OPENSUBDIV_VERSION;
\n-
83
\n-
84} // end namespace OpenSubdiv
\n+
81} // end namespace Osd
\n+
82
\n+
83} // end namespace OPENSUBDIV_VERSION
\n+
84using namespace OPENSUBDIV_VERSION;
\n
85
\n-
86#endif // OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H
\n+
86} // end namespace OpenSubdiv
\n+
87
\n+
88#endif // OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H
\n \n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n-\n-\n-\n-\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::Create
static GLLegacyGregoryPatchTable * Create(Far::PatchTable const *patchTable)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer
Concrete vertex buffer class for GLSL subdivision and OpenGL drawing.
Definition: glVertexBuffer.h:44
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::BindVBO
GLuint BindVBO(void *deviceContext=NULL)
Returns the GL buffer object.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GetNumElements
int GetNumElements() const
Returns how many elements defined in this vertex buffer.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::UpdateData
void UpdateData(const float *src, int startVertex, int numVertices, void *deviceContext=NULL)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GLVertexBuffer
GLVertexBuffer(int numElements, int numVertices)
Constructor.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GetNumVertices
int GetNumVertices() const
Returns how many vertices allocated in this vertex buffer.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::Create
static GLVertexBuffer * Create(int numElements, int numVertices, void *deviceContext=NULL)
Creator. Returns NULL if error.
\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-glLegacyGregoryPatchTable.h\n+glVertexBuffer.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2015 Pixar\n+ 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,113 +30,88 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H\n- 26#define OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H\n+ 25#ifndef OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H\n+ 26#define OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H\n 27\n 28#include \"../version.h\"\n 29\n- 30#include \"../far/patchTable.h\"\n- 31#include \"../osd/nonCopyable.h\"\n- 32#include \"../osd/opengl.h\"\n- 33\n- 34namespace OpenSubdiv {\n- 35namespace OPENSUBDIV_VERSION {\n- 36\n- 37namespace Osd {\n- 38\n-39class GLLegacyGregoryPatchTable\n- 40 : private NonCopyable {\n- 41public:\n-42 ~GLLegacyGregoryPatchTable();\n- 43\n-44 static GLLegacyGregoryPatchTable *Create(Far::PatchTable const *patchTable);\n- 45\n-46 void UpdateVertexBuffer(GLuint vbo);\n- 47\n-48 GLuint GetVertexTextureBuffer() const {\n- 49 return _vertexTextureBuffer;\n- 50 }\n- 51\n-52 GLuint GetVertexValenceTextureBuffer() const {\n- 53 return _vertexValenceTextureBuffer;\n- 54 }\n- 55\n-56 GLuint GetQuadOffsetsTextureBuffer() const {\n- 57 return _quadOffsetsTextureBuffer;\n- 58 }\n- 59\n-60 GLuint GetQuadOffsetsBase(Far::PatchDescriptor::Type type) {\n- 61 if (type == Far::PatchDescriptor::GREGORY_BOUNDARY) {\n- 62 return _quadOffsetsBase[1];\n- 63 }\n- 64 return _quadOffsetsBase[0];\n- 65 }\n+ 30#include \"../osd/opengl.h\"\n+ 31#include \n+ 32\n+ 33namespace OpenSubdiv {\n+ 34namespace OPENSUBDIV_VERSION {\n+ 35\n+ 36namespace Osd {\n+ 37\n+44class GLVertexBuffer {\n+ 45public:\n+47 static GLVertexBuffer * Create(int numElements, int numVertices,\n+ 48 void *deviceContext = NULL);\n+ 49\n+51 ~GLVertexBuffer();\n+ 52\n+55 void UpdateData(const float *src, int startVertex, int numVertices,\n+ 56 void *deviceContext = NULL);\n+ 57\n+59 int GetNumElements() const;\n+ 60\n+62 int GetNumVertices() const;\n+ 63\n+65 GLuint BindVBO(void *deviceContext = NULL);\n 66\n 67protected:\n-68 GLLegacyGregoryPatchTable();\n- 69\n- 70private:\n- 71 GLuint _vertexTextureBuffer;\n- 72 GLuint _vertexValenceTextureBuffer;\n- 73 GLuint _quadOffsetsTextureBuffer;\n- 74 GLuint _quadOffsetsBase[2]; // gregory, boundaryGregory\n- 75};\n- 76\n- 77\n- 78\n- 79} // end namespace Osd\n+69 GLVertexBuffer(int numElements, int numVertices);\n+ 70\n+73 bool allocate();\n+ 74\n+ 75private:\n+ 76 int _numElements;\n+ 77 int _numVertices;\n+ 78 GLuint _vbo;\n+ 79};\n 80\n- 81} // end namespace OPENSUBDIV_VERSION\n- 82using namespace OPENSUBDIV_VERSION;\n- 83\n- 84} // end namespace OpenSubdiv\n+ 81} // end namespace Osd\n+ 82\n+ 83} // end namespace OPENSUBDIV_VERSION\n+ 84using namespace OPENSUBDIV_VERSION;\n 85\n- 86#endif // OPENSUBDIV3_OSD_GL_LEGACY_GREGORY_PATCH_TABLE_H\n+ 86} // end namespace OpenSubdiv\n+ 87\n+ 88#endif // OPENSUBDIV3_OSD_GL_VERTEX_BUFFER_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::Type\n-Type\n-Definition: patchDescriptor.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GREGORY_BOUNDARY\n-@ GREGORY_BOUNDARY\n-Definition: patchDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n-Container for arrays of parametric patches.\n-Definition: patchTable.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable\n-Definition: glLegacyGregoryPatchTable.h:40\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n-GetVertexValenceTextureBuffer\n-GLuint GetVertexValenceTextureBuffer() const\n-Definition: glLegacyGregoryPatchTable.h:52\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n-GetQuadOffsetsTextureBuffer\n-GLuint GetQuadOffsetsTextureBuffer() const\n-Definition: glLegacyGregoryPatchTable.h:56\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n-~GLLegacyGregoryPatchTable\n-~GLLegacyGregoryPatchTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n-GLLegacyGregoryPatchTable\n-GLLegacyGregoryPatchTable()\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n-UpdateVertexBuffer\n-void UpdateVertexBuffer(GLuint vbo)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n-GetQuadOffsetsBase\n-GLuint GetQuadOffsetsBase(Far::PatchDescriptor::Type type)\n-Definition: glLegacyGregoryPatchTable.h:60\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::\n-GetVertexTextureBuffer\n-GLuint GetVertexTextureBuffer() const\n-Definition: glLegacyGregoryPatchTable.h:48\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::Create\n-static GLLegacyGregoryPatchTable * Create(Far::PatchTable const *patchTable)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer\n+Concrete vertex buffer class for GLSL subdivision and OpenGL drawing.\n+Definition: glVertexBuffer.h:44\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::~GLVertexBuffer\n+~GLVertexBuffer()\n+Destructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::BindVBO\n+GLuint BindVBO(void *deviceContext=NULL)\n+Returns the GL buffer object.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GetNumElements\n+int GetNumElements() const\n+Returns how many elements defined in this vertex buffer.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::UpdateData\n+void UpdateData(const float *src, int startVertex, int numVertices, void\n+*deviceContext=NULL)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GLVertexBuffer\n+GLVertexBuffer(int numElements, int numVertices)\n+Constructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GetNumVertices\n+int GetNumVertices() const\n+Returns how many vertices allocated in this vertex buffer.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::Create\n+static GLVertexBuffer * Create(int numElements, int numVertices, void\n+*deviceContext=NULL)\n+Creator. Returns NULL if error.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::allocate\n+bool allocate()\n * opensubdiv\n * osd\n- * glLegacyGregoryPatchTable.h\n+ * glVertexBuffer.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00869.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00869.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/tbbKernel.h File Reference\n+OpenSubdiv: opensubdiv/osd/tbbEvaluator.h File Reference\n \n \n \n \n \n \n \n@@ -81,51 +81,44 @@\n \n \n \n
\n
\n-Namespaces |\n-Functions
\n-
tbbKernel.h File Reference
\n+Classes |\n+Namespaces
\n+
tbbEvaluator.h File Reference
\n \n
\n
#include "../version.h"
\n-#include "../far/patchDescriptor.h"
\n-#include "../far/patchParam.h"
\n+#include "../osd/bufferDescriptor.h"
\n+#include "../osd/types.h"
\n+#include <cstddef>
\n
\n

Go to the source code of this file.

\n \n+\n+\n+\n+

\n+Classes

class  TbbEvaluator
 
\n \n \n \n \n \n \n \n-

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n

\n-Functions

void TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, int start, int end)
 
void TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, int start, int end)
 
void TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv, BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const &dstDvvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, float const *duuWeights, float const *duvWeights, float const *dvvWeights, int start, int end)
 
void TbbEvalPatches (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrayBuffer, const int *patchIndexBuffer, const PatchParam *patchParamBuffer)
 
void TbbEvalPatches (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv, BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const &dstDvvDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrayBuffer, const int *patchIndexBuffer, const PatchParam *patchParamBuffer)
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,62 +4,28 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-Namespaces | Functions\n-tbbKernel.h File Reference\n+Classes | Namespaces\n+tbbEvaluator.h File Reference\n #include \"../version.h\"\n-#include \"../far/patchDescriptor.h\"\n-#include \"../far/patchParam.h\"\n+#include \"../osd/bufferDescriptor.h\"\n+#include \"../osd/types.h\"\n+#include \n Go_to_the_source_code_of_this_file.\n+ Classes\n+class \u00a0TbbEvaluator\n+\u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n- Functions\n-void\u00a0TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n- *dst, BufferDescriptor const &dstDesc, int const *sizes, int const\n- *offsets, int const *indices, float const *weights, int start, int end)\n-\u00a0\n-void\u00a0TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n- *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n- const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int\n- const *sizes, int const *offsets, int const *indices, float const\n- *weights, float const *duWeights, float const *dvWeights, int start, int\n- end)\n-\u00a0\n-void\u00a0TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n- *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n- const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float\n- *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv,\n- BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const\n- &dstDvvDesc, int const *sizes, int const *offsets, int const *indices,\n- float const *weights, float const *duWeights, float const *dvWeights,\n- float const *duuWeights, float const *duvWeights, float const\n- *dvvWeights, int start, int end)\n-\u00a0\n-void\u00a0TbbEvalPatches (float const *src, BufferDescriptor const &srcDesc, float\n- *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n- const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int\n- numPatchCoords, const PatchCoord *patchCoords, const PatchArray\n- *patchArrayBuffer, const int *patchIndexBuffer, const PatchParam\n- *patchParamBuffer)\n-\u00a0\n-void\u00a0TbbEvalPatches (float const *src, BufferDescriptor const &srcDesc, float\n- *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n- const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float\n- *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv,\n- BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const\n- &dstDvvDesc, int numPatchCoords, const PatchCoord *patchCoords, const\n- PatchArray *patchArrayBuffer, const int *patchIndexBuffer, const\n- PatchParam *patchParamBuffer)\n-\u00a0\n * opensubdiv\n * osd\n- * tbbKernel.h\n+ * tbbEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00869.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00869.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,7 +1,3 @@\n var a00869 = [\n- [\"TbbEvalPatches\", \"a00869.html#aa0c3c264a43eeddc5cb3a58b006d3d32\", null],\n- [\"TbbEvalPatches\", \"a00869.html#a20e08be34d6183b6943ed7a1586eca4a\", null],\n- [\"TbbEvalStencils\", \"a00869.html#aa65ed860a2fa5935d99129a898a8f0b6\", null],\n- [\"TbbEvalStencils\", \"a00869.html#af23c35537952bfdd493441e5c42ff2ed\", null],\n- [\"TbbEvalStencils\", \"a00869.html#a4a6a0fcc01638f2a0316b32ad158b485\", null]\n+ [\"TbbEvaluator\", \"a01309.html\", null]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00869_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00869_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/tbbKernel.h Source File\n+OpenSubdiv: opensubdiv/osd/tbbEvaluator.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
tbbKernel.h
\n+
tbbEvaluator.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2013 Pixar
\n+
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,116 +107,502 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_TBB_KERNEL_H
\n-
26#define OPENSUBDIV3_OSD_TBB_KERNEL_H
\n+
25#ifndef OPENSUBDIV3_OSD_TBB_EVALUATOR_H
\n+
26#define OPENSUBDIV3_OSD_TBB_EVALUATOR_H
\n
27
\n
28#include "../version.h"
\n-
29#include "../far/patchDescriptor.h"
\n-
30#include "../far/patchParam.h"
\n+
29#include "../osd/bufferDescriptor.h"
\n+
30#include "../osd/types.h"
\n
31
\n-
32namespace OpenSubdiv {
\n-
33namespace OPENSUBDIV_VERSION {
\n-
34
\n-
35namespace Osd {
\n+
32#include <cstddef>
\n+
33
\n+
34namespace OpenSubdiv {
\n+
35namespace OPENSUBDIV_VERSION {
\n
36
\n-
37struct PatchArray;
\n-
38struct PatchCoord;
\n-
39struct PatchParam;
\n-
40struct BufferDescriptor;
\n-
41
\n-
42void
\n-
43TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n-
44 float * dst, BufferDescriptor const &dstDesc,
\n-
45 int const * sizes,
\n-
46 int const * offsets,
\n-
47 int const * indices,
\n-
48 float const * weights,
\n-
49 int start, int end);
\n-
50
\n-
51void
\n-
52TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n-
53 float * dst, BufferDescriptor const &dstDesc,
\n-
54 float * dstDu, BufferDescriptor const &dstDuDesc,
\n-
55 float * dstDv, BufferDescriptor const &dstDvDesc,
\n-
56 int const * sizes,
\n-
57 int const * offsets,
\n-
58 int const * indices,
\n-
59 float const * weights,
\n-
60 float const * duWeights,
\n-
61 float const * dvWeights,
\n-
62 int start, int end);
\n-
63
\n-
64void
\n-
65TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n-
66 float * dst, BufferDescriptor const &dstDesc,
\n-
67 float * dstDu, BufferDescriptor const &dstDuDesc,
\n-
68 float * dstDv, BufferDescriptor const &dstDvDesc,
\n-
69 float * dstDuu, BufferDescriptor const &dstDuuDesc,
\n-
70 float * dstDuv, BufferDescriptor const &dstDuvDesc,
\n-
71 float * dstDvv, BufferDescriptor const &dstDvvDesc,
\n-
72 int const * sizes,
\n-
73 int const * offsets,
\n-
74 int const * indices,
\n-
75 float const * weights,
\n-
76 float const * duWeights,
\n-
77 float const * dvWeights,
\n-
78 float const * duuWeights,
\n-
79 float const * duvWeights,
\n-
80 float const * dvvWeights,
\n-
81 int start, int end);
\n-
82
\n-
83void
\n-
84TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc,
\n-
85 float *dst, BufferDescriptor const &dstDesc,
\n-
86 float *dstDu, BufferDescriptor const &dstDuDesc,
\n-
87 float *dstDv, BufferDescriptor const &dstDvDesc,
\n-
88 int numPatchCoords,
\n-
89 const PatchCoord *patchCoords,
\n-
90 const PatchArray *patchArrayBuffer,
\n-
91 const int *patchIndexBuffer,
\n-
92 const PatchParam *patchParamBuffer);
\n-
93
\n-
94void
\n-
95TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc,
\n-
96 float *dst, BufferDescriptor const &dstDesc,
\n-
97 float *dstDu, BufferDescriptor const &dstDuDesc,
\n-
98 float *dstDv, BufferDescriptor const &dstDvDesc,
\n-
99 float *dstDuu, BufferDescriptor const &dstDuuDesc,
\n-
100 float *dstDuv, BufferDescriptor const &dstDuvDesc,
\n-
101 float *dstDvv, BufferDescriptor const &dstDvvDesc,
\n-
102 int numPatchCoords,
\n-
103 const PatchCoord *patchCoords,
\n-
104 const PatchArray *patchArrayBuffer,
\n-
105 const int *patchIndexBuffer,
\n-
106 const PatchParam *patchParamBuffer);
\n-
107
\n-
108} // end namespace Osd
\n-
109
\n-
110} // end namespace OPENSUBDIV_VERSION
\n-
111using namespace OPENSUBDIV_VERSION;
\n-
112
\n-
113} // end namespace OpenSubdiv
\n-
114
\n-
115#endif // OPENSUBDIV3_OSD_TBB_KERNEL_H
\n+
37namespace Osd {
\n+
38
\n+
39class TbbEvaluator {
\n+
40public:
\n+
46
\n+
71 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
72 static bool EvalStencils(
\n+
73 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
74 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
75 STENCIL_TABLE const *stencilTable,
\n+
76 TbbEvaluator const *instance = NULL,
\n+
77 void *deviceContext = NULL) {
\n+
78
\n+
79 (void)instance; // unused
\n+
80 (void)deviceContext; // unused
\n+
81
\n+
82 if (stencilTable->GetNumStencils() == 0)
\n+
83 return false;
\n+
84
\n+
85 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
86 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
87 &stencilTable->GetSizes()[0],
\n+
88 &stencilTable->GetOffsets()[0],
\n+
89 &stencilTable->GetControlIndices()[0],
\n+
90 &stencilTable->GetWeights()[0],
\n+
91 /*start = */ 0,
\n+
92 /*end = */ stencilTable->GetNumStencils());
\n+
93 }
\n+
94
\n+
121 static bool EvalStencils(
\n+
122 const float *src, BufferDescriptor const &srcDesc,
\n+
123 float *dst, BufferDescriptor const &dstDesc,
\n+
124 const int * sizes,
\n+
125 const int * offsets,
\n+
126 const int * indices,
\n+
127 const float * weights,
\n+
128 int start, int end);
\n+
129
\n+
167 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
168 static bool EvalStencils(
\n+
169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
171 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
172 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
173 STENCIL_TABLE const *stencilTable,
\n+
174 const TbbEvaluator *instance = NULL,
\n+
175 void * deviceContext = NULL) {
\n+
176
\n+
177 (void)instance; // unused
\n+
178 (void)deviceContext; // unused
\n+
179
\n+
180 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
181 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
182 duBuffer->BindCpuBuffer(), duDesc,
\n+
183 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
184 &stencilTable->GetSizes()[0],
\n+
185 &stencilTable->GetOffsets()[0],
\n+
186 &stencilTable->GetControlIndices()[0],
\n+
187 &stencilTable->GetWeights()[0],
\n+
188 &stencilTable->GetDuWeights()[0],
\n+
189 &stencilTable->GetDvWeights()[0],
\n+
190 /*start = */ 0,
\n+
191 /*end = */ stencilTable->GetNumStencils());
\n+
192 }
\n+
193
\n+
234 static bool EvalStencils(
\n+
235 const float *src, BufferDescriptor const &srcDesc,
\n+
236 float *dst, BufferDescriptor const &dstDesc,
\n+
237 float *du, BufferDescriptor const &duDesc,
\n+
238 float *dv, BufferDescriptor const &dvDesc,
\n+
239 const int * sizes,
\n+
240 const int * offsets,
\n+
241 const int * indices,
\n+
242 const float * weights,
\n+
243 const float * duWeights,
\n+
244 const float * dvWeights,
\n+
245 int start, int end);
\n+
246
\n+
302 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
303 static bool EvalStencils(
\n+
304 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
305 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
306 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
307 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
308 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
309 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
310 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
311 STENCIL_TABLE const *stencilTable,
\n+
312 const TbbEvaluator *instance = NULL,
\n+
313 void * deviceContext = NULL) {
\n+
314
\n+
315 (void)instance; // unused
\n+
316 (void)deviceContext; // unused
\n+
317
\n+
318 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
319 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
320 duBuffer->BindCpuBuffer(), duDesc,
\n+
321 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
322 duuBuffer->BindCpuBuffer(), duuDesc,
\n+
323 duvBuffer->BindCpuBuffer(), duvDesc,
\n+
324 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n+
325 &stencilTable->GetSizes()[0],
\n+
326 &stencilTable->GetOffsets()[0],
\n+
327 &stencilTable->GetControlIndices()[0],
\n+
328 &stencilTable->GetWeights()[0],
\n+
329 &stencilTable->GetDuWeights()[0],
\n+
330 &stencilTable->GetDvWeights()[0],
\n+
331 &stencilTable->GetDuuWeights()[0],
\n+
332 &stencilTable->GetDuvWeights()[0],
\n+
333 &stencilTable->GetDvvWeights()[0],
\n+
334 /*start = */ 0,
\n+
335 /*end = */ stencilTable->GetNumStencils());
\n+
336 }
\n+
337
\n+
399 static bool EvalStencils(
\n+
400 const float *src, BufferDescriptor const &srcDesc,
\n+
401 float *dst, BufferDescriptor const &dstDesc,
\n+
402 float *du, BufferDescriptor const &duDesc,
\n+
403 float *dv, BufferDescriptor const &dvDesc,
\n+
404 float *duu, BufferDescriptor const &duuDesc,
\n+
405 float *duv, BufferDescriptor const &duvDesc,
\n+
406 float *dvv, BufferDescriptor const &dvvDesc,
\n+
407 const int * sizes,
\n+
408 const int * offsets,
\n+
409 const int * indices,
\n+
410 const float * weights,
\n+
411 const float * duWeights,
\n+
412 const float * dvWeights,
\n+
413 const float * duuWeights,
\n+
414 const float * duvWeights,
\n+
415 const float * dvvWeights,
\n+
416 int start, int end);
\n+
417
\n+
423
\n+
452 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
453 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
454 static bool EvalPatches(
\n+
455 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
456 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
457 int numPatchCoords,
\n+
458 PATCHCOORD_BUFFER *patchCoords,
\n+
459 PATCH_TABLE *patchTable,
\n+
460 TbbEvaluator const *instance = NULL,
\n+
461 void * deviceContext = NULL) {
\n+
462
\n+
463 (void)instance; // unused
\n+
464 (void)deviceContext; // unused
\n+
465
\n+
466 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
467 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
468 numPatchCoords,
\n+
469 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
470 patchTable->GetPatchArrayBuffer(),
\n+
471 patchTable->GetPatchIndexBuffer(),
\n+
472 patchTable->GetPatchParamBuffer());
\n+
473 }
\n+
474
\n+
515 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
516 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
517 static bool EvalPatches(
\n+
518 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
519 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
520 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
521 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
522 int numPatchCoords,
\n+
523 PATCHCOORD_BUFFER *patchCoords,
\n+
524 PATCH_TABLE *patchTable,
\n+
525 TbbEvaluator const *instance = NULL,
\n+
526 void * deviceContext = NULL) {
\n+
527
\n+
528 (void)instance; // unused
\n+
529 (void)deviceContext; // unused
\n+
530
\n+
531 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.
\n+
532 // ideally all buffer classes should have templated by datatype
\n+
533 // so that downcast isn't needed there.
\n+
534 // (e.g. Osd::CpuBuffer<PatchCoord> )
\n+
535 //
\n+
536 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
537 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
538 duBuffer->BindCpuBuffer(), duDesc,
\n+
539 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
540 numPatchCoords,
\n+
541 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
542 patchTable->GetPatchArrayBuffer(),
\n+
543 patchTable->GetPatchIndexBuffer(),
\n+
544 patchTable->GetPatchParamBuffer());
\n+
545 }
\n+
546
\n+
605 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
606 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
607 static bool EvalPatches(
\n+
608 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
609 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
610 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
611 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
612 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
613 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
614 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
615 int numPatchCoords,
\n+
616 PATCHCOORD_BUFFER *patchCoords,
\n+
617 PATCH_TABLE *patchTable,
\n+
618 TbbEvaluator const *instance = NULL,
\n+
619 void * deviceContext = NULL) {
\n+
620
\n+
621 (void)instance; // unused
\n+
622 (void)deviceContext; // unused
\n+
623
\n+
624 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.
\n+
625 // ideally all buffer classes should have templated by datatype
\n+
626 // so that downcast isn't needed there.
\n+
627 // (e.g. Osd::CpuBuffer<PatchCoord> )
\n+
628 //
\n+
629 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
630 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
631 duBuffer->BindCpuBuffer(), duDesc,
\n+
632 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
633 duuBuffer->BindCpuBuffer(), duuDesc,
\n+
634 duvBuffer->BindCpuBuffer(), duvDesc,
\n+
635 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n+
636 numPatchCoords,
\n+
637 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
638 patchTable->GetPatchArrayBuffer(),
\n+
639 patchTable->GetPatchIndexBuffer(),
\n+
640 patchTable->GetPatchParamBuffer());
\n+
641 }
\n+
642
\n+
670 static bool EvalPatches(
\n+
671 const float *src, BufferDescriptor const &srcDesc,
\n+
672 float *dst, BufferDescriptor const &dstDesc,
\n+
673 int numPatchCoords,
\n+
674 const PatchCoord *patchCoords,
\n+
675 const PatchArray *patchArrays,
\n+
676 const int *patchIndexBuffer,
\n+
677 const PatchParam *patchParamBuffer);
\n+
678
\n+
716 static bool EvalPatches(
\n+
717 const float *src, BufferDescriptor const &srcDesc,
\n+
718 float *dst, BufferDescriptor const &dstDesc,
\n+
719 float *du, BufferDescriptor const &duDesc,
\n+
720 float *dv, BufferDescriptor const &dvDesc,
\n+
721 int numPatchCoords,
\n+
722 PatchCoord const *patchCoords,
\n+
723 PatchArray const *patchArrays,
\n+
724 const int *patchIndexBuffer,
\n+
725 PatchParam const *patchParamBuffer);
\n+
726
\n+
779 static bool EvalPatches(
\n+
780 const float *src, BufferDescriptor const &srcDesc,
\n+
781 float *dst, BufferDescriptor const &dstDesc,
\n+
782 float *du, BufferDescriptor const &duDesc,
\n+
783 float *dv, BufferDescriptor const &dvDesc,
\n+
784 float *duu, BufferDescriptor const &duuDesc,
\n+
785 float *duv, BufferDescriptor const &duvDesc,
\n+
786 float *dvv, BufferDescriptor const &dvvDesc,
\n+
787 int numPatchCoords,
\n+
788 PatchCoord const *patchCoords,
\n+
789 PatchArray const *patchArrays,
\n+
790 const int *patchIndexBuffer,
\n+
791 PatchParam const *patchParamBuffer);
\n+
792
\n+
821 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
822 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
823 static bool EvalPatchesVarying(
\n+
824 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
825 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
826 int numPatchCoords,
\n+
827 PATCHCOORD_BUFFER *patchCoords,
\n+
828 PATCH_TABLE *patchTable,
\n+
829 TbbEvaluator const *instance = NULL,
\n+
830 void * deviceContext = NULL) {
\n+
831
\n+
832 (void)instance; // unused
\n+
833 (void)deviceContext; // unused
\n+
834
\n+
835 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
836 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
837 numPatchCoords,
\n+
838 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
839 patchTable->GetVaryingPatchArrayBuffer(),
\n+
840 patchTable->GetVaryingPatchIndexBuffer(),
\n+
841 patchTable->GetPatchParamBuffer());
\n+
842 }
\n+
843
\n+
884 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
885 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
886 static bool EvalPatchesVarying(
\n+
887 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
888 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
889 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
890 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
891 int numPatchCoords,
\n+
892 PATCHCOORD_BUFFER *patchCoords,
\n+
893 PATCH_TABLE *patchTable,
\n+
894 TbbEvaluator const *instance = NULL,
\n+
895 void * deviceContext = NULL) {
\n+
896
\n+
897 (void)instance; // unused
\n+
898 (void)deviceContext; // unused
\n+
899
\n+
900 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
901 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
902 duBuffer->BindCpuBuffer(), duDesc,
\n+
903 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
904 numPatchCoords,
\n+
905 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
906 patchTable->GetVaryingPatchArrayBuffer(),
\n+
907 patchTable->GetVaryingPatchIndexBuffer(),
\n+
908 patchTable->GetPatchParamBuffer());
\n+
909 }
\n+
910
\n+
969 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
970 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
971 static bool EvalPatchesVarying(
\n+
972 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
973 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
974 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
975 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
976 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
977 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
978 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
979 int numPatchCoords,
\n+
980 PATCHCOORD_BUFFER *patchCoords,
\n+
981 PATCH_TABLE *patchTable,
\n+
982 TbbEvaluator const *instance = NULL,
\n+
983 void * deviceContext = NULL) {
\n+
984
\n+
985 (void)instance; // unused
\n+
986 (void)deviceContext; // unused
\n+
987
\n+
988 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
989 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
990 duBuffer->BindCpuBuffer(), duDesc,
\n+
991 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
992 duuBuffer->BindCpuBuffer(), duuDesc,
\n+
993 duvBuffer->BindCpuBuffer(), duvDesc,
\n+
994 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n+
995 numPatchCoords,
\n+
996 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
997 patchTable->GetVaryingPatchArrayBuffer(),
\n+
998 patchTable->GetVaryingPatchIndexBuffer(),
\n+
999 patchTable->GetPatchParamBuffer());
\n+
1000 }
\n+
1001
\n+
1032 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1033 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1034 static bool EvalPatchesFaceVarying(
\n+
1035 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1036 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1037 int numPatchCoords,
\n+
1038 PATCHCOORD_BUFFER *patchCoords,
\n+
1039 PATCH_TABLE *patchTable,
\n+
1040 int fvarChannel,
\n+
1041 TbbEvaluator const *instance = NULL,
\n+
1042 void * deviceContext = NULL) {
\n+
1043
\n+
1044 (void)instance; // unused
\n+
1045 (void)deviceContext; // unused
\n+
1046
\n+
1047 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
1048 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
1049 numPatchCoords,
\n+
1050 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
1051 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
1052 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1053 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
1054 }
\n+
1055
\n+
1098 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1099 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1100 static bool EvalPatchesFaceVarying(
\n+
1101 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1102 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1103 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1104 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1105 int numPatchCoords,
\n+
1106 PATCHCOORD_BUFFER *patchCoords,
\n+
1107 PATCH_TABLE *patchTable,
\n+
1108 int fvarChannel,
\n+
1109 TbbEvaluator const *instance = NULL,
\n+
1110 void * deviceContext = NULL) {
\n+
1111
\n+
1112 (void)instance; // unused
\n+
1113 (void)deviceContext; // unused
\n+
1114
\n+
1115 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
1116 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
1117 duBuffer->BindCpuBuffer(), duDesc,
\n+
1118 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
1119 numPatchCoords,
\n+
1120 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
1121 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
1122 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1123 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
1124 }
\n+
1125
\n+
1186 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1187 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1188 static bool EvalPatchesFaceVarying(
\n+
1189 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1190 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1191 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1192 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1193 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1194 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1195 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1196 int numPatchCoords,
\n+
1197 PATCHCOORD_BUFFER *patchCoords,
\n+
1198 PATCH_TABLE *patchTable,
\n+
1199 int fvarChannel,
\n+
1200 TbbEvaluator const *instance = NULL,
\n+
1201 void * deviceContext = NULL) {
\n+
1202
\n+
1203 (void)instance; // unused
\n+
1204 (void)deviceContext; // unused
\n+
1205
\n+
1206 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
1207 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
1208 duBuffer->BindCpuBuffer(), duDesc,
\n+
1209 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
1210 duuBuffer->BindCpuBuffer(), duuDesc,
\n+
1211 duvBuffer->BindCpuBuffer(), duvDesc,
\n+
1212 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n+
1213 numPatchCoords,
\n+
1214 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
1215 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
1216 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1217 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
1218 }
\n+
1219
\n+
1225
\n+
1227 static void Synchronize(void *deviceContext = NULL);
\n+
1228
\n+
1234 static void SetNumThreads(int numThreads);
\n+
1235};
\n+
1236
\n+
1237
\n+
1238} // end namespace Osd
\n+
1239
\n+
1240} // end namespace OPENSUBDIV_VERSION
\n+
1241using namespace OPENSUBDIV_VERSION;
\n+
1242
\n+
1243} // end namespace OpenSubdiv
\n+
1244
\n+
1245
\n+
1246#endif // OPENSUBDIV3_OSD_TBB_EVALUATOR_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalPatches
void TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrayBuffer, const int *patchIndexBuffer, const PatchParam *patchParamBuffer)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalStencils
void TbbEvalStencils(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, int start, int end)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function. This function has a same signature as other device kernels hav...
Definition: tbbEvaluator.h:72
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:454
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, const TbbEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: tbbEvaluator.h:303
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, int start, int end)
Static eval stencils function which takes raw CPU pointers for input and output.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:886
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: tbbEvaluator.h:607
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw CPU pointers for input and output.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, const float *duuWeights, const float *duvWeights, const float *dvvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw CPU pointers for input and output.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer, PatchParam const *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::Synchronize
static void Synchronize(void *deviceContext=NULL)
synchronize all asynchronous computation invoked on this device.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:971
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:1100
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: tbbEvaluator.h:517
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:1034
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer, PatchParam const *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::SetNumThreads
static void SetNumThreads(int numThreads)
initialize tbb task schedular (optional: client may use tbb::task_scheduler_init)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrays, const int *patchIndexBuffer, const PatchParam *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const TbbEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: tbbEvaluator.h:168
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:1188
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:823
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord
Coordinates set on a patch table.
Definition: types.h:42
\n \n \n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-tbbKernel.h\n+tbbEvaluator.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2013 Pixar\n+ 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,126 +30,660 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_TBB_KERNEL_H\n- 26#define OPENSUBDIV3_OSD_TBB_KERNEL_H\n+ 25#ifndef OPENSUBDIV3_OSD_TBB_EVALUATOR_H\n+ 26#define OPENSUBDIV3_OSD_TBB_EVALUATOR_H\n 27\n 28#include \"../version.h\"\n- 29#include \"../far/patchDescriptor.h\"\n- 30#include \"../far/patchParam.h\"\n+ 29#include \"../osd/bufferDescriptor.h\"\n+ 30#include \"../osd/types.h\"\n 31\n- 32namespace OpenSubdiv {\n- 33namespace OPENSUBDIV_VERSION {\n- 34\n- 35namespace Osd {\n+ 32#include \n+ 33\n+ 34namespace OpenSubdiv {\n+ 35namespace OPENSUBDIV_VERSION {\n 36\n- 37struct PatchArray;\n- 38struct PatchCoord;\n- 39struct PatchParam;\n- 40struct BufferDescriptor;\n- 41\n- 42void\n-43TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n- 44 float * dst, BufferDescriptor const &dstDesc,\n- 45 int const * sizes,\n- 46 int const * offsets,\n- 47 int const * indices,\n- 48 float const * weights,\n- 49 int start, int end);\n- 50\n- 51void\n-52TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n- 53 float * dst, BufferDescriptor const &dstDesc,\n- 54 float * dstDu, BufferDescriptor const &dstDuDesc,\n- 55 float * dstDv, BufferDescriptor const &dstDvDesc,\n- 56 int const * sizes,\n- 57 int const * offsets,\n- 58 int const * indices,\n- 59 float const * weights,\n- 60 float const * duWeights,\n- 61 float const * dvWeights,\n- 62 int start, int end);\n- 63\n- 64void\n-65TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n- 66 float * dst, BufferDescriptor const &dstDesc,\n- 67 float * dstDu, BufferDescriptor const &dstDuDesc,\n- 68 float * dstDv, BufferDescriptor const &dstDvDesc,\n- 69 float * dstDuu, BufferDescriptor const &dstDuuDesc,\n- 70 float * dstDuv, BufferDescriptor const &dstDuvDesc,\n- 71 float * dstDvv, BufferDescriptor const &dstDvvDesc,\n- 72 int const * sizes,\n- 73 int const * offsets,\n- 74 int const * indices,\n- 75 float const * weights,\n- 76 float const * duWeights,\n- 77 float const * dvWeights,\n- 78 float const * duuWeights,\n- 79 float const * duvWeights,\n- 80 float const * dvvWeights,\n- 81 int start, int end);\n- 82\n- 83void\n-84TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc,\n- 85 float *dst, BufferDescriptor const &dstDesc,\n- 86 float *dstDu, BufferDescriptor const &dstDuDesc,\n- 87 float *dstDv, BufferDescriptor const &dstDvDesc,\n- 88 int numPatchCoords,\n- 89 const PatchCoord *patchCoords,\n- 90 const PatchArray *patchArrayBuffer,\n- 91 const int *patchIndexBuffer,\n- 92 const PatchParam *patchParamBuffer);\n- 93\n- 94void\n-95TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc,\n- 96 float *dst, BufferDescriptor const &dstDesc,\n- 97 float *dstDu, BufferDescriptor const &dstDuDesc,\n- 98 float *dstDv, BufferDescriptor const &dstDvDesc,\n- 99 float *dstDuu, BufferDescriptor const &dstDuuDesc,\n- 100 float *dstDuv, BufferDescriptor const &dstDuvDesc,\n- 101 float *dstDvv, BufferDescriptor const &dstDvvDesc,\n- 102 int numPatchCoords,\n- 103 const PatchCoord *patchCoords,\n- 104 const PatchArray *patchArrayBuffer,\n- 105 const int *patchIndexBuffer,\n- 106 const PatchParam *patchParamBuffer);\n- 107\n- 108} // end namespace Osd\n- 109\n- 110} // end namespace OPENSUBDIV_VERSION\n- 111using namespace OPENSUBDIV_VERSION;\n- 112\n- 113} // end namespace OpenSubdiv\n- 114\n- 115#endif // OPENSUBDIV3_OSD_TBB_KERNEL_H\n+ 37namespace Osd {\n+ 38\n+39class TbbEvaluator {\n+ 40public:\n+ 46\n+ 71 template \n+72 static bool EvalStencils(\n+ 73 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 74 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 75 STENCIL_TABLE const *stencilTable,\n+ 76 TbbEvaluator const *instance = NULL,\n+ 77 void *deviceContext = NULL) {\n+ 78\n+ 79 (void)instance; // unused\n+ 80 (void)deviceContext; // unused\n+ 81\n+ 82 if (stencilTable->GetNumStencils() == 0)\n+ 83 return false;\n+ 84\n+ 85 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 86 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 87 &stencilTable->GetSizes()[0],\n+ 88 &stencilTable->GetOffsets()[0],\n+ 89 &stencilTable->GetControlIndices()[0],\n+ 90 &stencilTable->GetWeights()[0],\n+ 91 /*start = */ 0,\n+ 92 /*end = */ stencilTable->GetNumStencils());\n+ 93 }\n+ 94\n+121 static bool EvalStencils(\n+ 122 const float *src, BufferDescriptor const &srcDesc,\n+ 123 float *dst, BufferDescriptor const &dstDesc,\n+ 124 const int * sizes,\n+ 125 const int * offsets,\n+ 126 const int * indices,\n+ 127 const float * weights,\n+ 128 int start, int end);\n+ 129\n+ 167 template \n+168 static bool EvalStencils(\n+ 169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 171 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 172 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 173 STENCIL_TABLE const *stencilTable,\n+ 174 const TbbEvaluator *instance = NULL,\n+ 175 void * deviceContext = NULL) {\n+ 176\n+ 177 (void)instance; // unused\n+ 178 (void)deviceContext; // unused\n+ 179\n+ 180 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 181 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 182 duBuffer->BindCpuBuffer(), duDesc,\n+ 183 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 184 &stencilTable->GetSizes()[0],\n+ 185 &stencilTable->GetOffsets()[0],\n+ 186 &stencilTable->GetControlIndices()[0],\n+ 187 &stencilTable->GetWeights()[0],\n+ 188 &stencilTable->GetDuWeights()[0],\n+ 189 &stencilTable->GetDvWeights()[0],\n+ 190 /*start = */ 0,\n+ 191 /*end = */ stencilTable->GetNumStencils());\n+ 192 }\n+ 193\n+234 static bool EvalStencils(\n+ 235 const float *src, BufferDescriptor const &srcDesc,\n+ 236 float *dst, BufferDescriptor const &dstDesc,\n+ 237 float *du, BufferDescriptor const &duDesc,\n+ 238 float *dv, BufferDescriptor const &dvDesc,\n+ 239 const int * sizes,\n+ 240 const int * offsets,\n+ 241 const int * indices,\n+ 242 const float * weights,\n+ 243 const float * duWeights,\n+ 244 const float * dvWeights,\n+ 245 int start, int end);\n+ 246\n+ 302 template \n+303 static bool EvalStencils(\n+ 304 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 305 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 306 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 307 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 308 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 309 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 310 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 311 STENCIL_TABLE const *stencilTable,\n+ 312 const TbbEvaluator *instance = NULL,\n+ 313 void * deviceContext = NULL) {\n+ 314\n+ 315 (void)instance; // unused\n+ 316 (void)deviceContext; // unused\n+ 317\n+ 318 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 319 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 320 duBuffer->BindCpuBuffer(), duDesc,\n+ 321 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 322 duuBuffer->BindCpuBuffer(), duuDesc,\n+ 323 duvBuffer->BindCpuBuffer(), duvDesc,\n+ 324 dvvBuffer->BindCpuBuffer(), dvvDesc,\n+ 325 &stencilTable->GetSizes()[0],\n+ 326 &stencilTable->GetOffsets()[0],\n+ 327 &stencilTable->GetControlIndices()[0],\n+ 328 &stencilTable->GetWeights()[0],\n+ 329 &stencilTable->GetDuWeights()[0],\n+ 330 &stencilTable->GetDvWeights()[0],\n+ 331 &stencilTable->GetDuuWeights()[0],\n+ 332 &stencilTable->GetDuvWeights()[0],\n+ 333 &stencilTable->GetDvvWeights()[0],\n+ 334 /*start = */ 0,\n+ 335 /*end = */ stencilTable->GetNumStencils());\n+ 336 }\n+ 337\n+399 static bool EvalStencils(\n+ 400 const float *src, BufferDescriptor const &srcDesc,\n+ 401 float *dst, BufferDescriptor const &dstDesc,\n+ 402 float *du, BufferDescriptor const &duDesc,\n+ 403 float *dv, BufferDescriptor const &dvDesc,\n+ 404 float *duu, BufferDescriptor const &duuDesc,\n+ 405 float *duv, BufferDescriptor const &duvDesc,\n+ 406 float *dvv, BufferDescriptor const &dvvDesc,\n+ 407 const int * sizes,\n+ 408 const int * offsets,\n+ 409 const int * indices,\n+ 410 const float * weights,\n+ 411 const float * duWeights,\n+ 412 const float * dvWeights,\n+ 413 const float * duuWeights,\n+ 414 const float * duvWeights,\n+ 415 const float * dvvWeights,\n+ 416 int start, int end);\n+ 417\n+ 423\n+ 452 template \n+454 static bool EvalPatches(\n+ 455 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 456 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 457 int numPatchCoords,\n+ 458 PATCHCOORD_BUFFER *patchCoords,\n+ 459 PATCH_TABLE *patchTable,\n+ 460 TbbEvaluator const *instance = NULL,\n+ 461 void * deviceContext = NULL) {\n+ 462\n+ 463 (void)instance; // unused\n+ 464 (void)deviceContext; // unused\n+ 465\n+ 466 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 467 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 468 numPatchCoords,\n+ 469 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 470 patchTable->GetPatchArrayBuffer(),\n+ 471 patchTable->GetPatchIndexBuffer(),\n+ 472 patchTable->GetPatchParamBuffer());\n+ 473 }\n+ 474\n+ 515 template \n+517 static bool EvalPatches(\n+ 518 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 519 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 520 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 521 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 522 int numPatchCoords,\n+ 523 PATCHCOORD_BUFFER *patchCoords,\n+ 524 PATCH_TABLE *patchTable,\n+ 525 TbbEvaluator const *instance = NULL,\n+ 526 void * deviceContext = NULL) {\n+ 527\n+ 528 (void)instance; // unused\n+ 529 (void)deviceContext; // unused\n+ 530\n+ 531 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.\n+ 532 // ideally all buffer classes should have templated by datatype\n+ 533 // so that downcast isn't needed there.\n+ 534 // (e.g. Osd::CpuBuffer )\n+ 535 //\n+ 536 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 537 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 538 duBuffer->BindCpuBuffer(), duDesc,\n+ 539 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 540 numPatchCoords,\n+ 541 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 542 patchTable->GetPatchArrayBuffer(),\n+ 543 patchTable->GetPatchIndexBuffer(),\n+ 544 patchTable->GetPatchParamBuffer());\n+ 545 }\n+ 546\n+ 605 template \n+607 static bool EvalPatches(\n+ 608 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 609 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 610 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 611 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 612 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 613 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 614 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 615 int numPatchCoords,\n+ 616 PATCHCOORD_BUFFER *patchCoords,\n+ 617 PATCH_TABLE *patchTable,\n+ 618 TbbEvaluator const *instance = NULL,\n+ 619 void * deviceContext = NULL) {\n+ 620\n+ 621 (void)instance; // unused\n+ 622 (void)deviceContext; // unused\n+ 623\n+ 624 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.\n+ 625 // ideally all buffer classes should have templated by datatype\n+ 626 // so that downcast isn't needed there.\n+ 627 // (e.g. Osd::CpuBuffer )\n+ 628 //\n+ 629 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 630 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 631 duBuffer->BindCpuBuffer(), duDesc,\n+ 632 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 633 duuBuffer->BindCpuBuffer(), duuDesc,\n+ 634 duvBuffer->BindCpuBuffer(), duvDesc,\n+ 635 dvvBuffer->BindCpuBuffer(), dvvDesc,\n+ 636 numPatchCoords,\n+ 637 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 638 patchTable->GetPatchArrayBuffer(),\n+ 639 patchTable->GetPatchIndexBuffer(),\n+ 640 patchTable->GetPatchParamBuffer());\n+ 641 }\n+ 642\n+670 static bool EvalPatches(\n+ 671 const float *src, BufferDescriptor const &srcDesc,\n+ 672 float *dst, BufferDescriptor const &dstDesc,\n+ 673 int numPatchCoords,\n+ 674 const PatchCoord *patchCoords,\n+ 675 const PatchArray *patchArrays,\n+ 676 const int *patchIndexBuffer,\n+ 677 const PatchParam *patchParamBuffer);\n+ 678\n+716 static bool EvalPatches(\n+ 717 const float *src, BufferDescriptor const &srcDesc,\n+ 718 float *dst, BufferDescriptor const &dstDesc,\n+ 719 float *du, BufferDescriptor const &duDesc,\n+ 720 float *dv, BufferDescriptor const &dvDesc,\n+ 721 int numPatchCoords,\n+ 722 PatchCoord const *patchCoords,\n+ 723 PatchArray const *patchArrays,\n+ 724 const int *patchIndexBuffer,\n+ 725 PatchParam const *patchParamBuffer);\n+ 726\n+779 static bool EvalPatches(\n+ 780 const float *src, BufferDescriptor const &srcDesc,\n+ 781 float *dst, BufferDescriptor const &dstDesc,\n+ 782 float *du, BufferDescriptor const &duDesc,\n+ 783 float *dv, BufferDescriptor const &dvDesc,\n+ 784 float *duu, BufferDescriptor const &duuDesc,\n+ 785 float *duv, BufferDescriptor const &duvDesc,\n+ 786 float *dvv, BufferDescriptor const &dvvDesc,\n+ 787 int numPatchCoords,\n+ 788 PatchCoord const *patchCoords,\n+ 789 PatchArray const *patchArrays,\n+ 790 const int *patchIndexBuffer,\n+ 791 PatchParam const *patchParamBuffer);\n+ 792\n+ 821 template \n+823 static bool EvalPatchesVarying(\n+ 824 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 825 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 826 int numPatchCoords,\n+ 827 PATCHCOORD_BUFFER *patchCoords,\n+ 828 PATCH_TABLE *patchTable,\n+ 829 TbbEvaluator const *instance = NULL,\n+ 830 void * deviceContext = NULL) {\n+ 831\n+ 832 (void)instance; // unused\n+ 833 (void)deviceContext; // unused\n+ 834\n+ 835 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 836 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 837 numPatchCoords,\n+ 838 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 839 patchTable->GetVaryingPatchArrayBuffer(),\n+ 840 patchTable->GetVaryingPatchIndexBuffer(),\n+ 841 patchTable->GetPatchParamBuffer());\n+ 842 }\n+ 843\n+ 884 template \n+886 static bool EvalPatchesVarying(\n+ 887 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 888 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 889 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 890 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 891 int numPatchCoords,\n+ 892 PATCHCOORD_BUFFER *patchCoords,\n+ 893 PATCH_TABLE *patchTable,\n+ 894 TbbEvaluator const *instance = NULL,\n+ 895 void * deviceContext = NULL) {\n+ 896\n+ 897 (void)instance; // unused\n+ 898 (void)deviceContext; // unused\n+ 899\n+ 900 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 901 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 902 duBuffer->BindCpuBuffer(), duDesc,\n+ 903 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 904 numPatchCoords,\n+ 905 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 906 patchTable->GetVaryingPatchArrayBuffer(),\n+ 907 patchTable->GetVaryingPatchIndexBuffer(),\n+ 908 patchTable->GetPatchParamBuffer());\n+ 909 }\n+ 910\n+ 969 template \n+971 static bool EvalPatchesVarying(\n+ 972 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 973 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 974 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 975 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 976 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 977 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 978 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 979 int numPatchCoords,\n+ 980 PATCHCOORD_BUFFER *patchCoords,\n+ 981 PATCH_TABLE *patchTable,\n+ 982 TbbEvaluator const *instance = NULL,\n+ 983 void * deviceContext = NULL) {\n+ 984\n+ 985 (void)instance; // unused\n+ 986 (void)deviceContext; // unused\n+ 987\n+ 988 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 989 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 990 duBuffer->BindCpuBuffer(), duDesc,\n+ 991 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 992 duuBuffer->BindCpuBuffer(), duuDesc,\n+ 993 duvBuffer->BindCpuBuffer(), duvDesc,\n+ 994 dvvBuffer->BindCpuBuffer(), dvvDesc,\n+ 995 numPatchCoords,\n+ 996 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 997 patchTable->GetVaryingPatchArrayBuffer(),\n+ 998 patchTable->GetVaryingPatchIndexBuffer(),\n+ 999 patchTable->GetPatchParamBuffer());\n+ 1000 }\n+ 1001\n+ 1032 template \n+1034 static bool EvalPatchesFaceVarying(\n+ 1035 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1036 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1037 int numPatchCoords,\n+ 1038 PATCHCOORD_BUFFER *patchCoords,\n+ 1039 PATCH_TABLE *patchTable,\n+ 1040 int fvarChannel,\n+ 1041 TbbEvaluator const *instance = NULL,\n+ 1042 void * deviceContext = NULL) {\n+ 1043\n+ 1044 (void)instance; // unused\n+ 1045 (void)deviceContext; // unused\n+ 1046\n+ 1047 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 1048 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 1049 numPatchCoords,\n+ 1050 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 1051 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 1052 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1053 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 1054 }\n+ 1055\n+ 1098 template \n+1100 static bool EvalPatchesFaceVarying(\n+ 1101 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1102 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1103 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1104 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1105 int numPatchCoords,\n+ 1106 PATCHCOORD_BUFFER *patchCoords,\n+ 1107 PATCH_TABLE *patchTable,\n+ 1108 int fvarChannel,\n+ 1109 TbbEvaluator const *instance = NULL,\n+ 1110 void * deviceContext = NULL) {\n+ 1111\n+ 1112 (void)instance; // unused\n+ 1113 (void)deviceContext; // unused\n+ 1114\n+ 1115 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 1116 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 1117 duBuffer->BindCpuBuffer(), duDesc,\n+ 1118 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 1119 numPatchCoords,\n+ 1120 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 1121 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 1122 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1123 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 1124 }\n+ 1125\n+ 1186 template \n+1188 static bool EvalPatchesFaceVarying(\n+ 1189 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1190 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1191 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1192 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1193 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1194 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1195 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1196 int numPatchCoords,\n+ 1197 PATCHCOORD_BUFFER *patchCoords,\n+ 1198 PATCH_TABLE *patchTable,\n+ 1199 int fvarChannel,\n+ 1200 TbbEvaluator const *instance = NULL,\n+ 1201 void * deviceContext = NULL) {\n+ 1202\n+ 1203 (void)instance; // unused\n+ 1204 (void)deviceContext; // unused\n+ 1205\n+ 1206 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 1207 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 1208 duBuffer->BindCpuBuffer(), duDesc,\n+ 1209 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 1210 duuBuffer->BindCpuBuffer(), duuDesc,\n+ 1211 duvBuffer->BindCpuBuffer(), duvDesc,\n+ 1212 dvvBuffer->BindCpuBuffer(), dvvDesc,\n+ 1213 numPatchCoords,\n+ 1214 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 1215 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 1216 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1217 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 1218 }\n+ 1219\n+ 1225\n+1227 static void Synchronize(void *deviceContext = NULL);\n+ 1228\n+1234 static void SetNumThreads(int numThreads);\n+ 1235};\n+ 1236\n+ 1237\n+ 1238} // end namespace Osd\n+ 1239\n+ 1240} // end namespace OPENSUBDIV_VERSION\n+ 1241using namespace OPENSUBDIV_VERSION;\n+ 1242\n+ 1243} // end namespace OpenSubdiv\n+ 1244\n+ 1245\n+ 1246#endif // OPENSUBDIV3_OSD_TBB_EVALUATOR_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalPatches\n-void TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc, float\n-*dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const\n-&dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int\n-numPatchCoords, const PatchCoord *patchCoords, const PatchArray\n-*patchArrayBuffer, const int *patchIndexBuffer, const PatchParam\n-*patchParamBuffer)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalStencils\n-void TbbEvalStencils(float const *src, BufferDescriptor const &srcDesc, float\n-*dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets,\n-int const *indices, float const *weights, int start, int end)\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n BufferDescriptor is a struct which describes buffer elements in interleaved\n data buffers....\n Definition: bufferDescriptor.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator\n+Definition: tbbEvaluator.h:39\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n+const *stencilTable, TbbEvaluator const *instance=NULL, void\n+*deviceContext=NULL)\n+Generic static eval stencils function. This function has a same signature as\n+other device kernels hav...\n+Definition: tbbEvaluator.h:72\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const\n+*instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: tbbEvaluator.h:454\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n+const TbbEvaluator *instance=NULL, void *deviceContext=NULL)\n+Generic static eval stencils function with derivatives. This function has a\n+same signature as other d...\n+Definition: tbbEvaluator.h:303\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n+static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int\n+*offsets, const int *indices, const float *weights, int start, int end)\n+Static eval stencils function which takes raw CPU pointers for input and\n+output.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void\n+*deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: tbbEvaluator.h:886\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void\n+*deviceContext=NULL)\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: tbbEvaluator.h:607\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n+static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int\n+*offsets, const int *indices, const float *weights, const float *duWeights,\n+const float *dvWeights, int start, int end)\n+Static eval stencils function with derivatives, which takes raw CPU pointers\n+for input and output.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n+static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n+BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n+float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int\n+*offsets, const int *indices, const float *weights, const float *duWeights,\n+const float *dvWeights, const float *duuWeights, const float *duvWeights, const\n+float *dvvWeights, int start, int end)\n+Static eval stencils function with derivatives, which takes raw CPU pointers\n+for input and output.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n+static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n+BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n+float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord\n+const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer,\n+PatchParam const *patchParamBuffer)\n+Static limit eval function. It takes an array of PatchCoord and evaluate limit\n+values on given PatchT...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::Synchronize\n+static void Synchronize(void *deviceContext=NULL)\n+synchronize all asynchronous computation invoked on this device.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const\n+*instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: tbbEvaluator.h:971\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, TbbEvaluator const\n+*instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: tbbEvaluator.h:1100\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: tbbEvaluator.h:517\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n+fvarChannel, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: tbbEvaluator.h:1034\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n+static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords,\n+PatchCoord const *patchCoords, PatchArray const *patchArrays, const int\n+*patchIndexBuffer, PatchParam const *patchParamBuffer)\n+Static limit eval function. It takes an array of PatchCoord and evaluate limit\n+values on given PatchT...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::SetNumThreads\n+static void SetNumThreads(int numThreads)\n+initialize tbb task schedular (optional: client may use tbb::\n+task_scheduler_init)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n+static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const\n+PatchCoord *patchCoords, const PatchArray *patchArrays, const int\n+*patchIndexBuffer, const PatchParam *patchParamBuffer)\n+Static limit eval function. It takes an array of PatchCoord and evaluate limit\n+values on given PatchT...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const\n+TbbEvaluator *instance=NULL, void *deviceContext=NULL)\n+Generic static eval stencils function with derivatives. This function has a\n+same signature as other d...\n+Definition: tbbEvaluator.h:168\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n+TbbEvaluator const *instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: tbbEvaluator.h:1188\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n+TbbEvaluator const *instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: tbbEvaluator.h:823\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord\n Coordinates set on a patch table.\n Definition: types.h:42\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArray\n Definition: types.h:66\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParam\n Definition: types.h:127\n * opensubdiv\n * osd\n- * tbbKernel.h\n+ * tbbEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00872.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00872.html", "unified_diff": "@@ -88,15 +88,15 @@\n Classes |\n Namespaces \n
cpuGLVertexBuffer.h File Reference
\n \n
\n
#include "../version.h"
\n #include <cstddef>
\n-#include "../osd/opengl.h"
\n+#include "../osd/opengl.h"
\n
\n

Go to the source code of this file.

\n \n \n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00875.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00875.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/tbbEvaluator.h File Reference\n+OpenSubdiv: opensubdiv/osd/tbbKernel.h File Reference\n \n \n \n \n \n \n \n@@ -81,44 +81,51 @@\n \n \n \n
\n
\n-Classes |\n-Namespaces
\n-
tbbEvaluator.h File Reference
\n+Namespaces |\n+Functions
\n+
tbbKernel.h File Reference
\n \n
\n
#include "../version.h"
\n-#include "../osd/bufferDescriptor.h"
\n-#include "../osd/types.h"
\n-#include <cstddef>
\n+#include "../far/patchDescriptor.h"
\n+#include "../far/patchParam.h"
\n
\n

Go to the source code of this file.

\n

\n Classes

class  CpuGLVertexBuffer
 Concrete vertex buffer class for cpu subdivision and OpenGL drawing. More...
\n-\n-\n-\n-

\n-Classes

class  TbbEvaluator
 
\n \n \n \n \n \n \n \n+

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n

\n+Functions

void TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, int start, int end)
 
void TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, int start, int end)
 
void TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv, BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const &dstDvvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, float const *duuWeights, float const *duvWeights, float const *dvvWeights, int start, int end)
 
void TbbEvalPatches (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrayBuffer, const int *patchIndexBuffer, const PatchParam *patchParamBuffer)
 
void TbbEvalPatches (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv, BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const &dstDvvDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrayBuffer, const int *patchIndexBuffer, const PatchParam *patchParamBuffer)
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,28 +4,62 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-Classes | Namespaces\n-tbbEvaluator.h File Reference\n+Namespaces | Functions\n+tbbKernel.h File Reference\n #include \"../version.h\"\n-#include \"../osd/bufferDescriptor.h\"\n-#include \"../osd/types.h\"\n-#include \n+#include \"../far/patchDescriptor.h\"\n+#include \"../far/patchParam.h\"\n Go_to_the_source_code_of_this_file.\n- Classes\n-class \u00a0TbbEvaluator\n-\u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n+ Functions\n+void\u00a0TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n+ *dst, BufferDescriptor const &dstDesc, int const *sizes, int const\n+ *offsets, int const *indices, float const *weights, int start, int end)\n+\u00a0\n+void\u00a0TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n+ *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n+ const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int\n+ const *sizes, int const *offsets, int const *indices, float const\n+ *weights, float const *duWeights, float const *dvWeights, int start, int\n+ end)\n+\u00a0\n+void\u00a0TbbEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n+ *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n+ const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float\n+ *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv,\n+ BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const\n+ &dstDvvDesc, int const *sizes, int const *offsets, int const *indices,\n+ float const *weights, float const *duWeights, float const *dvWeights,\n+ float const *duuWeights, float const *duvWeights, float const\n+ *dvvWeights, int start, int end)\n+\u00a0\n+void\u00a0TbbEvalPatches (float const *src, BufferDescriptor const &srcDesc, float\n+ *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n+ const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int\n+ numPatchCoords, const PatchCoord *patchCoords, const PatchArray\n+ *patchArrayBuffer, const int *patchIndexBuffer, const PatchParam\n+ *patchParamBuffer)\n+\u00a0\n+void\u00a0TbbEvalPatches (float const *src, BufferDescriptor const &srcDesc, float\n+ *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n+ const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float\n+ *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv,\n+ BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const\n+ &dstDvvDesc, int numPatchCoords, const PatchCoord *patchCoords, const\n+ PatchArray *patchArrayBuffer, const int *patchIndexBuffer, const\n+ PatchParam *patchParamBuffer)\n+\u00a0\n * opensubdiv\n * osd\n- * tbbEvaluator.h\n+ * tbbKernel.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00875.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00875.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,7 @@\n var a00875 = [\n- [\"TbbEvaluator\", \"a01309.html\", null]\n+ [\"TbbEvalPatches\", \"a00875.html#aa0c3c264a43eeddc5cb3a58b006d3d32\", null],\n+ [\"TbbEvalPatches\", \"a00875.html#a20e08be34d6183b6943ed7a1586eca4a\", null],\n+ [\"TbbEvalStencils\", \"a00875.html#aa65ed860a2fa5935d99129a898a8f0b6\", null],\n+ [\"TbbEvalStencils\", \"a00875.html#af23c35537952bfdd493441e5c42ff2ed\", null],\n+ [\"TbbEvalStencils\", \"a00875.html#a4a6a0fcc01638f2a0316b32ad158b485\", null]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00875_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00875_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/tbbEvaluator.h Source File\n+OpenSubdiv: opensubdiv/osd/tbbKernel.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
tbbEvaluator.h
\n+
tbbKernel.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2015 Pixar
\n+
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,502 +107,116 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_TBB_EVALUATOR_H
\n-
26#define OPENSUBDIV3_OSD_TBB_EVALUATOR_H
\n+
25#ifndef OPENSUBDIV3_OSD_TBB_KERNEL_H
\n+
26#define OPENSUBDIV3_OSD_TBB_KERNEL_H
\n
27
\n
28#include "../version.h"
\n-
29#include "../osd/bufferDescriptor.h"
\n-
30#include "../osd/types.h"
\n+
29#include "../far/patchDescriptor.h"
\n+
30#include "../far/patchParam.h"
\n
31
\n-
32#include <cstddef>
\n-
33
\n-
34namespace OpenSubdiv {
\n-
35namespace OPENSUBDIV_VERSION {
\n+
32namespace OpenSubdiv {
\n+
33namespace OPENSUBDIV_VERSION {
\n+
34
\n+
35namespace Osd {
\n
36
\n-
37namespace Osd {
\n-
38
\n-
39class TbbEvaluator {
\n-
40public:
\n-
46
\n-
71 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
72 static bool EvalStencils(
\n-
73 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
74 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
75 STENCIL_TABLE const *stencilTable,
\n-
76 TbbEvaluator const *instance = NULL,
\n-
77 void *deviceContext = NULL) {
\n-
78
\n-
79 (void)instance; // unused
\n-
80 (void)deviceContext; // unused
\n-
81
\n-
82 if (stencilTable->GetNumStencils() == 0)
\n-
83 return false;
\n-
84
\n-
85 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
86 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
87 &stencilTable->GetSizes()[0],
\n-
88 &stencilTable->GetOffsets()[0],
\n-
89 &stencilTable->GetControlIndices()[0],
\n-
90 &stencilTable->GetWeights()[0],
\n-
91 /*start = */ 0,
\n-
92 /*end = */ stencilTable->GetNumStencils());
\n-
93 }
\n-
94
\n-
121 static bool EvalStencils(
\n-
122 const float *src, BufferDescriptor const &srcDesc,
\n-
123 float *dst, BufferDescriptor const &dstDesc,
\n-
124 const int * sizes,
\n-
125 const int * offsets,
\n-
126 const int * indices,
\n-
127 const float * weights,
\n-
128 int start, int end);
\n-
129
\n-
167 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
168 static bool EvalStencils(
\n-
169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
171 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
172 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
173 STENCIL_TABLE const *stencilTable,
\n-
174 const TbbEvaluator *instance = NULL,
\n-
175 void * deviceContext = NULL) {
\n-
176
\n-
177 (void)instance; // unused
\n-
178 (void)deviceContext; // unused
\n-
179
\n-
180 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
181 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
182 duBuffer->BindCpuBuffer(), duDesc,
\n-
183 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
184 &stencilTable->GetSizes()[0],
\n-
185 &stencilTable->GetOffsets()[0],
\n-
186 &stencilTable->GetControlIndices()[0],
\n-
187 &stencilTable->GetWeights()[0],
\n-
188 &stencilTable->GetDuWeights()[0],
\n-
189 &stencilTable->GetDvWeights()[0],
\n-
190 /*start = */ 0,
\n-
191 /*end = */ stencilTable->GetNumStencils());
\n-
192 }
\n-
193
\n-
234 static bool EvalStencils(
\n-
235 const float *src, BufferDescriptor const &srcDesc,
\n-
236 float *dst, BufferDescriptor const &dstDesc,
\n-
237 float *du, BufferDescriptor const &duDesc,
\n-
238 float *dv, BufferDescriptor const &dvDesc,
\n-
239 const int * sizes,
\n-
240 const int * offsets,
\n-
241 const int * indices,
\n-
242 const float * weights,
\n-
243 const float * duWeights,
\n-
244 const float * dvWeights,
\n-
245 int start, int end);
\n-
246
\n-
302 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
303 static bool EvalStencils(
\n-
304 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
305 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
306 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
307 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
308 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
309 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
310 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
311 STENCIL_TABLE const *stencilTable,
\n-
312 const TbbEvaluator *instance = NULL,
\n-
313 void * deviceContext = NULL) {
\n-
314
\n-
315 (void)instance; // unused
\n-
316 (void)deviceContext; // unused
\n-
317
\n-
318 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
319 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
320 duBuffer->BindCpuBuffer(), duDesc,
\n-
321 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
322 duuBuffer->BindCpuBuffer(), duuDesc,
\n-
323 duvBuffer->BindCpuBuffer(), duvDesc,
\n-
324 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n-
325 &stencilTable->GetSizes()[0],
\n-
326 &stencilTable->GetOffsets()[0],
\n-
327 &stencilTable->GetControlIndices()[0],
\n-
328 &stencilTable->GetWeights()[0],
\n-
329 &stencilTable->GetDuWeights()[0],
\n-
330 &stencilTable->GetDvWeights()[0],
\n-
331 &stencilTable->GetDuuWeights()[0],
\n-
332 &stencilTable->GetDuvWeights()[0],
\n-
333 &stencilTable->GetDvvWeights()[0],
\n-
334 /*start = */ 0,
\n-
335 /*end = */ stencilTable->GetNumStencils());
\n-
336 }
\n-
337
\n-
399 static bool EvalStencils(
\n-
400 const float *src, BufferDescriptor const &srcDesc,
\n-
401 float *dst, BufferDescriptor const &dstDesc,
\n-
402 float *du, BufferDescriptor const &duDesc,
\n-
403 float *dv, BufferDescriptor const &dvDesc,
\n-
404 float *duu, BufferDescriptor const &duuDesc,
\n-
405 float *duv, BufferDescriptor const &duvDesc,
\n-
406 float *dvv, BufferDescriptor const &dvvDesc,
\n-
407 const int * sizes,
\n-
408 const int * offsets,
\n-
409 const int * indices,
\n-
410 const float * weights,
\n-
411 const float * duWeights,
\n-
412 const float * dvWeights,
\n-
413 const float * duuWeights,
\n-
414 const float * duvWeights,
\n-
415 const float * dvvWeights,
\n-
416 int start, int end);
\n-
417
\n-
423
\n-
452 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
453 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
454 static bool EvalPatches(
\n-
455 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
456 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
457 int numPatchCoords,
\n-
458 PATCHCOORD_BUFFER *patchCoords,
\n-
459 PATCH_TABLE *patchTable,
\n-
460 TbbEvaluator const *instance = NULL,
\n-
461 void * deviceContext = NULL) {
\n-
462
\n-
463 (void)instance; // unused
\n-
464 (void)deviceContext; // unused
\n-
465
\n-
466 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
467 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
468 numPatchCoords,
\n-
469 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
470 patchTable->GetPatchArrayBuffer(),
\n-
471 patchTable->GetPatchIndexBuffer(),
\n-
472 patchTable->GetPatchParamBuffer());
\n-
473 }
\n-
474
\n-
515 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
516 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
517 static bool EvalPatches(
\n-
518 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
519 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
520 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
521 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
522 int numPatchCoords,
\n-
523 PATCHCOORD_BUFFER *patchCoords,
\n-
524 PATCH_TABLE *patchTable,
\n-
525 TbbEvaluator const *instance = NULL,
\n-
526 void * deviceContext = NULL) {
\n-
527
\n-
528 (void)instance; // unused
\n-
529 (void)deviceContext; // unused
\n-
530
\n-
531 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.
\n-
532 // ideally all buffer classes should have templated by datatype
\n-
533 // so that downcast isn't needed there.
\n-
534 // (e.g. Osd::CpuBuffer<PatchCoord> )
\n-
535 //
\n-
536 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
537 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
538 duBuffer->BindCpuBuffer(), duDesc,
\n-
539 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
540 numPatchCoords,
\n-
541 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
542 patchTable->GetPatchArrayBuffer(),
\n-
543 patchTable->GetPatchIndexBuffer(),
\n-
544 patchTable->GetPatchParamBuffer());
\n-
545 }
\n-
546
\n-
605 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
606 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
607 static bool EvalPatches(
\n-
608 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
609 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
610 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
611 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
612 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
613 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
614 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
615 int numPatchCoords,
\n-
616 PATCHCOORD_BUFFER *patchCoords,
\n-
617 PATCH_TABLE *patchTable,
\n-
618 TbbEvaluator const *instance = NULL,
\n-
619 void * deviceContext = NULL) {
\n-
620
\n-
621 (void)instance; // unused
\n-
622 (void)deviceContext; // unused
\n-
623
\n-
624 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.
\n-
625 // ideally all buffer classes should have templated by datatype
\n-
626 // so that downcast isn't needed there.
\n-
627 // (e.g. Osd::CpuBuffer<PatchCoord> )
\n-
628 //
\n-
629 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
630 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
631 duBuffer->BindCpuBuffer(), duDesc,
\n-
632 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
633 duuBuffer->BindCpuBuffer(), duuDesc,
\n-
634 duvBuffer->BindCpuBuffer(), duvDesc,
\n-
635 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n-
636 numPatchCoords,
\n-
637 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
638 patchTable->GetPatchArrayBuffer(),
\n-
639 patchTable->GetPatchIndexBuffer(),
\n-
640 patchTable->GetPatchParamBuffer());
\n-
641 }
\n-
642
\n-
670 static bool EvalPatches(
\n-
671 const float *src, BufferDescriptor const &srcDesc,
\n-
672 float *dst, BufferDescriptor const &dstDesc,
\n-
673 int numPatchCoords,
\n-
674 const PatchCoord *patchCoords,
\n-
675 const PatchArray *patchArrays,
\n-
676 const int *patchIndexBuffer,
\n-
677 const PatchParam *patchParamBuffer);
\n-
678
\n-
716 static bool EvalPatches(
\n-
717 const float *src, BufferDescriptor const &srcDesc,
\n-
718 float *dst, BufferDescriptor const &dstDesc,
\n-
719 float *du, BufferDescriptor const &duDesc,
\n-
720 float *dv, BufferDescriptor const &dvDesc,
\n-
721 int numPatchCoords,
\n-
722 PatchCoord const *patchCoords,
\n-
723 PatchArray const *patchArrays,
\n-
724 const int *patchIndexBuffer,
\n-
725 PatchParam const *patchParamBuffer);
\n-
726
\n-
779 static bool EvalPatches(
\n-
780 const float *src, BufferDescriptor const &srcDesc,
\n-
781 float *dst, BufferDescriptor const &dstDesc,
\n-
782 float *du, BufferDescriptor const &duDesc,
\n-
783 float *dv, BufferDescriptor const &dvDesc,
\n-
784 float *duu, BufferDescriptor const &duuDesc,
\n-
785 float *duv, BufferDescriptor const &duvDesc,
\n-
786 float *dvv, BufferDescriptor const &dvvDesc,
\n-
787 int numPatchCoords,
\n-
788 PatchCoord const *patchCoords,
\n-
789 PatchArray const *patchArrays,
\n-
790 const int *patchIndexBuffer,
\n-
791 PatchParam const *patchParamBuffer);
\n-
792
\n-
821 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
822 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
823 static bool EvalPatchesVarying(
\n-
824 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
825 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
826 int numPatchCoords,
\n-
827 PATCHCOORD_BUFFER *patchCoords,
\n-
828 PATCH_TABLE *patchTable,
\n-
829 TbbEvaluator const *instance = NULL,
\n-
830 void * deviceContext = NULL) {
\n-
831
\n-
832 (void)instance; // unused
\n-
833 (void)deviceContext; // unused
\n-
834
\n-
835 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
836 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
837 numPatchCoords,
\n-
838 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
839 patchTable->GetVaryingPatchArrayBuffer(),
\n-
840 patchTable->GetVaryingPatchIndexBuffer(),
\n-
841 patchTable->GetPatchParamBuffer());
\n-
842 }
\n-
843
\n-
884 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
885 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
886 static bool EvalPatchesVarying(
\n-
887 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
888 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
889 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
890 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
891 int numPatchCoords,
\n-
892 PATCHCOORD_BUFFER *patchCoords,
\n-
893 PATCH_TABLE *patchTable,
\n-
894 TbbEvaluator const *instance = NULL,
\n-
895 void * deviceContext = NULL) {
\n-
896
\n-
897 (void)instance; // unused
\n-
898 (void)deviceContext; // unused
\n-
899
\n-
900 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
901 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
902 duBuffer->BindCpuBuffer(), duDesc,
\n-
903 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
904 numPatchCoords,
\n-
905 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
906 patchTable->GetVaryingPatchArrayBuffer(),
\n-
907 patchTable->GetVaryingPatchIndexBuffer(),
\n-
908 patchTable->GetPatchParamBuffer());
\n-
909 }
\n-
910
\n-
969 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
970 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
971 static bool EvalPatchesVarying(
\n-
972 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
973 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
974 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
975 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
976 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
977 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
978 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
979 int numPatchCoords,
\n-
980 PATCHCOORD_BUFFER *patchCoords,
\n-
981 PATCH_TABLE *patchTable,
\n-
982 TbbEvaluator const *instance = NULL,
\n-
983 void * deviceContext = NULL) {
\n-
984
\n-
985 (void)instance; // unused
\n-
986 (void)deviceContext; // unused
\n-
987
\n-
988 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
989 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
990 duBuffer->BindCpuBuffer(), duDesc,
\n-
991 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
992 duuBuffer->BindCpuBuffer(), duuDesc,
\n-
993 duvBuffer->BindCpuBuffer(), duvDesc,
\n-
994 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n-
995 numPatchCoords,
\n-
996 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
997 patchTable->GetVaryingPatchArrayBuffer(),
\n-
998 patchTable->GetVaryingPatchIndexBuffer(),
\n-
999 patchTable->GetPatchParamBuffer());
\n-
1000 }
\n-
1001
\n-
1032 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1033 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1034 static bool EvalPatchesFaceVarying(
\n-
1035 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1036 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1037 int numPatchCoords,
\n-
1038 PATCHCOORD_BUFFER *patchCoords,
\n-
1039 PATCH_TABLE *patchTable,
\n-
1040 int fvarChannel,
\n-
1041 TbbEvaluator const *instance = NULL,
\n-
1042 void * deviceContext = NULL) {
\n-
1043
\n-
1044 (void)instance; // unused
\n-
1045 (void)deviceContext; // unused
\n-
1046
\n-
1047 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
1048 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
1049 numPatchCoords,
\n-
1050 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
1051 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
1052 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1053 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
1054 }
\n-
1055
\n-
1098 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1099 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1100 static bool EvalPatchesFaceVarying(
\n-
1101 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1102 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1103 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1104 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1105 int numPatchCoords,
\n-
1106 PATCHCOORD_BUFFER *patchCoords,
\n-
1107 PATCH_TABLE *patchTable,
\n-
1108 int fvarChannel,
\n-
1109 TbbEvaluator const *instance = NULL,
\n-
1110 void * deviceContext = NULL) {
\n-
1111
\n-
1112 (void)instance; // unused
\n-
1113 (void)deviceContext; // unused
\n-
1114
\n-
1115 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
1116 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
1117 duBuffer->BindCpuBuffer(), duDesc,
\n-
1118 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
1119 numPatchCoords,
\n-
1120 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
1121 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
1122 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1123 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
1124 }
\n-
1125
\n-
1186 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1187 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1188 static bool EvalPatchesFaceVarying(
\n-
1189 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1190 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1191 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1192 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1193 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1194 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1195 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1196 int numPatchCoords,
\n-
1197 PATCHCOORD_BUFFER *patchCoords,
\n-
1198 PATCH_TABLE *patchTable,
\n-
1199 int fvarChannel,
\n-
1200 TbbEvaluator const *instance = NULL,
\n-
1201 void * deviceContext = NULL) {
\n-
1202
\n-
1203 (void)instance; // unused
\n-
1204 (void)deviceContext; // unused
\n-
1205
\n-
1206 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
1207 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
1208 duBuffer->BindCpuBuffer(), duDesc,
\n-
1209 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
1210 duuBuffer->BindCpuBuffer(), duuDesc,
\n-
1211 duvBuffer->BindCpuBuffer(), duvDesc,
\n-
1212 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n-
1213 numPatchCoords,
\n-
1214 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
1215 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
1216 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1217 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
1218 }
\n-
1219
\n-
1225
\n-
1227 static void Synchronize(void *deviceContext = NULL);
\n-
1228
\n-
1234 static void SetNumThreads(int numThreads);
\n-
1235};
\n-
1236
\n-
1237
\n-
1238} // end namespace Osd
\n-
1239
\n-
1240} // end namespace OPENSUBDIV_VERSION
\n-
1241using namespace OPENSUBDIV_VERSION;
\n-
1242
\n-
1243} // end namespace OpenSubdiv
\n-
1244
\n-
1245
\n-
1246#endif // OPENSUBDIV3_OSD_TBB_EVALUATOR_H
\n+
37struct PatchArray;
\n+
38struct PatchCoord;
\n+
39struct PatchParam;
\n+
40struct BufferDescriptor;
\n+
41
\n+
42void
\n+
43TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n+
44 float * dst, BufferDescriptor const &dstDesc,
\n+
45 int const * sizes,
\n+
46 int const * offsets,
\n+
47 int const * indices,
\n+
48 float const * weights,
\n+
49 int start, int end);
\n+
50
\n+
51void
\n+
52TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n+
53 float * dst, BufferDescriptor const &dstDesc,
\n+
54 float * dstDu, BufferDescriptor const &dstDuDesc,
\n+
55 float * dstDv, BufferDescriptor const &dstDvDesc,
\n+
56 int const * sizes,
\n+
57 int const * offsets,
\n+
58 int const * indices,
\n+
59 float const * weights,
\n+
60 float const * duWeights,
\n+
61 float const * dvWeights,
\n+
62 int start, int end);
\n+
63
\n+
64void
\n+
65TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n+
66 float * dst, BufferDescriptor const &dstDesc,
\n+
67 float * dstDu, BufferDescriptor const &dstDuDesc,
\n+
68 float * dstDv, BufferDescriptor const &dstDvDesc,
\n+
69 float * dstDuu, BufferDescriptor const &dstDuuDesc,
\n+
70 float * dstDuv, BufferDescriptor const &dstDuvDesc,
\n+
71 float * dstDvv, BufferDescriptor const &dstDvvDesc,
\n+
72 int const * sizes,
\n+
73 int const * offsets,
\n+
74 int const * indices,
\n+
75 float const * weights,
\n+
76 float const * duWeights,
\n+
77 float const * dvWeights,
\n+
78 float const * duuWeights,
\n+
79 float const * duvWeights,
\n+
80 float const * dvvWeights,
\n+
81 int start, int end);
\n+
82
\n+
83void
\n+
84TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc,
\n+
85 float *dst, BufferDescriptor const &dstDesc,
\n+
86 float *dstDu, BufferDescriptor const &dstDuDesc,
\n+
87 float *dstDv, BufferDescriptor const &dstDvDesc,
\n+
88 int numPatchCoords,
\n+
89 const PatchCoord *patchCoords,
\n+
90 const PatchArray *patchArrayBuffer,
\n+
91 const int *patchIndexBuffer,
\n+
92 const PatchParam *patchParamBuffer);
\n+
93
\n+
94void
\n+
95TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc,
\n+
96 float *dst, BufferDescriptor const &dstDesc,
\n+
97 float *dstDu, BufferDescriptor const &dstDuDesc,
\n+
98 float *dstDv, BufferDescriptor const &dstDvDesc,
\n+
99 float *dstDuu, BufferDescriptor const &dstDuuDesc,
\n+
100 float *dstDuv, BufferDescriptor const &dstDuvDesc,
\n+
101 float *dstDvv, BufferDescriptor const &dstDvvDesc,
\n+
102 int numPatchCoords,
\n+
103 const PatchCoord *patchCoords,
\n+
104 const PatchArray *patchArrayBuffer,
\n+
105 const int *patchIndexBuffer,
\n+
106 const PatchParam *patchParamBuffer);
\n+
107
\n+
108} // end namespace Osd
\n+
109
\n+
110} // end namespace OPENSUBDIV_VERSION
\n+
111using namespace OPENSUBDIV_VERSION;
\n+
112
\n+
113} // end namespace OpenSubdiv
\n+
114
\n+
115#endif // OPENSUBDIV3_OSD_TBB_KERNEL_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function. This function has a same signature as other device kernels hav...
Definition: tbbEvaluator.h:72
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:454
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, const TbbEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: tbbEvaluator.h:303
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, int start, int end)
Static eval stencils function which takes raw CPU pointers for input and output.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:886
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: tbbEvaluator.h:607
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw CPU pointers for input and output.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, const float *duuWeights, const float *duvWeights, const float *dvvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw CPU pointers for input and output.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer, PatchParam const *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::Synchronize
static void Synchronize(void *deviceContext=NULL)
synchronize all asynchronous computation invoked on this device.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:971
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:1100
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: tbbEvaluator.h:517
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:1034
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer, PatchParam const *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::SetNumThreads
static void SetNumThreads(int numThreads)
initialize tbb task schedular (optional: client may use tbb::task_scheduler_init)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrays, const int *patchIndexBuffer, const PatchParam *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const TbbEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: tbbEvaluator.h:168
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:1188
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: tbbEvaluator.h:823
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalPatches
void TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrayBuffer, const int *patchIndexBuffer, const PatchParam *patchParamBuffer)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalStencils
void TbbEvalStencils(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, int start, int end)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord
Coordinates set on a patch table.
Definition: types.h:42
\n \n \n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-tbbEvaluator.h\n+tbbKernel.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2015 Pixar\n+ 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,660 +30,126 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_TBB_EVALUATOR_H\n- 26#define OPENSUBDIV3_OSD_TBB_EVALUATOR_H\n+ 25#ifndef OPENSUBDIV3_OSD_TBB_KERNEL_H\n+ 26#define OPENSUBDIV3_OSD_TBB_KERNEL_H\n 27\n 28#include \"../version.h\"\n- 29#include \"../osd/bufferDescriptor.h\"\n- 30#include \"../osd/types.h\"\n+ 29#include \"../far/patchDescriptor.h\"\n+ 30#include \"../far/patchParam.h\"\n 31\n- 32#include \n- 33\n- 34namespace OpenSubdiv {\n- 35namespace OPENSUBDIV_VERSION {\n+ 32namespace OpenSubdiv {\n+ 33namespace OPENSUBDIV_VERSION {\n+ 34\n+ 35namespace Osd {\n 36\n- 37namespace Osd {\n- 38\n-39class TbbEvaluator {\n- 40public:\n- 46\n- 71 template \n-72 static bool EvalStencils(\n- 73 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 74 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 75 STENCIL_TABLE const *stencilTable,\n- 76 TbbEvaluator const *instance = NULL,\n- 77 void *deviceContext = NULL) {\n- 78\n- 79 (void)instance; // unused\n- 80 (void)deviceContext; // unused\n- 81\n- 82 if (stencilTable->GetNumStencils() == 0)\n- 83 return false;\n- 84\n- 85 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n- 86 dstBuffer->BindCpuBuffer(), dstDesc,\n- 87 &stencilTable->GetSizes()[0],\n- 88 &stencilTable->GetOffsets()[0],\n- 89 &stencilTable->GetControlIndices()[0],\n- 90 &stencilTable->GetWeights()[0],\n- 91 /*start = */ 0,\n- 92 /*end = */ stencilTable->GetNumStencils());\n- 93 }\n- 94\n-121 static bool EvalStencils(\n- 122 const float *src, BufferDescriptor const &srcDesc,\n- 123 float *dst, BufferDescriptor const &dstDesc,\n- 124 const int * sizes,\n- 125 const int * offsets,\n- 126 const int * indices,\n- 127 const float * weights,\n- 128 int start, int end);\n- 129\n- 167 template \n-168 static bool EvalStencils(\n- 169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 171 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 172 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 173 STENCIL_TABLE const *stencilTable,\n- 174 const TbbEvaluator *instance = NULL,\n- 175 void * deviceContext = NULL) {\n- 176\n- 177 (void)instance; // unused\n- 178 (void)deviceContext; // unused\n- 179\n- 180 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n- 181 dstBuffer->BindCpuBuffer(), dstDesc,\n- 182 duBuffer->BindCpuBuffer(), duDesc,\n- 183 dvBuffer->BindCpuBuffer(), dvDesc,\n- 184 &stencilTable->GetSizes()[0],\n- 185 &stencilTable->GetOffsets()[0],\n- 186 &stencilTable->GetControlIndices()[0],\n- 187 &stencilTable->GetWeights()[0],\n- 188 &stencilTable->GetDuWeights()[0],\n- 189 &stencilTable->GetDvWeights()[0],\n- 190 /*start = */ 0,\n- 191 /*end = */ stencilTable->GetNumStencils());\n- 192 }\n- 193\n-234 static bool EvalStencils(\n- 235 const float *src, BufferDescriptor const &srcDesc,\n- 236 float *dst, BufferDescriptor const &dstDesc,\n- 237 float *du, BufferDescriptor const &duDesc,\n- 238 float *dv, BufferDescriptor const &dvDesc,\n- 239 const int * sizes,\n- 240 const int * offsets,\n- 241 const int * indices,\n- 242 const float * weights,\n- 243 const float * duWeights,\n- 244 const float * dvWeights,\n- 245 int start, int end);\n- 246\n- 302 template \n-303 static bool EvalStencils(\n- 304 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 305 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 306 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 307 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 308 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 309 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 310 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 311 STENCIL_TABLE const *stencilTable,\n- 312 const TbbEvaluator *instance = NULL,\n- 313 void * deviceContext = NULL) {\n- 314\n- 315 (void)instance; // unused\n- 316 (void)deviceContext; // unused\n- 317\n- 318 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n- 319 dstBuffer->BindCpuBuffer(), dstDesc,\n- 320 duBuffer->BindCpuBuffer(), duDesc,\n- 321 dvBuffer->BindCpuBuffer(), dvDesc,\n- 322 duuBuffer->BindCpuBuffer(), duuDesc,\n- 323 duvBuffer->BindCpuBuffer(), duvDesc,\n- 324 dvvBuffer->BindCpuBuffer(), dvvDesc,\n- 325 &stencilTable->GetSizes()[0],\n- 326 &stencilTable->GetOffsets()[0],\n- 327 &stencilTable->GetControlIndices()[0],\n- 328 &stencilTable->GetWeights()[0],\n- 329 &stencilTable->GetDuWeights()[0],\n- 330 &stencilTable->GetDvWeights()[0],\n- 331 &stencilTable->GetDuuWeights()[0],\n- 332 &stencilTable->GetDuvWeights()[0],\n- 333 &stencilTable->GetDvvWeights()[0],\n- 334 /*start = */ 0,\n- 335 /*end = */ stencilTable->GetNumStencils());\n- 336 }\n- 337\n-399 static bool EvalStencils(\n- 400 const float *src, BufferDescriptor const &srcDesc,\n- 401 float *dst, BufferDescriptor const &dstDesc,\n- 402 float *du, BufferDescriptor const &duDesc,\n- 403 float *dv, BufferDescriptor const &dvDesc,\n- 404 float *duu, BufferDescriptor const &duuDesc,\n- 405 float *duv, BufferDescriptor const &duvDesc,\n- 406 float *dvv, BufferDescriptor const &dvvDesc,\n- 407 const int * sizes,\n- 408 const int * offsets,\n- 409 const int * indices,\n- 410 const float * weights,\n- 411 const float * duWeights,\n- 412 const float * dvWeights,\n- 413 const float * duuWeights,\n- 414 const float * duvWeights,\n- 415 const float * dvvWeights,\n- 416 int start, int end);\n- 417\n- 423\n- 452 template \n-454 static bool EvalPatches(\n- 455 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 456 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 457 int numPatchCoords,\n- 458 PATCHCOORD_BUFFER *patchCoords,\n- 459 PATCH_TABLE *patchTable,\n- 460 TbbEvaluator const *instance = NULL,\n- 461 void * deviceContext = NULL) {\n- 462\n- 463 (void)instance; // unused\n- 464 (void)deviceContext; // unused\n- 465\n- 466 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 467 dstBuffer->BindCpuBuffer(), dstDesc,\n- 468 numPatchCoords,\n- 469 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 470 patchTable->GetPatchArrayBuffer(),\n- 471 patchTable->GetPatchIndexBuffer(),\n- 472 patchTable->GetPatchParamBuffer());\n- 473 }\n- 474\n- 515 template \n-517 static bool EvalPatches(\n- 518 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 519 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 520 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 521 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 522 int numPatchCoords,\n- 523 PATCHCOORD_BUFFER *patchCoords,\n- 524 PATCH_TABLE *patchTable,\n- 525 TbbEvaluator const *instance = NULL,\n- 526 void * deviceContext = NULL) {\n- 527\n- 528 (void)instance; // unused\n- 529 (void)deviceContext; // unused\n- 530\n- 531 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.\n- 532 // ideally all buffer classes should have templated by datatype\n- 533 // so that downcast isn't needed there.\n- 534 // (e.g. Osd::CpuBuffer )\n- 535 //\n- 536 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 537 dstBuffer->BindCpuBuffer(), dstDesc,\n- 538 duBuffer->BindCpuBuffer(), duDesc,\n- 539 dvBuffer->BindCpuBuffer(), dvDesc,\n- 540 numPatchCoords,\n- 541 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 542 patchTable->GetPatchArrayBuffer(),\n- 543 patchTable->GetPatchIndexBuffer(),\n- 544 patchTable->GetPatchParamBuffer());\n- 545 }\n- 546\n- 605 template \n-607 static bool EvalPatches(\n- 608 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 609 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 610 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 611 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 612 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 613 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 614 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 615 int numPatchCoords,\n- 616 PATCHCOORD_BUFFER *patchCoords,\n- 617 PATCH_TABLE *patchTable,\n- 618 TbbEvaluator const *instance = NULL,\n- 619 void * deviceContext = NULL) {\n- 620\n- 621 (void)instance; // unused\n- 622 (void)deviceContext; // unused\n- 623\n- 624 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.\n- 625 // ideally all buffer classes should have templated by datatype\n- 626 // so that downcast isn't needed there.\n- 627 // (e.g. Osd::CpuBuffer )\n- 628 //\n- 629 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 630 dstBuffer->BindCpuBuffer(), dstDesc,\n- 631 duBuffer->BindCpuBuffer(), duDesc,\n- 632 dvBuffer->BindCpuBuffer(), dvDesc,\n- 633 duuBuffer->BindCpuBuffer(), duuDesc,\n- 634 duvBuffer->BindCpuBuffer(), duvDesc,\n- 635 dvvBuffer->BindCpuBuffer(), dvvDesc,\n- 636 numPatchCoords,\n- 637 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 638 patchTable->GetPatchArrayBuffer(),\n- 639 patchTable->GetPatchIndexBuffer(),\n- 640 patchTable->GetPatchParamBuffer());\n- 641 }\n- 642\n-670 static bool EvalPatches(\n- 671 const float *src, BufferDescriptor const &srcDesc,\n- 672 float *dst, BufferDescriptor const &dstDesc,\n- 673 int numPatchCoords,\n- 674 const PatchCoord *patchCoords,\n- 675 const PatchArray *patchArrays,\n- 676 const int *patchIndexBuffer,\n- 677 const PatchParam *patchParamBuffer);\n- 678\n-716 static bool EvalPatches(\n- 717 const float *src, BufferDescriptor const &srcDesc,\n- 718 float *dst, BufferDescriptor const &dstDesc,\n- 719 float *du, BufferDescriptor const &duDesc,\n- 720 float *dv, BufferDescriptor const &dvDesc,\n- 721 int numPatchCoords,\n- 722 PatchCoord const *patchCoords,\n- 723 PatchArray const *patchArrays,\n- 724 const int *patchIndexBuffer,\n- 725 PatchParam const *patchParamBuffer);\n- 726\n-779 static bool EvalPatches(\n- 780 const float *src, BufferDescriptor const &srcDesc,\n- 781 float *dst, BufferDescriptor const &dstDesc,\n- 782 float *du, BufferDescriptor const &duDesc,\n- 783 float *dv, BufferDescriptor const &dvDesc,\n- 784 float *duu, BufferDescriptor const &duuDesc,\n- 785 float *duv, BufferDescriptor const &duvDesc,\n- 786 float *dvv, BufferDescriptor const &dvvDesc,\n- 787 int numPatchCoords,\n- 788 PatchCoord const *patchCoords,\n- 789 PatchArray const *patchArrays,\n- 790 const int *patchIndexBuffer,\n- 791 PatchParam const *patchParamBuffer);\n- 792\n- 821 template \n-823 static bool EvalPatchesVarying(\n- 824 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 825 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 826 int numPatchCoords,\n- 827 PATCHCOORD_BUFFER *patchCoords,\n- 828 PATCH_TABLE *patchTable,\n- 829 TbbEvaluator const *instance = NULL,\n- 830 void * deviceContext = NULL) {\n- 831\n- 832 (void)instance; // unused\n- 833 (void)deviceContext; // unused\n- 834\n- 835 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 836 dstBuffer->BindCpuBuffer(), dstDesc,\n- 837 numPatchCoords,\n- 838 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 839 patchTable->GetVaryingPatchArrayBuffer(),\n- 840 patchTable->GetVaryingPatchIndexBuffer(),\n- 841 patchTable->GetPatchParamBuffer());\n- 842 }\n- 843\n- 884 template \n-886 static bool EvalPatchesVarying(\n- 887 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 888 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 889 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 890 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 891 int numPatchCoords,\n- 892 PATCHCOORD_BUFFER *patchCoords,\n- 893 PATCH_TABLE *patchTable,\n- 894 TbbEvaluator const *instance = NULL,\n- 895 void * deviceContext = NULL) {\n- 896\n- 897 (void)instance; // unused\n- 898 (void)deviceContext; // unused\n- 899\n- 900 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 901 dstBuffer->BindCpuBuffer(), dstDesc,\n- 902 duBuffer->BindCpuBuffer(), duDesc,\n- 903 dvBuffer->BindCpuBuffer(), dvDesc,\n- 904 numPatchCoords,\n- 905 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 906 patchTable->GetVaryingPatchArrayBuffer(),\n- 907 patchTable->GetVaryingPatchIndexBuffer(),\n- 908 patchTable->GetPatchParamBuffer());\n- 909 }\n- 910\n- 969 template \n-971 static bool EvalPatchesVarying(\n- 972 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 973 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 974 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 975 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 976 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 977 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 978 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 979 int numPatchCoords,\n- 980 PATCHCOORD_BUFFER *patchCoords,\n- 981 PATCH_TABLE *patchTable,\n- 982 TbbEvaluator const *instance = NULL,\n- 983 void * deviceContext = NULL) {\n- 984\n- 985 (void)instance; // unused\n- 986 (void)deviceContext; // unused\n- 987\n- 988 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 989 dstBuffer->BindCpuBuffer(), dstDesc,\n- 990 duBuffer->BindCpuBuffer(), duDesc,\n- 991 dvBuffer->BindCpuBuffer(), dvDesc,\n- 992 duuBuffer->BindCpuBuffer(), duuDesc,\n- 993 duvBuffer->BindCpuBuffer(), duvDesc,\n- 994 dvvBuffer->BindCpuBuffer(), dvvDesc,\n- 995 numPatchCoords,\n- 996 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 997 patchTable->GetVaryingPatchArrayBuffer(),\n- 998 patchTable->GetVaryingPatchIndexBuffer(),\n- 999 patchTable->GetPatchParamBuffer());\n- 1000 }\n- 1001\n- 1032 template \n-1034 static bool EvalPatchesFaceVarying(\n- 1035 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1036 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1037 int numPatchCoords,\n- 1038 PATCHCOORD_BUFFER *patchCoords,\n- 1039 PATCH_TABLE *patchTable,\n- 1040 int fvarChannel,\n- 1041 TbbEvaluator const *instance = NULL,\n- 1042 void * deviceContext = NULL) {\n- 1043\n- 1044 (void)instance; // unused\n- 1045 (void)deviceContext; // unused\n- 1046\n- 1047 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 1048 dstBuffer->BindCpuBuffer(), dstDesc,\n- 1049 numPatchCoords,\n- 1050 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 1051 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 1052 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1053 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 1054 }\n- 1055\n- 1098 template \n-1100 static bool EvalPatchesFaceVarying(\n- 1101 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1102 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1103 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1104 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1105 int numPatchCoords,\n- 1106 PATCHCOORD_BUFFER *patchCoords,\n- 1107 PATCH_TABLE *patchTable,\n- 1108 int fvarChannel,\n- 1109 TbbEvaluator const *instance = NULL,\n- 1110 void * deviceContext = NULL) {\n- 1111\n- 1112 (void)instance; // unused\n- 1113 (void)deviceContext; // unused\n- 1114\n- 1115 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 1116 dstBuffer->BindCpuBuffer(), dstDesc,\n- 1117 duBuffer->BindCpuBuffer(), duDesc,\n- 1118 dvBuffer->BindCpuBuffer(), dvDesc,\n- 1119 numPatchCoords,\n- 1120 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 1121 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 1122 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1123 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 1124 }\n- 1125\n- 1186 template \n-1188 static bool EvalPatchesFaceVarying(\n- 1189 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1190 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1191 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1192 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1193 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1194 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1195 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1196 int numPatchCoords,\n- 1197 PATCHCOORD_BUFFER *patchCoords,\n- 1198 PATCH_TABLE *patchTable,\n- 1199 int fvarChannel,\n- 1200 TbbEvaluator const *instance = NULL,\n- 1201 void * deviceContext = NULL) {\n- 1202\n- 1203 (void)instance; // unused\n- 1204 (void)deviceContext; // unused\n- 1205\n- 1206 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 1207 dstBuffer->BindCpuBuffer(), dstDesc,\n- 1208 duBuffer->BindCpuBuffer(), duDesc,\n- 1209 dvBuffer->BindCpuBuffer(), dvDesc,\n- 1210 duuBuffer->BindCpuBuffer(), duuDesc,\n- 1211 duvBuffer->BindCpuBuffer(), duvDesc,\n- 1212 dvvBuffer->BindCpuBuffer(), dvvDesc,\n- 1213 numPatchCoords,\n- 1214 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 1215 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 1216 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1217 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 1218 }\n- 1219\n- 1225\n-1227 static void Synchronize(void *deviceContext = NULL);\n- 1228\n-1234 static void SetNumThreads(int numThreads);\n- 1235};\n- 1236\n- 1237\n- 1238} // end namespace Osd\n- 1239\n- 1240} // end namespace OPENSUBDIV_VERSION\n- 1241using namespace OPENSUBDIV_VERSION;\n- 1242\n- 1243} // end namespace OpenSubdiv\n- 1244\n- 1245\n- 1246#endif // OPENSUBDIV3_OSD_TBB_EVALUATOR_H\n+ 37struct PatchArray;\n+ 38struct PatchCoord;\n+ 39struct PatchParam;\n+ 40struct BufferDescriptor;\n+ 41\n+ 42void\n+43TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n+ 44 float * dst, BufferDescriptor const &dstDesc,\n+ 45 int const * sizes,\n+ 46 int const * offsets,\n+ 47 int const * indices,\n+ 48 float const * weights,\n+ 49 int start, int end);\n+ 50\n+ 51void\n+52TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n+ 53 float * dst, BufferDescriptor const &dstDesc,\n+ 54 float * dstDu, BufferDescriptor const &dstDuDesc,\n+ 55 float * dstDv, BufferDescriptor const &dstDvDesc,\n+ 56 int const * sizes,\n+ 57 int const * offsets,\n+ 58 int const * indices,\n+ 59 float const * weights,\n+ 60 float const * duWeights,\n+ 61 float const * dvWeights,\n+ 62 int start, int end);\n+ 63\n+ 64void\n+65TbbEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n+ 66 float * dst, BufferDescriptor const &dstDesc,\n+ 67 float * dstDu, BufferDescriptor const &dstDuDesc,\n+ 68 float * dstDv, BufferDescriptor const &dstDvDesc,\n+ 69 float * dstDuu, BufferDescriptor const &dstDuuDesc,\n+ 70 float * dstDuv, BufferDescriptor const &dstDuvDesc,\n+ 71 float * dstDvv, BufferDescriptor const &dstDvvDesc,\n+ 72 int const * sizes,\n+ 73 int const * offsets,\n+ 74 int const * indices,\n+ 75 float const * weights,\n+ 76 float const * duWeights,\n+ 77 float const * dvWeights,\n+ 78 float const * duuWeights,\n+ 79 float const * duvWeights,\n+ 80 float const * dvvWeights,\n+ 81 int start, int end);\n+ 82\n+ 83void\n+84TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc,\n+ 85 float *dst, BufferDescriptor const &dstDesc,\n+ 86 float *dstDu, BufferDescriptor const &dstDuDesc,\n+ 87 float *dstDv, BufferDescriptor const &dstDvDesc,\n+ 88 int numPatchCoords,\n+ 89 const PatchCoord *patchCoords,\n+ 90 const PatchArray *patchArrayBuffer,\n+ 91 const int *patchIndexBuffer,\n+ 92 const PatchParam *patchParamBuffer);\n+ 93\n+ 94void\n+95TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc,\n+ 96 float *dst, BufferDescriptor const &dstDesc,\n+ 97 float *dstDu, BufferDescriptor const &dstDuDesc,\n+ 98 float *dstDv, BufferDescriptor const &dstDvDesc,\n+ 99 float *dstDuu, BufferDescriptor const &dstDuuDesc,\n+ 100 float *dstDuv, BufferDescriptor const &dstDuvDesc,\n+ 101 float *dstDvv, BufferDescriptor const &dstDvvDesc,\n+ 102 int numPatchCoords,\n+ 103 const PatchCoord *patchCoords,\n+ 104 const PatchArray *patchArrayBuffer,\n+ 105 const int *patchIndexBuffer,\n+ 106 const PatchParam *patchParamBuffer);\n+ 107\n+ 108} // end namespace Osd\n+ 109\n+ 110} // end namespace OPENSUBDIV_VERSION\n+ 111using namespace OPENSUBDIV_VERSION;\n+ 112\n+ 113} // end namespace OpenSubdiv\n+ 114\n+ 115#endif // OPENSUBDIV3_OSD_TBB_KERNEL_H\n OpenSubdiv\n Definition: limits.h:32\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalPatches\n+void TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc, float\n+*dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const\n+&dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int\n+numPatchCoords, const PatchCoord *patchCoords, const PatchArray\n+*patchArrayBuffer, const int *patchIndexBuffer, const PatchParam\n+*patchParamBuffer)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalStencils\n+void TbbEvalStencils(float const *src, BufferDescriptor const &srcDesc, float\n+*dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets,\n+int const *indices, float const *weights, int start, int end)\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n BufferDescriptor is a struct which describes buffer elements in interleaved\n data buffers....\n Definition: bufferDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator\n-Definition: tbbEvaluator.h:39\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n-const *stencilTable, TbbEvaluator const *instance=NULL, void\n-*deviceContext=NULL)\n-Generic static eval stencils function. This function has a same signature as\n-other device kernels hav...\n-Definition: tbbEvaluator.h:72\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const\n-*instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: tbbEvaluator.h:454\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n-const TbbEvaluator *instance=NULL, void *deviceContext=NULL)\n-Generic static eval stencils function with derivatives. This function has a\n-same signature as other d...\n-Definition: tbbEvaluator.h:303\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n-static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int\n-*offsets, const int *indices, const float *weights, int start, int end)\n-Static eval stencils function which takes raw CPU pointers for input and\n-output.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void\n-*deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: tbbEvaluator.h:886\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const *instance=NULL, void\n-*deviceContext=NULL)\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: tbbEvaluator.h:607\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n-static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int\n-*offsets, const int *indices, const float *weights, const float *duWeights,\n-const float *dvWeights, int start, int end)\n-Static eval stencils function with derivatives, which takes raw CPU pointers\n-for input and output.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n-static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n-BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n-float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int\n-*offsets, const int *indices, const float *weights, const float *duWeights,\n-const float *dvWeights, const float *duuWeights, const float *duvWeights, const\n-float *dvvWeights, int start, int end)\n-Static eval stencils function with derivatives, which takes raw CPU pointers\n-for input and output.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n-static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n-BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n-float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord\n-const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer,\n-PatchParam const *patchParamBuffer)\n-Static limit eval function. It takes an array of PatchCoord and evaluate limit\n-values on given PatchT...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::Synchronize\n-static void Synchronize(void *deviceContext=NULL)\n-synchronize all asynchronous computation invoked on this device.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, TbbEvaluator const\n-*instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: tbbEvaluator.h:971\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, TbbEvaluator const\n-*instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: tbbEvaluator.h:1100\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: tbbEvaluator.h:517\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n-fvarChannel, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: tbbEvaluator.h:1034\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n-static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords,\n-PatchCoord const *patchCoords, PatchArray const *patchArrays, const int\n-*patchIndexBuffer, PatchParam const *patchParamBuffer)\n-Static limit eval function. It takes an array of PatchCoord and evaluate limit\n-values on given PatchT...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::SetNumThreads\n-static void SetNumThreads(int numThreads)\n-initialize tbb task schedular (optional: client may use tbb::\n-task_scheduler_init)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatches\n-static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const\n-PatchCoord *patchCoords, const PatchArray *patchArrays, const int\n-*patchIndexBuffer, const PatchParam *patchParamBuffer)\n-Static limit eval function. It takes an array of PatchCoord and evaluate limit\n-values on given PatchT...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const\n-TbbEvaluator *instance=NULL, void *deviceContext=NULL)\n-Generic static eval stencils function with derivatives. This function has a\n-same signature as other d...\n-Definition: tbbEvaluator.h:168\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n-TbbEvaluator const *instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: tbbEvaluator.h:1188\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n-TbbEvaluator const *instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: tbbEvaluator.h:823\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord\n Coordinates set on a patch table.\n Definition: types.h:42\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArray\n Definition: types.h:66\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParam\n Definition: types.h:127\n * opensubdiv\n * osd\n- * tbbEvaluator.h\n+ * tbbKernel.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00878.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00878.html", "unified_diff": "@@ -87,15 +87,15 @@\n
\n Classes |\n Namespaces
\n
ompEvaluator.h File Reference
\n \n
\n
#include "../version.h"
\n-#include "../osd/bufferDescriptor.h"
\n+#include "../osd/bufferDescriptor.h"
\n #include "../osd/types.h"
\n #include <cstddef>
\n
\n

Go to the source code of this file.

\n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00878_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00878_source.html", "unified_diff": "@@ -566,15 +566,15 @@\n
1235using namespace OPENSUBDIV_VERSION;
\n
1236
\n
1237} // end namespace OpenSubdiv
\n
1238
\n
1239
\n
1240#endif // OPENSUBDIV3_OSD_OMP_EVALUATOR_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, OmpEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: ompEvaluator.h:1034
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, int start, int end)
Static eval stencils function which takes raw CPU pointers for input and output.
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, OmpEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: ompEvaluator.h:971
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const OmpEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: ompEvaluator.h:168
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw CPU pointers for input and output.
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, OmpEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: ompEvaluator.h:823
\n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00881.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00881.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/ompKernel.h File Reference\n+OpenSubdiv: opensubdiv/osd/opengl.h File Reference\n \n \n \n \n \n \n \n@@ -80,46 +80,24 @@\n
\n \n
\n \n
\n-
\n-Namespaces |\n-Functions
\n-
ompKernel.h File Reference
\n+
opengl.h File Reference
\n
\n
\n-
#include "../version.h"
\n+
#include <GL/gl.h>
\n
\n

Go to the source code of this file.

\n-

\n Classes

\n-\n-\n-\n-\n-\n-\n-\n-

\n-Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n-\n-\n-\n-\n-\n-\n-\n-

\n-Functions

void OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, int start, int end)
 
void OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, int start, int end)
 
void OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv, BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const &dstDvvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, float const *duuWeights, float const *duvWeights, float const *dvvWeights, int start, int end)
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,44 +4,14 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-Namespaces | Functions\n-ompKernel.h File Reference\n-#include \"../version.h\"\n+opengl.h File Reference\n+#include \n Go_to_the_source_code_of_this_file.\n- Namespaces\n-namespace \u00a0OpenSubdiv\n-\u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n-\u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n-\u00a0\n- Functions\n-void\u00a0OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n- *dst, BufferDescriptor const &dstDesc, int const *sizes, int const\n- *offsets, int const *indices, float const *weights, int start, int end)\n-\u00a0\n-void\u00a0OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n- *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n- const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int\n- const *sizes, int const *offsets, int const *indices, float const\n- *weights, float const *duWeights, float const *dvWeights, int start, int\n- end)\n-\u00a0\n-void\u00a0OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n- *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n- const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float\n- *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv,\n- BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const\n- &dstDvvDesc, int const *sizes, int const *offsets, int const *indices,\n- float const *weights, float const *duWeights, float const *dvWeights,\n- float const *duuWeights, float const *duvWeights, float const\n- *dvvWeights, int start, int end)\n-\u00a0\n * opensubdiv\n * osd\n- * ompKernel.h\n+ * opengl.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00881_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00881_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/ompKernel.h Source File\n+OpenSubdiv: opensubdiv/osd/opengl.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
ompKernel.h
\n+
opengl.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,82 +107,47 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_OMP_KERNEL_H
\n-
26#define OPENSUBDIV3_OSD_OMP_KERNEL_H
\n+
25#ifndef OPENSUBDIV3_OSD_OPENGL_H
\n+
26#define OPENSUBDIV3_OSD_OPENGL_H
\n
27
\n-
28#include "../version.h"
\n-
29
\n-
30namespace OpenSubdiv {
\n-
31namespace OPENSUBDIV_VERSION {
\n-
32
\n-
33namespace Osd {
\n-
34
\n-
35struct BufferDescriptor;
\n-
36
\n-
37void
\n-
38OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n-
39 float * dst, BufferDescriptor const &dstDesc,
\n-
40 int const * sizes,
\n-
41 int const * offsets,
\n-
42 int const * indices,
\n-
43 float const * weights,
\n-
44 int start, int end);
\n-
45
\n-
46void
\n-
47OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n-
48 float * dst, BufferDescriptor const &dstDesc,
\n-
49 float * dstDu, BufferDescriptor const &dstDuDesc,
\n-
50 float * dstDv, BufferDescriptor const &dstDvDesc,
\n-
51 int const * sizes,
\n-
52 int const * offsets,
\n-
53 int const * indices,
\n-
54 float const * weights,
\n-
55 float const * duWeights,
\n-
56 float const * dvWeights,
\n-
57 int start, int end);
\n-
58
\n-
59void
\n-
60OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n-
61 float * dst, BufferDescriptor const &dstDesc,
\n-
62 float * dstDu, BufferDescriptor const &dstDuDesc,
\n-
63 float * dstDv, BufferDescriptor const &dstDvDesc,
\n-
64 float * dstDuu, BufferDescriptor const &dstDuuDesc,
\n-
65 float * dstDuv, BufferDescriptor const &dstDuvDesc,
\n-
66 float * dstDvv, BufferDescriptor const &dstDvvDesc,
\n-
67 int const * sizes,
\n-
68 int const * offsets,
\n-
69 int const * indices,
\n-
70 float const * weights,
\n-
71 float const * duWeights,
\n-
72 float const * dvWeights,
\n-
73 float const * duuWeights,
\n-
74 float const * duvWeights,
\n-
75 float const * dvvWeights,
\n-
76 int start, int end);
\n-
77
\n-
78} // end namespace Osd
\n-
79
\n-
80} // end namespace OPENSUBDIV_VERSION
\n-
81using namespace OPENSUBDIV_VERSION;
\n-
82
\n-
83} // end namespace OpenSubdiv
\n-
84
\n-
85#endif // OPENSUBDIV3_OSD_OMP_KERNEL_H
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvalStencils
void OmpEvalStencils(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, int start, int end)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+
28#if defined(__APPLE__)
\n+
29 #include "TargetConditionals.h"
\n+
30 #if TARGET_OS_IPHONE || TARGET_IPHONE_SIMULATOR
\n+
31 #include <OpenGLES/ES2/gl.h>
\n+
32 #else
\n+
33 #if defined(OSD_USES_GLEW)
\n+
34 #include <GL/glew.h>
\n+
35 #else
\n+
36 #include <OpenGL/gl3.h>
\n+
37 #endif
\n+
38 #endif
\n+
39#elif defined(ANDROID)
\n+
40 #include <GLES2/gl2.h>
\n+
41#else
\n+
42 #if defined(_WIN32)
\n+
43 #define WIN32_LEAN_AND_MEAN
\n+
44 #include <windows.h>
\n+
45 #endif
\n+
46 #if defined(OSD_USES_GLEW)
\n+
47 #include <GL/glew.h>
\n+
48 #else
\n+
49 #include <GL/gl.h>
\n+
50 #endif
\n+
51#endif
\n+
52
\n+
53#endif // OPENSUBDIV3_OSD_OPENGL_H
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-ompKernel.h\n+opengl.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,82 +30,40 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_OMP_KERNEL_H\n- 26#define OPENSUBDIV3_OSD_OMP_KERNEL_H\n+ 25#ifndef OPENSUBDIV3_OSD_OPENGL_H\n+ 26#define OPENSUBDIV3_OSD_OPENGL_H\n 27\n- 28#include \"../version.h\"\n- 29\n- 30namespace OpenSubdiv {\n- 31namespace OPENSUBDIV_VERSION {\n- 32\n- 33namespace Osd {\n- 34\n- 35struct BufferDescriptor;\n- 36\n- 37void\n-38OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n- 39 float * dst, BufferDescriptor const &dstDesc,\n- 40 int const * sizes,\n- 41 int const * offsets,\n- 42 int const * indices,\n- 43 float const * weights,\n- 44 int start, int end);\n- 45\n- 46void\n-47OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n- 48 float * dst, BufferDescriptor const &dstDesc,\n- 49 float * dstDu, BufferDescriptor const &dstDuDesc,\n- 50 float * dstDv, BufferDescriptor const &dstDvDesc,\n- 51 int const * sizes,\n- 52 int const * offsets,\n- 53 int const * indices,\n- 54 float const * weights,\n- 55 float const * duWeights,\n- 56 float const * dvWeights,\n- 57 int start, int end);\n- 58\n- 59void\n-60OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n- 61 float * dst, BufferDescriptor const &dstDesc,\n- 62 float * dstDu, BufferDescriptor const &dstDuDesc,\n- 63 float * dstDv, BufferDescriptor const &dstDvDesc,\n- 64 float * dstDuu, BufferDescriptor const &dstDuuDesc,\n- 65 float * dstDuv, BufferDescriptor const &dstDuvDesc,\n- 66 float * dstDvv, BufferDescriptor const &dstDvvDesc,\n- 67 int const * sizes,\n- 68 int const * offsets,\n- 69 int const * indices,\n- 70 float const * weights,\n- 71 float const * duWeights,\n- 72 float const * dvWeights,\n- 73 float const * duuWeights,\n- 74 float const * duvWeights,\n- 75 float const * dvvWeights,\n- 76 int start, int end);\n- 77\n- 78} // end namespace Osd\n- 79\n- 80} // end namespace OPENSUBDIV_VERSION\n- 81using namespace OPENSUBDIV_VERSION;\n- 82\n- 83} // end namespace OpenSubdiv\n- 84\n- 85#endif // OPENSUBDIV3_OSD_OMP_KERNEL_H\n-OpenSubdiv\n-Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvalStencils\n-void OmpEvalStencils(float const *src, BufferDescriptor const &srcDesc, float\n-*dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets,\n-int const *indices, float const *weights, int start, int end)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n-BufferDescriptor is a struct which describes buffer elements in interleaved\n-data buffers....\n-Definition: bufferDescriptor.h:61\n+ 28#if defined(__APPLE__)\n+ 29 #include \"TargetConditionals.h\"\n+ 30 #if TARGET_OS_IPHONE || TARGET_IPHONE_SIMULATOR\n+ 31 #include \n+ 32 #else\n+ 33 #if defined(OSD_USES_GLEW)\n+ 34 #include \n+ 35 #else\n+ 36 #include \n+ 37 #endif\n+ 38 #endif\n+ 39#elif defined(ANDROID)\n+ 40 #include \n+ 41#else\n+ 42 #if defined(_WIN32)\n+ 43 #define WIN32_LEAN_AND_MEAN\n+ 44 #include \n+ 45 #endif\n+ 46 #if defined(OSD_USES_GLEW)\n+ 47 #include \n+ 48 #else\n+ 49 #include \n+ 50 #endif\n+ 51#endif\n+ 52\n+ 53#endif // OPENSUBDIV3_OSD_OPENGL_H\n * opensubdiv\n * osd\n- * ompKernel.h\n+ * opengl.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00884.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00884.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/opengl.h File Reference\n+OpenSubdiv: opensubdiv/osd/ompKernel.h File Reference\n \n \n \n \n \n \n \n@@ -80,24 +80,46 @@\n
\n \n
\n \n
\n-
opengl.h File Reference
\n+
\n+Namespaces |\n+Functions
\n+
ompKernel.h File Reference
\n
\n
\n-
#include <GL/gl.h>
\n+
#include "../version.h"
\n
\n

Go to the source code of this file.

\n+\n+\n+\n+\n+\n+\n+\n+\n+

\n+Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n+\n+\n+\n+\n+\n+\n+\n+

\n+Functions

void OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, int start, int end)
 
void OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, int start, int end)
 
void OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv, BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const &dstDvvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, float const *duuWeights, float const *duvWeights, float const *dvvWeights, int start, int end)
 
\n
\n
\n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,14 +4,44 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-opengl.h File Reference\n-#include \n+Namespaces | Functions\n+ompKernel.h File Reference\n+#include \"../version.h\"\n Go_to_the_source_code_of_this_file.\n+ Namespaces\n+namespace \u00a0OpenSubdiv\n+\u00a0\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n+\u00a0\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n+\u00a0\n+ Functions\n+void\u00a0OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n+ *dst, BufferDescriptor const &dstDesc, int const *sizes, int const\n+ *offsets, int const *indices, float const *weights, int start, int end)\n+\u00a0\n+void\u00a0OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n+ *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n+ const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int\n+ const *sizes, int const *offsets, int const *indices, float const\n+ *weights, float const *duWeights, float const *dvWeights, int start, int\n+ end)\n+\u00a0\n+void\u00a0OmpEvalStencils (float const *src, BufferDescriptor const &srcDesc, float\n+ *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor\n+ const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float\n+ *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv,\n+ BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const\n+ &dstDvvDesc, int const *sizes, int const *offsets, int const *indices,\n+ float const *weights, float const *duWeights, float const *dvWeights,\n+ float const *duuWeights, float const *duvWeights, float const\n+ *dvvWeights, int start, int end)\n+\u00a0\n * opensubdiv\n * osd\n- * opengl.h\n+ * ompKernel.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00884_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00884_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/opengl.h Source File\n+OpenSubdiv: opensubdiv/osd/ompKernel.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
opengl.h
\n+
ompKernel.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,47 +107,82 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_OPENGL_H
\n-
26#define OPENSUBDIV3_OSD_OPENGL_H
\n+
25#ifndef OPENSUBDIV3_OSD_OMP_KERNEL_H
\n+
26#define OPENSUBDIV3_OSD_OMP_KERNEL_H
\n
27
\n-
28#if defined(__APPLE__)
\n-
29 #include "TargetConditionals.h"
\n-
30 #if TARGET_OS_IPHONE || TARGET_IPHONE_SIMULATOR
\n-
31 #include <OpenGLES/ES2/gl.h>
\n-
32 #else
\n-
33 #if defined(OSD_USES_GLEW)
\n-
34 #include <GL/glew.h>
\n-
35 #else
\n-
36 #include <OpenGL/gl3.h>
\n-
37 #endif
\n-
38 #endif
\n-
39#elif defined(ANDROID)
\n-
40 #include <GLES2/gl2.h>
\n-
41#else
\n-
42 #if defined(_WIN32)
\n-
43 #define WIN32_LEAN_AND_MEAN
\n-
44 #include <windows.h>
\n-
45 #endif
\n-
46 #if defined(OSD_USES_GLEW)
\n-
47 #include <GL/glew.h>
\n-
48 #else
\n-
49 #include <GL/gl.h>
\n-
50 #endif
\n-
51#endif
\n-
52
\n-
53#endif // OPENSUBDIV3_OSD_OPENGL_H
\n+
28#include "../version.h"
\n+
29
\n+
30namespace OpenSubdiv {
\n+
31namespace OPENSUBDIV_VERSION {
\n+
32
\n+
33namespace Osd {
\n+
34
\n+
35struct BufferDescriptor;
\n+
36
\n+
37void
\n+
38OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n+
39 float * dst, BufferDescriptor const &dstDesc,
\n+
40 int const * sizes,
\n+
41 int const * offsets,
\n+
42 int const * indices,
\n+
43 float const * weights,
\n+
44 int start, int end);
\n+
45
\n+
46void
\n+
47OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n+
48 float * dst, BufferDescriptor const &dstDesc,
\n+
49 float * dstDu, BufferDescriptor const &dstDuDesc,
\n+
50 float * dstDv, BufferDescriptor const &dstDvDesc,
\n+
51 int const * sizes,
\n+
52 int const * offsets,
\n+
53 int const * indices,
\n+
54 float const * weights,
\n+
55 float const * duWeights,
\n+
56 float const * dvWeights,
\n+
57 int start, int end);
\n+
58
\n+
59void
\n+
60OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,
\n+
61 float * dst, BufferDescriptor const &dstDesc,
\n+
62 float * dstDu, BufferDescriptor const &dstDuDesc,
\n+
63 float * dstDv, BufferDescriptor const &dstDvDesc,
\n+
64 float * dstDuu, BufferDescriptor const &dstDuuDesc,
\n+
65 float * dstDuv, BufferDescriptor const &dstDuvDesc,
\n+
66 float * dstDvv, BufferDescriptor const &dstDvvDesc,
\n+
67 int const * sizes,
\n+
68 int const * offsets,
\n+
69 int const * indices,
\n+
70 float const * weights,
\n+
71 float const * duWeights,
\n+
72 float const * dvWeights,
\n+
73 float const * duuWeights,
\n+
74 float const * duvWeights,
\n+
75 float const * dvvWeights,
\n+
76 int start, int end);
\n+
77
\n+
78} // end namespace Osd
\n+
79
\n+
80} // end namespace OPENSUBDIV_VERSION
\n+
81using namespace OPENSUBDIV_VERSION;
\n+
82
\n+
83} // end namespace OpenSubdiv
\n+
84
\n+
85#endif // OPENSUBDIV3_OSD_OMP_KERNEL_H
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvalStencils
void OmpEvalStencils(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, int start, int end)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-opengl.h\n+ompKernel.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,40 +30,82 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_OPENGL_H\n- 26#define OPENSUBDIV3_OSD_OPENGL_H\n+ 25#ifndef OPENSUBDIV3_OSD_OMP_KERNEL_H\n+ 26#define OPENSUBDIV3_OSD_OMP_KERNEL_H\n 27\n- 28#if defined(__APPLE__)\n- 29 #include \"TargetConditionals.h\"\n- 30 #if TARGET_OS_IPHONE || TARGET_IPHONE_SIMULATOR\n- 31 #include \n- 32 #else\n- 33 #if defined(OSD_USES_GLEW)\n- 34 #include \n- 35 #else\n- 36 #include \n- 37 #endif\n- 38 #endif\n- 39#elif defined(ANDROID)\n- 40 #include \n- 41#else\n- 42 #if defined(_WIN32)\n- 43 #define WIN32_LEAN_AND_MEAN\n- 44 #include \n- 45 #endif\n- 46 #if defined(OSD_USES_GLEW)\n- 47 #include \n- 48 #else\n- 49 #include \n- 50 #endif\n- 51#endif\n- 52\n- 53#endif // OPENSUBDIV3_OSD_OPENGL_H\n+ 28#include \"../version.h\"\n+ 29\n+ 30namespace OpenSubdiv {\n+ 31namespace OPENSUBDIV_VERSION {\n+ 32\n+ 33namespace Osd {\n+ 34\n+ 35struct BufferDescriptor;\n+ 36\n+ 37void\n+38OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n+ 39 float * dst, BufferDescriptor const &dstDesc,\n+ 40 int const * sizes,\n+ 41 int const * offsets,\n+ 42 int const * indices,\n+ 43 float const * weights,\n+ 44 int start, int end);\n+ 45\n+ 46void\n+47OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n+ 48 float * dst, BufferDescriptor const &dstDesc,\n+ 49 float * dstDu, BufferDescriptor const &dstDuDesc,\n+ 50 float * dstDv, BufferDescriptor const &dstDvDesc,\n+ 51 int const * sizes,\n+ 52 int const * offsets,\n+ 53 int const * indices,\n+ 54 float const * weights,\n+ 55 float const * duWeights,\n+ 56 float const * dvWeights,\n+ 57 int start, int end);\n+ 58\n+ 59void\n+60OmpEvalStencils(float const * src, BufferDescriptor const &srcDesc,\n+ 61 float * dst, BufferDescriptor const &dstDesc,\n+ 62 float * dstDu, BufferDescriptor const &dstDuDesc,\n+ 63 float * dstDv, BufferDescriptor const &dstDvDesc,\n+ 64 float * dstDuu, BufferDescriptor const &dstDuuDesc,\n+ 65 float * dstDuv, BufferDescriptor const &dstDuvDesc,\n+ 66 float * dstDvv, BufferDescriptor const &dstDvvDesc,\n+ 67 int const * sizes,\n+ 68 int const * offsets,\n+ 69 int const * indices,\n+ 70 float const * weights,\n+ 71 float const * duWeights,\n+ 72 float const * dvWeights,\n+ 73 float const * duuWeights,\n+ 74 float const * duvWeights,\n+ 75 float const * dvvWeights,\n+ 76 int start, int end);\n+ 77\n+ 78} // end namespace Osd\n+ 79\n+ 80} // end namespace OPENSUBDIV_VERSION\n+ 81using namespace OPENSUBDIV_VERSION;\n+ 82\n+ 83} // end namespace OpenSubdiv\n+ 84\n+ 85#endif // OPENSUBDIV3_OSD_OMP_KERNEL_H\n+OpenSubdiv\n+Definition: limits.h:32\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvalStencils\n+void OmpEvalStencils(float const *src, BufferDescriptor const &srcDesc, float\n+*dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets,\n+int const *indices, float const *weights, int start, int end)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n+BufferDescriptor is a struct which describes buffer elements in interleaved\n+data buffers....\n+Definition: bufferDescriptor.h:61\n * opensubdiv\n * osd\n- * opengl.h\n+ * ompKernel.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00887.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00887.html", "unified_diff": "@@ -96,17 +96,17 @@\n
#include "../version.h"
\n #include <bitset>
\n #include <cassert>
\n #include <cstring>
\n #include <vector>
\n #include "../far/topologyRefiner.h"
\n #include "../far/patchTableFactory.h"
\n-#include "../far/stencilTable.h"
\n-#include "../far/stencilTableFactory.h"
\n-#include "../osd/bufferDescriptor.h"
\n+#include "../far/stencilTable.h"
\n+#include "../far/stencilTableFactory.h"
\n+#include "../osd/bufferDescriptor.h"
\n
\n

Go to the source code of this file.

\n \n \n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00887_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00887_source.html", "unified_diff": "@@ -830,38 +830,38 @@\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::GetLocalPointVaryingStencilTable
StencilTable const * GetLocalPointVaryingStencilTable() const
Returns the stencil table to compute local point varying values.
Definition: patchTable.h:735
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTableFactory::Create
static PatchTable * Create(TopologyRefiner const &refiner, Options options=Options(), ConstIndexArray selectedFaces=ConstIndexArray())
Instantiates a PatchTable from a client-provided TopologyRefiner.
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTableFactory::Options::ENDCAP_LEGACY_GREGORY
@ ENDCAP_LEGACY_GREGORY
legacy option for 2.x style Gregory patches (Catmark only)
Definition: patchTableFactory.h:60
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTableFactory::Options::ENDCAP_GREGORY_BASIS
@ ENDCAP_GREGORY_BASIS
use Gregory patches (highest quality, recommended default)
Definition: patchTableFactory.h:59
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTableFactory::Options::ENDCAP_BILINEAR_BASIS
@ ENDCAP_BILINEAR_BASIS
use linear patches (simple quads or tris)
Definition: patchTableFactory.h:57
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTableFactory::Options::ENDCAP_BSPLINE_BASIS
@ ENDCAP_BSPLINE_BASIS
use BSpline-like patches (same patch type as regular)
Definition: patchTableFactory.h:58
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumControlVertices
int GetNumControlVertices() const
Returns the number of control vertices indexed in the table.
Definition: stencilTable.h:161
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumStencils
int GetNumStencils() const
Returns the number of stencils in the table.
Definition: stencilTable.h:156
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::Create
static StencilTable const * Create(TopologyRefiner const &refiner, Options options=Options())
Definition: stencilTableFactory.h:293
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::AppendLocalPointStencilTable
static StencilTable const * AppendLocalPointStencilTable(TopologyRefiner const &refiner, StencilTable const *baseStencilTable, StencilTable const *localPointStencilTable, bool factorize=true)
Definition: stencilTableFactory.h:308
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumVertices
int GetNumVertices() const
Return the number of vertices in this level.
Definition: topologyLevel.h:62
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumControlVertices
int GetNumControlVertices() const
Returns the number of control vertices indexed in the table.
Definition: stencilTable.h:161
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::GetNumStencils
int GetNumStencils() const
Returns the number of stencils in the table.
Definition: stencilTable.h:156
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable
Stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:273
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable
Limit stencil table class wrapping the template for compatibility.
Definition: stencilTable.h:583
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::Create
static StencilTable const * Create(TopologyRefiner const &refiner, Options options=Options())
Definition: stencilTableFactory.h:293
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactory::AppendLocalPointStencilTable
static StencilTable const * AppendLocalPointStencilTable(TopologyRefiner const &refiner, StencilTable const *baseStencilTable, StencilTable const *localPointStencilTable, bool factorize=true)
Definition: stencilTableFactory.h:308
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::GetNumVertices
int GetNumVertices() const
Return the number of vertices in this level.
Definition: topologyLevel.h:62
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner
Stores topology data for a specified set of refinement options.
Definition: topologyRefiner.h:51
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::IsUniform
bool IsUniform() const
Returns true if uniform refinement has been applied.
Definition: topologyRefiner.h:68
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::GetNumFVarChannels
int GetNumFVarChannels() const
Returns the number of face-varying channels in the tables.
Definition: topologyRefiner.h:293
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::RefineUniform
void RefineUniform(UniformOptions options)
Refine the topology uniformly.
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::GetLevel
TopologyLevel const & GetLevel(int level) const
Returns a handle to access data specific to a particular level.
Definition: topologyRefiner.h:95
\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::RefineAdaptive
void RefineAdaptive(AdaptiveOptions options, ConstIndexArray selectedFaces=ConstIndexArray())
Feature Adaptive topology refinement.
\n \n \n \n \n \n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n-\n-\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+\n+\n+\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface::GetFarPatchTable
virtual Far::PatchTable const * GetFarPatchTable() const =0
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface::refineMesh
static void refineMesh(Far::TopologyRefiner &refiner, int level, MeshBitset bits)
Definition: mesh.h:117
\n \n \n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface::UpdateVaryingBuffer
virtual void UpdateVaryingBuffer(float const *varyingData, int startVertex, int numVerts)=0
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface::refineMesh
static void refineMesh(Far::TopologyRefiner &refiner, int level, bool adaptive, bool singleCreasePatch)
Definition: mesh.h:101
\n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00890.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00890.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cpuPatchTable.h File Reference\n+OpenSubdiv: opensubdiv/osd/cpuVertexBuffer.h File Reference\n \n \n \n \n \n \n \n@@ -83,46 +83,41 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
cpuPatchTable.h File Reference
\n+
cpuVertexBuffer.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include <vector>
\n-#include "../far/patchDescriptor.h"
\n-#include "../osd/nonCopyable.h"
\n-#include "../osd/types.h"
\n+#include <cstddef>
\n
\n

Go to the source code of this file.

\n

\n Classes

class  MeshInterface< PATCH_TABLE >
 
\n \n-\n-\n+\n+\n \n

\n Classes

class  CpuPatchTable
 Cpu patch table. More...
class  CpuVertexBuffer
 Concrete vertex buffer class for CPU subdivision. More...
 
\n \n \n \n \n \n-\n-\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n \n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,31 +5,26 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-cpuPatchTable.h File Reference\n+cpuVertexBuffer.h File Reference\n #include \"../version.h\"\n-#include \n-#include \"../far/patchDescriptor.h\"\n-#include \"../osd/nonCopyable.h\"\n-#include \"../osd/types.h\"\n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0CpuPatchTable\n-\u00a0 Cpu patch table. More...\n+class \u00a0CpuVertexBuffer\n+\u00a0 Concrete vertex buffer class for CPU subdivision. More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n-namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n-\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * cpuPatchTable.h\n+ * cpuVertexBuffer.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00890.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00890.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00890 = [\n- [\"CpuPatchTable\", \"a01165.html\", \"a01165\"]\n+ [\"CpuVertexBuffer\", \"a01169.html\", \"a01169\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00890_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00890_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cpuPatchTable.h Source File\n+OpenSubdiv: opensubdiv/osd/cpuVertexBuffer.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
cpuPatchTable.h
\n+
cpuVertexBuffer.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2015 Pixar
\n+
2// Copyright 2013 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,159 +107,73 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H
\n-
26#define OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H
\n+
25#ifndef OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H
\n+
26#define OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30#include <vector>
\n-
31#include "../far/patchDescriptor.h"
\n-
32#include "../osd/nonCopyable.h"
\n-
33#include "../osd/types.h"
\n+
30#include <cstddef>
\n+
31
\n+
32namespace OpenSubdiv {
\n+
33namespace OPENSUBDIV_VERSION {
\n
34
\n-
35#include <vector>
\n+
35namespace Osd {
\n
36
\n-
37namespace OpenSubdiv {
\n-
38namespace OPENSUBDIV_VERSION {
\n-
39
\n-
40namespace Far{
\n-
41 class PatchTable;
\n-
42};
\n-
43
\n-
44namespace Osd {
\n-
45
\n-
59class CpuPatchTable {
\n-
60public:
\n-
61 static CpuPatchTable *Create(const Far::PatchTable *patchTable,
\n-
62 void *deviceContext = NULL) {
\n-
63 (void)deviceContext; // unused
\n-
64 return new CpuPatchTable(patchTable);
\n-
65 }
\n-
66
\n-
67 explicit CpuPatchTable(const Far::PatchTable *patchTable);
\n-
68 ~CpuPatchTable() {}
\n-
69
\n-
70 const PatchArray *GetPatchArrayBuffer() const {
\n-
71 return &_patchArrays[0];
\n-
72 }
\n-
73 const int *GetPatchIndexBuffer() const {
\n-
74 return &_indexBuffer[0];
\n-
75 }
\n-
76 const PatchParam *GetPatchParamBuffer() const {
\n-
77 return &_patchParamBuffer[0];
\n-
78 }
\n-
79
\n-
80 size_t GetNumPatchArrays() const {
\n-
81 return _patchArrays.size();
\n-
82 }
\n-
83 size_t GetPatchIndexSize() const {
\n-
84 return _indexBuffer.size();
\n-
85 }
\n-
86 size_t GetPatchParamSize() const {
\n-
87 return _patchParamBuffer.size();
\n-
88 }
\n-
89
\n-
90 const PatchArray *GetVaryingPatchArrayBuffer() const {
\n-
91 if (_varyingPatchArrays.empty()) {
\n-
92 return NULL;
\n-
93 }
\n-
94 return &_varyingPatchArrays[0];
\n-
95 }
\n-
96 const int *GetVaryingPatchIndexBuffer() const {
\n-
97 if (_varyingIndexBuffer.empty()) {
\n-
98 return NULL;
\n-
99 }
\n-
100 return &_varyingIndexBuffer[0];
\n-
101 }
\n-
102 size_t GetVaryingPatchIndexSize() const {
\n-
103 return _varyingIndexBuffer.size();
\n-
104 }
\n-
105
\n-
106 int GetNumFVarChannels() const {
\n-
107 return (int)_fvarPatchArrays.size();
\n-
108 }
\n-
109 const PatchArray *GetFVarPatchArrayBuffer(int fvarChannel = 0) const {
\n-
110 return &_fvarPatchArrays[fvarChannel][0];
\n-
111 }
\n-
112 const int *GetFVarPatchIndexBuffer(int fvarChannel = 0) const {
\n-
113 return &_fvarIndexBuffers[fvarChannel][0];
\n-
114 }
\n-
115 size_t GetFVarPatchIndexSize(int fvarChannel = 0) const {
\n-
116 return _fvarIndexBuffers[fvarChannel].size();
\n-
117 }
\n-
118 const PatchParam *GetFVarPatchParamBuffer(int fvarChannel= 0) const {
\n-
119 return &_fvarParamBuffers[fvarChannel][0];
\n-
120 }
\n-
121 size_t GetFVarPatchParamSize(int fvarChannel = 0) const {
\n-
122 return _fvarParamBuffers[fvarChannel].size();
\n-
123 }
\n-
124
\n-
125protected:
\n-
126 PatchArrayVector _patchArrays;
\n-
127 std::vector<int> _indexBuffer;
\n-
128 PatchParamVector _patchParamBuffer;
\n-
129
\n-
130 PatchArrayVector _varyingPatchArrays;
\n-
131 std::vector<int> _varyingIndexBuffer;
\n-
132
\n-
133 std::vector< PatchArrayVector > _fvarPatchArrays;
\n-
134 std::vector< std::vector<int> > _fvarIndexBuffers;
\n-
135 std::vector< PatchParamVector > _fvarParamBuffers;
\n-
136};
\n-
137
\n-
138} // end namespace Osd
\n-
139
\n-
140} // end namespace OPENSUBDIV_VERSION
\n-
141using namespace OPENSUBDIV_VERSION;
\n-
142
\n-
143} // end namespace OpenSubdiv
\n-
144
\n-
145#endif // OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H
\n+
42class CpuVertexBuffer {
\n+
43public:
\n+
45 static CpuVertexBuffer * Create(int numElements, int numVertices,
\n+
46 void *deviceContext = NULL);
\n+
47
\n+
49 ~CpuVertexBuffer();
\n+
50
\n+
53 void UpdateData(const float *src, int startVertex, int numVertices,
\n+
54 void *deviceContext = NULL);
\n+
55
\n+
57 int GetNumElements() const;
\n+
58
\n+
60 int GetNumVertices() const;
\n+
61
\n+
63 float * BindCpuBuffer();
\n+
64
\n+
65protected:
\n+
67 CpuVertexBuffer(int numElements, int numVertices);
\n+
68
\n+
69private:
\n+
70 int _numElements;
\n+
71 int _numVertices;
\n+
72 float *_cpuBuffer;
\n+
73};
\n+
74
\n+
75
\n+
76} // end namespace Osd
\n+
77
\n+
78} // end namespace OPENSUBDIV_VERSION
\n+
79using namespace OPENSUBDIV_VERSION;
\n+
80
\n+
81} // end namespace OpenSubdiv
\n+
82
\n+
83#endif // OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParamVector
std::vector< PatchParam > PatchParamVector
Definition: types.h:133
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchArrayBuffer
const PatchArray * GetFVarPatchArrayBuffer(int fvarChannel=0) const
Definition: cpuPatchTable.h:109
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchParamSize
size_t GetFVarPatchParamSize(int fvarChannel=0) const
Definition: cpuPatchTable.h:121
\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::Create
static CpuPatchTable * Create(const Far::PatchTable *patchTable, void *deviceContext=NULL)
Definition: cpuPatchTable.h:61
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarParamBuffers
std::vector< PatchParamVector > _fvarParamBuffers
Definition: cpuPatchTable.h:135
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::CpuPatchTable
CpuPatchTable(const Far::PatchTable *patchTable)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarIndexBuffers
std::vector< std::vector< int > > _fvarIndexBuffers
Definition: cpuPatchTable.h:134
\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchIndexSize
size_t GetFVarPatchIndexSize(int fvarChannel=0) const
Definition: cpuPatchTable.h:115
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchIndexBuffer
const int * GetFVarPatchIndexBuffer(int fvarChannel=0) const
Definition: cpuPatchTable.h:112
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchParamBuffer
const PatchParam * GetFVarPatchParamBuffer(int fvarChannel=0) const
Definition: cpuPatchTable.h:118
\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetVaryingPatchArrayBuffer
const PatchArray * GetVaryingPatchArrayBuffer() const
Definition: cpuPatchTable.h:90
\n-\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarPatchArrays
std::vector< PatchArrayVector > _fvarPatchArrays
Definition: cpuPatchTable.h:133
\n-\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer
Concrete vertex buffer class for CPU subdivision.
Definition: cpuVertexBuffer.h:42
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::GetNumElements
int GetNumElements() const
Returns how many elements defined in this vertex buffer.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::Create
static CpuVertexBuffer * Create(int numElements, int numVertices, void *deviceContext=NULL)
Creator. Returns NULL if error.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::UpdateData
void UpdateData(const float *src, int startVertex, int numVertices, void *deviceContext=NULL)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::GetNumVertices
int GetNumVertices() const
Returns how many vertices allocated in this vertex buffer.
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::BindCpuBuffer
float * BindCpuBuffer()
Returns the address of CPU buffer.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::CpuVertexBuffer
CpuVertexBuffer(int numElements, int numVertices)
Constructor.
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-cpuPatchTable.h\n+cpuVertexBuffer.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2015 Pixar\n+ 2// Copyright 2013 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,215 +30,84 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H\n- 26#define OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H\n+ 25#ifndef OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H\n+ 26#define OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H\n 27\n 28#include \"../version.h\"\n 29\n- 30#include \n- 31#include \"../far/patchDescriptor.h\"\n- 32#include \"../osd/nonCopyable.h\"\n- 33#include \"../osd/types.h\"\n+ 30#include \n+ 31\n+ 32namespace OpenSubdiv {\n+ 33namespace OPENSUBDIV_VERSION {\n 34\n- 35#include \n+ 35namespace Osd {\n 36\n- 37namespace OpenSubdiv {\n- 38namespace OPENSUBDIV_VERSION {\n- 39\n- 40namespace Far{\n- 41 class PatchTable;\n- 42};\n- 43\n- 44namespace Osd {\n- 45\n-59class CpuPatchTable {\n- 60public:\n-61 static CpuPatchTable *Create(const Far::PatchTable *patchTable,\n- 62 void *deviceContext = NULL) {\n- 63 (void)deviceContext; // unused\n- 64 return new CpuPatchTable(patchTable);\n- 65 }\n- 66\n-67 explicit CpuPatchTable(const Far::PatchTable *patchTable);\n-68 ~CpuPatchTable() {}\n- 69\n-70 const PatchArray *GetPatchArrayBuffer() const {\n- 71 return &_patchArrays[0];\n- 72 }\n-73 const int *GetPatchIndexBuffer() const {\n- 74 return &_indexBuffer[0];\n- 75 }\n-76 const PatchParam *GetPatchParamBuffer() const {\n- 77 return &_patchParamBuffer[0];\n- 78 }\n- 79\n-80 size_t GetNumPatchArrays() const {\n- 81 return _patchArrays.size();\n- 82 }\n-83 size_t GetPatchIndexSize() const {\n- 84 return _indexBuffer.size();\n- 85 }\n-86 size_t GetPatchParamSize() const {\n- 87 return _patchParamBuffer.size();\n- 88 }\n- 89\n-90 const PatchArray *GetVaryingPatchArrayBuffer() const {\n- 91 if (_varyingPatchArrays.empty()) {\n- 92 return NULL;\n- 93 }\n- 94 return &_varyingPatchArrays[0];\n- 95 }\n-96 const int *GetVaryingPatchIndexBuffer() const {\n- 97 if (_varyingIndexBuffer.empty()) {\n- 98 return NULL;\n- 99 }\n- 100 return &_varyingIndexBuffer[0];\n- 101 }\n-102 size_t GetVaryingPatchIndexSize() const {\n- 103 return _varyingIndexBuffer.size();\n- 104 }\n- 105\n-106 int GetNumFVarChannels() const {\n- 107 return (int)_fvarPatchArrays.size();\n- 108 }\n-109 const PatchArray *GetFVarPatchArrayBuffer(int fvarChannel = 0) const {\n- 110 return &_fvarPatchArrays[fvarChannel][0];\n- 111 }\n-112 const int *GetFVarPatchIndexBuffer(int fvarChannel = 0) const {\n- 113 return &_fvarIndexBuffers[fvarChannel][0];\n- 114 }\n-115 size_t GetFVarPatchIndexSize(int fvarChannel = 0) const {\n- 116 return _fvarIndexBuffers[fvarChannel].size();\n- 117 }\n-118 const PatchParam *GetFVarPatchParamBuffer(int fvarChannel= 0) const {\n- 119 return &_fvarParamBuffers[fvarChannel][0];\n- 120 }\n-121 size_t GetFVarPatchParamSize(int fvarChannel = 0) const {\n- 122 return _fvarParamBuffers[fvarChannel].size();\n- 123 }\n- 124\n- 125protected:\n-126 PatchArrayVector _patchArrays;\n-127 std::vector _indexBuffer;\n-128 PatchParamVector _patchParamBuffer;\n- 129\n-130 PatchArrayVector _varyingPatchArrays;\n-131 std::vector _varyingIndexBuffer;\n- 132\n-133 std::vector< PatchArrayVector > _fvarPatchArrays;\n-134 std::vector< std::vector > _fvarIndexBuffers;\n-135 std::vector< PatchParamVector > _fvarParamBuffers;\n- 136};\n- 137\n- 138} // end namespace Osd\n- 139\n- 140} // end namespace OPENSUBDIV_VERSION\n- 141using namespace OPENSUBDIV_VERSION;\n- 142\n- 143} // end namespace OpenSubdiv\n- 144\n- 145#endif // OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H\n+42class CpuVertexBuffer {\n+ 43public:\n+45 static CpuVertexBuffer * Create(int numElements, int numVertices,\n+ 46 void *deviceContext = NULL);\n+ 47\n+49 ~CpuVertexBuffer();\n+ 50\n+53 void UpdateData(const float *src, int startVertex, int numVertices,\n+ 54 void *deviceContext = NULL);\n+ 55\n+57 int GetNumElements() const;\n+ 58\n+60 int GetNumVertices() const;\n+ 61\n+63 float * BindCpuBuffer();\n+ 64\n+ 65protected:\n+67 CpuVertexBuffer(int numElements, int numVertices);\n+ 68\n+ 69private:\n+ 70 int _numElements;\n+ 71 int _numVertices;\n+ 72 float *_cpuBuffer;\n+ 73};\n+ 74\n+ 75\n+ 76} // end namespace Osd\n+ 77\n+ 78} // end namespace OPENSUBDIV_VERSION\n+ 79using namespace OPENSUBDIV_VERSION;\n+ 80\n+ 81} // end namespace OpenSubdiv\n+ 82\n+ 83#endif // OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParamVector\n-std::vector< PatchParam > PatchParamVector\n-Definition: types.h:133\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n-std::vector< PatchArray > PatchArrayVector\n-Definition: types.h:132\n-OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n-Container for arrays of parametric patches.\n-Definition: patchTable.h:55\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable\n-Cpu patch table.\n-Definition: cpuPatchTable.h:59\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchArrayBuffer\n-const PatchArray * GetFVarPatchArrayBuffer(int fvarChannel=0) const\n-Definition: cpuPatchTable.h:109\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetVaryingPatchIndexBuffer\n-const int * GetVaryingPatchIndexBuffer() const\n-Definition: cpuPatchTable.h:96\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_varyingIndexBuffer\n-std::vector< int > _varyingIndexBuffer\n-Definition: cpuPatchTable.h:131\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchParamSize\n-size_t GetFVarPatchParamSize(int fvarChannel=0) const\n-Definition: cpuPatchTable.h:121\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::~CpuPatchTable\n-~CpuPatchTable()\n-Definition: cpuPatchTable.h:68\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_patchParamBuffer\n-PatchParamVector _patchParamBuffer\n-Definition: cpuPatchTable.h:128\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetPatchIndexBuffer\n-const int * GetPatchIndexBuffer() const\n-Definition: cpuPatchTable.h:73\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::Create\n-static CpuPatchTable * Create(const Far::PatchTable *patchTable, void\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer\n+Concrete vertex buffer class for CPU subdivision.\n+Definition: cpuVertexBuffer.h:42\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::GetNumElements\n+int GetNumElements() const\n+Returns how many elements defined in this vertex buffer.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::Create\n+static CpuVertexBuffer * Create(int numElements, int numVertices, void\n *deviceContext=NULL)\n-Definition: cpuPatchTable.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarParamBuffers\n-std::vector< PatchParamVector > _fvarParamBuffers\n-Definition: cpuPatchTable.h:135\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::CpuPatchTable\n-CpuPatchTable(const Far::PatchTable *patchTable)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarIndexBuffers\n-std::vector< std::vector< int > > _fvarIndexBuffers\n-Definition: cpuPatchTable.h:134\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetPatchIndexSize\n-size_t GetPatchIndexSize() const\n-Definition: cpuPatchTable.h:83\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_patchArrays\n-PatchArrayVector _patchArrays\n-Definition: cpuPatchTable.h:126\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetPatchParamBuffer\n-const PatchParam * GetPatchParamBuffer() const\n-Definition: cpuPatchTable.h:76\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_indexBuffer\n-std::vector< int > _indexBuffer\n-Definition: cpuPatchTable.h:127\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchIndexSize\n-size_t GetFVarPatchIndexSize(int fvarChannel=0) const\n-Definition: cpuPatchTable.h:115\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchIndexBuffer\n-const int * GetFVarPatchIndexBuffer(int fvarChannel=0) const\n-Definition: cpuPatchTable.h:112\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchParamBuffer\n-const PatchParam * GetFVarPatchParamBuffer(int fvarChannel=0) const\n-Definition: cpuPatchTable.h:118\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetPatchParamSize\n-size_t GetPatchParamSize() const\n-Definition: cpuPatchTable.h:86\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetVaryingPatchIndexSize\n-size_t GetVaryingPatchIndexSize() const\n-Definition: cpuPatchTable.h:102\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetVaryingPatchArrayBuffer\n-const PatchArray * GetVaryingPatchArrayBuffer() const\n-Definition: cpuPatchTable.h:90\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetNumFVarChannels\n-int GetNumFVarChannels() const\n-Definition: cpuPatchTable.h:106\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetNumPatchArrays\n-size_t GetNumPatchArrays() const\n-Definition: cpuPatchTable.h:80\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetPatchArrayBuffer\n-const PatchArray * GetPatchArrayBuffer() const\n-Definition: cpuPatchTable.h:70\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_varyingPatchArrays\n-PatchArrayVector _varyingPatchArrays\n-Definition: cpuPatchTable.h:130\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarPatchArrays\n-std::vector< PatchArrayVector > _fvarPatchArrays\n-Definition: cpuPatchTable.h:133\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArray\n-Definition: types.h:66\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParam\n-Definition: types.h:127\n+Creator. Returns NULL if error.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::UpdateData\n+void UpdateData(const float *src, int startVertex, int numVertices, void\n+*deviceContext=NULL)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::GetNumVertices\n+int GetNumVertices() const\n+Returns how many vertices allocated in this vertex buffer.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::~CpuVertexBuffer\n+~CpuVertexBuffer()\n+Destructor.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::BindCpuBuffer\n+float * BindCpuBuffer()\n+Returns the address of CPU buffer.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::CpuVertexBuffer\n+CpuVertexBuffer(int numElements, int numVertices)\n+Constructor.\n * opensubdiv\n * osd\n- * cpuPatchTable.h\n+ * cpuVertexBuffer.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00893.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00893.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cpuVertexBuffer.h File Reference\n+OpenSubdiv: opensubdiv/osd/cpuPatchTable.h File Reference\n \n \n \n \n \n \n \n@@ -83,41 +83,46 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
cpuVertexBuffer.h File Reference
\n+
cpuPatchTable.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include <cstddef>
\n+#include <vector>
\n+#include "../far/patchDescriptor.h"
\n+#include "../osd/nonCopyable.h"
\n+#include "../osd/types.h"
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n+\n+\n \n

\n Classes

class  CpuVertexBuffer
 Concrete vertex buffer class for CPU subdivision. More...
class  CpuPatchTable
 Cpu patch table. More...
 
\n \n \n \n \n \n+\n+\n \n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Far
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION::Osd
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,26 +5,31 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-cpuVertexBuffer.h File Reference\n+cpuPatchTable.h File Reference\n #include \"../version.h\"\n-#include \n+#include \n+#include \"../far/patchDescriptor.h\"\n+#include \"../osd/nonCopyable.h\"\n+#include \"../osd/types.h\"\n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0CpuVertexBuffer\n-\u00a0 Concrete vertex buffer class for CPU subdivision. More...\n+class \u00a0CpuPatchTable\n+\u00a0 Cpu patch table. More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n+namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Far\n+\u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * cpuVertexBuffer.h\n+ * cpuPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00893.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00893.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00893 = [\n- [\"CpuVertexBuffer\", \"a01169.html\", \"a01169\"]\n+ [\"CpuPatchTable\", \"a01165.html\", \"a01165\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00893_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00893_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cpuVertexBuffer.h Source File\n+OpenSubdiv: opensubdiv/osd/cpuPatchTable.h Source File\n \n \n \n \n \n \n \n@@ -80,19 +80,19 @@\n
\n \n
\n \n
\n-
cpuVertexBuffer.h
\n+
cpuPatchTable.h
\n
\n
\n Go to the documentation of this file.
1//
\n-
2// Copyright 2013 Pixar
\n+
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n
6// compliance with the Apache License and the following modification to it:
\n
7// Section 6. Trademarks. is deleted and replaced with:
\n
8//
\n
9// 6. Trademarks. This License does not grant permission to use the trade
\n@@ -107,73 +107,159 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H
\n-
26#define OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H
\n+
25#ifndef OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H
\n+
26#define OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H
\n
27
\n
28#include "../version.h"
\n
29
\n-
30#include <cstddef>
\n-
31
\n-
32namespace OpenSubdiv {
\n-
33namespace OPENSUBDIV_VERSION {
\n+
30#include <vector>
\n+
31#include "../far/patchDescriptor.h"
\n+
32#include "../osd/nonCopyable.h"
\n+
33#include "../osd/types.h"
\n
34
\n-
35namespace Osd {
\n+
35#include <vector>
\n
36
\n-
42class CpuVertexBuffer {
\n-
43public:
\n-
45 static CpuVertexBuffer * Create(int numElements, int numVertices,
\n-
46 void *deviceContext = NULL);
\n-
47
\n-
49 ~CpuVertexBuffer();
\n-
50
\n-
53 void UpdateData(const float *src, int startVertex, int numVertices,
\n-
54 void *deviceContext = NULL);
\n-
55
\n-
57 int GetNumElements() const;
\n-
58
\n-
60 int GetNumVertices() const;
\n-
61
\n-
63 float * BindCpuBuffer();
\n-
64
\n-
65protected:
\n-
67 CpuVertexBuffer(int numElements, int numVertices);
\n-
68
\n-
69private:
\n-
70 int _numElements;
\n-
71 int _numVertices;
\n-
72 float *_cpuBuffer;
\n-
73};
\n-
74
\n-
75
\n-
76} // end namespace Osd
\n-
77
\n-
78} // end namespace OPENSUBDIV_VERSION
\n-
79using namespace OPENSUBDIV_VERSION;
\n-
80
\n-
81} // end namespace OpenSubdiv
\n-
82
\n-
83#endif // OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H
\n+
37namespace OpenSubdiv {
\n+
38namespace OPENSUBDIV_VERSION {
\n+
39
\n+
40namespace Far{
\n+
41 class PatchTable;
\n+
42};
\n+
43
\n+
44namespace Osd {
\n+
45
\n+
59class CpuPatchTable {
\n+
60public:
\n+
61 static CpuPatchTable *Create(const Far::PatchTable *patchTable,
\n+
62 void *deviceContext = NULL) {
\n+
63 (void)deviceContext; // unused
\n+
64 return new CpuPatchTable(patchTable);
\n+
65 }
\n+
66
\n+
67 explicit CpuPatchTable(const Far::PatchTable *patchTable);
\n+
68 ~CpuPatchTable() {}
\n+
69
\n+
70 const PatchArray *GetPatchArrayBuffer() const {
\n+
71 return &_patchArrays[0];
\n+
72 }
\n+
73 const int *GetPatchIndexBuffer() const {
\n+
74 return &_indexBuffer[0];
\n+
75 }
\n+
76 const PatchParam *GetPatchParamBuffer() const {
\n+
77 return &_patchParamBuffer[0];
\n+
78 }
\n+
79
\n+
80 size_t GetNumPatchArrays() const {
\n+
81 return _patchArrays.size();
\n+
82 }
\n+
83 size_t GetPatchIndexSize() const {
\n+
84 return _indexBuffer.size();
\n+
85 }
\n+
86 size_t GetPatchParamSize() const {
\n+
87 return _patchParamBuffer.size();
\n+
88 }
\n+
89
\n+
90 const PatchArray *GetVaryingPatchArrayBuffer() const {
\n+
91 if (_varyingPatchArrays.empty()) {
\n+
92 return NULL;
\n+
93 }
\n+
94 return &_varyingPatchArrays[0];
\n+
95 }
\n+
96 const int *GetVaryingPatchIndexBuffer() const {
\n+
97 if (_varyingIndexBuffer.empty()) {
\n+
98 return NULL;
\n+
99 }
\n+
100 return &_varyingIndexBuffer[0];
\n+
101 }
\n+
102 size_t GetVaryingPatchIndexSize() const {
\n+
103 return _varyingIndexBuffer.size();
\n+
104 }
\n+
105
\n+
106 int GetNumFVarChannels() const {
\n+
107 return (int)_fvarPatchArrays.size();
\n+
108 }
\n+
109 const PatchArray *GetFVarPatchArrayBuffer(int fvarChannel = 0) const {
\n+
110 return &_fvarPatchArrays[fvarChannel][0];
\n+
111 }
\n+
112 const int *GetFVarPatchIndexBuffer(int fvarChannel = 0) const {
\n+
113 return &_fvarIndexBuffers[fvarChannel][0];
\n+
114 }
\n+
115 size_t GetFVarPatchIndexSize(int fvarChannel = 0) const {
\n+
116 return _fvarIndexBuffers[fvarChannel].size();
\n+
117 }
\n+
118 const PatchParam *GetFVarPatchParamBuffer(int fvarChannel= 0) const {
\n+
119 return &_fvarParamBuffers[fvarChannel][0];
\n+
120 }
\n+
121 size_t GetFVarPatchParamSize(int fvarChannel = 0) const {
\n+
122 return _fvarParamBuffers[fvarChannel].size();
\n+
123 }
\n+
124
\n+
125protected:
\n+
126 PatchArrayVector _patchArrays;
\n+
127 std::vector<int> _indexBuffer;
\n+
128 PatchParamVector _patchParamBuffer;
\n+
129
\n+
130 PatchArrayVector _varyingPatchArrays;
\n+
131 std::vector<int> _varyingIndexBuffer;
\n+
132
\n+
133 std::vector< PatchArrayVector > _fvarPatchArrays;
\n+
134 std::vector< std::vector<int> > _fvarIndexBuffers;
\n+
135 std::vector< PatchParamVector > _fvarParamBuffers;
\n+
136};
\n+
137
\n+
138} // end namespace Osd
\n+
139
\n+
140} // end namespace OPENSUBDIV_VERSION
\n+
141using namespace OPENSUBDIV_VERSION;
\n+
142
\n+
143} // end namespace OpenSubdiv
\n+
144
\n+
145#endif // OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer
Concrete vertex buffer class for CPU subdivision.
Definition: cpuVertexBuffer.h:42
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::GetNumElements
int GetNumElements() const
Returns how many elements defined in this vertex buffer.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::Create
static CpuVertexBuffer * Create(int numElements, int numVertices, void *deviceContext=NULL)
Creator. Returns NULL if error.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::UpdateData
void UpdateData(const float *src, int startVertex, int numVertices, void *deviceContext=NULL)
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::GetNumVertices
int GetNumVertices() const
Returns how many vertices allocated in this vertex buffer.
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::BindCpuBuffer
float * BindCpuBuffer()
Returns the address of CPU buffer.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::CpuVertexBuffer
CpuVertexBuffer(int numElements, int numVertices)
Constructor.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParamVector
std::vector< PatchParam > PatchParamVector
Definition: types.h:133
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable
Container for arrays of parametric patches.
Definition: patchTable.h:55
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchArrayBuffer
const PatchArray * GetFVarPatchArrayBuffer(int fvarChannel=0) const
Definition: cpuPatchTable.h:109
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchParamSize
size_t GetFVarPatchParamSize(int fvarChannel=0) const
Definition: cpuPatchTable.h:121
\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::Create
static CpuPatchTable * Create(const Far::PatchTable *patchTable, void *deviceContext=NULL)
Definition: cpuPatchTable.h:61
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarParamBuffers
std::vector< PatchParamVector > _fvarParamBuffers
Definition: cpuPatchTable.h:135
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::CpuPatchTable
CpuPatchTable(const Far::PatchTable *patchTable)
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarIndexBuffers
std::vector< std::vector< int > > _fvarIndexBuffers
Definition: cpuPatchTable.h:134
\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchIndexSize
size_t GetFVarPatchIndexSize(int fvarChannel=0) const
Definition: cpuPatchTable.h:115
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchIndexBuffer
const int * GetFVarPatchIndexBuffer(int fvarChannel=0) const
Definition: cpuPatchTable.h:112
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchParamBuffer
const PatchParam * GetFVarPatchParamBuffer(int fvarChannel=0) const
Definition: cpuPatchTable.h:118
\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetVaryingPatchArrayBuffer
const PatchArray * GetVaryingPatchArrayBuffer() const
Definition: cpuPatchTable.h:90
\n+\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarPatchArrays
std::vector< PatchArrayVector > _fvarPatchArrays
Definition: cpuPatchTable.h:133
\n+\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,18 +4,18 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-cpuVertexBuffer.h\n+cpuPatchTable.h\n Go_to_the_documentation_of_this_file.\n 1//\n- 2// Copyright 2013 Pixar\n+ 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n 7// Section 6. Trademarks. is deleted and replaced with:\n 8//\n 9// 6. Trademarks. This License does not grant permission to use the trade\n@@ -30,84 +30,215 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H\n- 26#define OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H\n+ 25#ifndef OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H\n+ 26#define OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H\n 27\n 28#include \"../version.h\"\n 29\n- 30#include \n- 31\n- 32namespace OpenSubdiv {\n- 33namespace OPENSUBDIV_VERSION {\n+ 30#include \n+ 31#include \"../far/patchDescriptor.h\"\n+ 32#include \"../osd/nonCopyable.h\"\n+ 33#include \"../osd/types.h\"\n 34\n- 35namespace Osd {\n+ 35#include \n 36\n-42class CpuVertexBuffer {\n- 43public:\n-45 static CpuVertexBuffer * Create(int numElements, int numVertices,\n- 46 void *deviceContext = NULL);\n- 47\n-49 ~CpuVertexBuffer();\n- 50\n-53 void UpdateData(const float *src, int startVertex, int numVertices,\n- 54 void *deviceContext = NULL);\n- 55\n-57 int GetNumElements() const;\n- 58\n-60 int GetNumVertices() const;\n- 61\n-63 float * BindCpuBuffer();\n- 64\n- 65protected:\n-67 CpuVertexBuffer(int numElements, int numVertices);\n- 68\n- 69private:\n- 70 int _numElements;\n- 71 int _numVertices;\n- 72 float *_cpuBuffer;\n- 73};\n- 74\n- 75\n- 76} // end namespace Osd\n- 77\n- 78} // end namespace OPENSUBDIV_VERSION\n- 79using namespace OPENSUBDIV_VERSION;\n- 80\n- 81} // end namespace OpenSubdiv\n- 82\n- 83#endif // OPENSUBDIV3_OSD_CPU_VERTEX_BUFFER_H\n+ 37namespace OpenSubdiv {\n+ 38namespace OPENSUBDIV_VERSION {\n+ 39\n+ 40namespace Far{\n+ 41 class PatchTable;\n+ 42};\n+ 43\n+ 44namespace Osd {\n+ 45\n+59class CpuPatchTable {\n+ 60public:\n+61 static CpuPatchTable *Create(const Far::PatchTable *patchTable,\n+ 62 void *deviceContext = NULL) {\n+ 63 (void)deviceContext; // unused\n+ 64 return new CpuPatchTable(patchTable);\n+ 65 }\n+ 66\n+67 explicit CpuPatchTable(const Far::PatchTable *patchTable);\n+68 ~CpuPatchTable() {}\n+ 69\n+70 const PatchArray *GetPatchArrayBuffer() const {\n+ 71 return &_patchArrays[0];\n+ 72 }\n+73 const int *GetPatchIndexBuffer() const {\n+ 74 return &_indexBuffer[0];\n+ 75 }\n+76 const PatchParam *GetPatchParamBuffer() const {\n+ 77 return &_patchParamBuffer[0];\n+ 78 }\n+ 79\n+80 size_t GetNumPatchArrays() const {\n+ 81 return _patchArrays.size();\n+ 82 }\n+83 size_t GetPatchIndexSize() const {\n+ 84 return _indexBuffer.size();\n+ 85 }\n+86 size_t GetPatchParamSize() const {\n+ 87 return _patchParamBuffer.size();\n+ 88 }\n+ 89\n+90 const PatchArray *GetVaryingPatchArrayBuffer() const {\n+ 91 if (_varyingPatchArrays.empty()) {\n+ 92 return NULL;\n+ 93 }\n+ 94 return &_varyingPatchArrays[0];\n+ 95 }\n+96 const int *GetVaryingPatchIndexBuffer() const {\n+ 97 if (_varyingIndexBuffer.empty()) {\n+ 98 return NULL;\n+ 99 }\n+ 100 return &_varyingIndexBuffer[0];\n+ 101 }\n+102 size_t GetVaryingPatchIndexSize() const {\n+ 103 return _varyingIndexBuffer.size();\n+ 104 }\n+ 105\n+106 int GetNumFVarChannels() const {\n+ 107 return (int)_fvarPatchArrays.size();\n+ 108 }\n+109 const PatchArray *GetFVarPatchArrayBuffer(int fvarChannel = 0) const {\n+ 110 return &_fvarPatchArrays[fvarChannel][0];\n+ 111 }\n+112 const int *GetFVarPatchIndexBuffer(int fvarChannel = 0) const {\n+ 113 return &_fvarIndexBuffers[fvarChannel][0];\n+ 114 }\n+115 size_t GetFVarPatchIndexSize(int fvarChannel = 0) const {\n+ 116 return _fvarIndexBuffers[fvarChannel].size();\n+ 117 }\n+118 const PatchParam *GetFVarPatchParamBuffer(int fvarChannel= 0) const {\n+ 119 return &_fvarParamBuffers[fvarChannel][0];\n+ 120 }\n+121 size_t GetFVarPatchParamSize(int fvarChannel = 0) const {\n+ 122 return _fvarParamBuffers[fvarChannel].size();\n+ 123 }\n+ 124\n+ 125protected:\n+126 PatchArrayVector _patchArrays;\n+127 std::vector _indexBuffer;\n+128 PatchParamVector _patchParamBuffer;\n+ 129\n+130 PatchArrayVector _varyingPatchArrays;\n+131 std::vector _varyingIndexBuffer;\n+ 132\n+133 std::vector< PatchArrayVector > _fvarPatchArrays;\n+134 std::vector< std::vector > _fvarIndexBuffers;\n+135 std::vector< PatchParamVector > _fvarParamBuffers;\n+ 136};\n+ 137\n+ 138} // end namespace Osd\n+ 139\n+ 140} // end namespace OPENSUBDIV_VERSION\n+ 141using namespace OPENSUBDIV_VERSION;\n+ 142\n+ 143} // end namespace OpenSubdiv\n+ 144\n+ 145#endif // OPENSUBDIV3_OSD_CPU_PATCH_TABLE_H\n OpenSubdiv\n Definition: limits.h:32\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer\n-Concrete vertex buffer class for CPU subdivision.\n-Definition: cpuVertexBuffer.h:42\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::GetNumElements\n-int GetNumElements() const\n-Returns how many elements defined in this vertex buffer.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::Create\n-static CpuVertexBuffer * Create(int numElements, int numVertices, void\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParamVector\n+std::vector< PatchParam > PatchParamVector\n+Definition: types.h:133\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector\n+std::vector< PatchArray > PatchArrayVector\n+Definition: types.h:132\n+OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable\n+Container for arrays of parametric patches.\n+Definition: patchTable.h:55\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable\n+Cpu patch table.\n+Definition: cpuPatchTable.h:59\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchArrayBuffer\n+const PatchArray * GetFVarPatchArrayBuffer(int fvarChannel=0) const\n+Definition: cpuPatchTable.h:109\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetVaryingPatchIndexBuffer\n+const int * GetVaryingPatchIndexBuffer() const\n+Definition: cpuPatchTable.h:96\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_varyingIndexBuffer\n+std::vector< int > _varyingIndexBuffer\n+Definition: cpuPatchTable.h:131\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchParamSize\n+size_t GetFVarPatchParamSize(int fvarChannel=0) const\n+Definition: cpuPatchTable.h:121\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::~CpuPatchTable\n+~CpuPatchTable()\n+Definition: cpuPatchTable.h:68\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_patchParamBuffer\n+PatchParamVector _patchParamBuffer\n+Definition: cpuPatchTable.h:128\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetPatchIndexBuffer\n+const int * GetPatchIndexBuffer() const\n+Definition: cpuPatchTable.h:73\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::Create\n+static CpuPatchTable * Create(const Far::PatchTable *patchTable, void\n *deviceContext=NULL)\n-Creator. Returns NULL if error.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::UpdateData\n-void UpdateData(const float *src, int startVertex, int numVertices, void\n-*deviceContext=NULL)\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::GetNumVertices\n-int GetNumVertices() const\n-Returns how many vertices allocated in this vertex buffer.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::~CpuVertexBuffer\n-~CpuVertexBuffer()\n-Destructor.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::BindCpuBuffer\n-float * BindCpuBuffer()\n-Returns the address of CPU buffer.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::CpuVertexBuffer\n-CpuVertexBuffer(int numElements, int numVertices)\n-Constructor.\n+Definition: cpuPatchTable.h:61\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarParamBuffers\n+std::vector< PatchParamVector > _fvarParamBuffers\n+Definition: cpuPatchTable.h:135\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::CpuPatchTable\n+CpuPatchTable(const Far::PatchTable *patchTable)\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarIndexBuffers\n+std::vector< std::vector< int > > _fvarIndexBuffers\n+Definition: cpuPatchTable.h:134\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetPatchIndexSize\n+size_t GetPatchIndexSize() const\n+Definition: cpuPatchTable.h:83\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_patchArrays\n+PatchArrayVector _patchArrays\n+Definition: cpuPatchTable.h:126\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetPatchParamBuffer\n+const PatchParam * GetPatchParamBuffer() const\n+Definition: cpuPatchTable.h:76\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_indexBuffer\n+std::vector< int > _indexBuffer\n+Definition: cpuPatchTable.h:127\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchIndexSize\n+size_t GetFVarPatchIndexSize(int fvarChannel=0) const\n+Definition: cpuPatchTable.h:115\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchIndexBuffer\n+const int * GetFVarPatchIndexBuffer(int fvarChannel=0) const\n+Definition: cpuPatchTable.h:112\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetFVarPatchParamBuffer\n+const PatchParam * GetFVarPatchParamBuffer(int fvarChannel=0) const\n+Definition: cpuPatchTable.h:118\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetPatchParamSize\n+size_t GetPatchParamSize() const\n+Definition: cpuPatchTable.h:86\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetVaryingPatchIndexSize\n+size_t GetVaryingPatchIndexSize() const\n+Definition: cpuPatchTable.h:102\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetVaryingPatchArrayBuffer\n+const PatchArray * GetVaryingPatchArrayBuffer() const\n+Definition: cpuPatchTable.h:90\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetNumFVarChannels\n+int GetNumFVarChannels() const\n+Definition: cpuPatchTable.h:106\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetNumPatchArrays\n+size_t GetNumPatchArrays() const\n+Definition: cpuPatchTable.h:80\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::GetPatchArrayBuffer\n+const PatchArray * GetPatchArrayBuffer() const\n+Definition: cpuPatchTable.h:70\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_varyingPatchArrays\n+PatchArrayVector _varyingPatchArrays\n+Definition: cpuPatchTable.h:130\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::_fvarPatchArrays\n+std::vector< PatchArrayVector > _fvarPatchArrays\n+Definition: cpuPatchTable.h:133\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArray\n+Definition: types.h:66\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParam\n+Definition: types.h:127\n * opensubdiv\n * osd\n- * cpuVertexBuffer.h\n+ * cpuPatchTable.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00896.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00896.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/bufferDescriptor.h File Reference\n+OpenSubdiv: opensubdiv/osd/cpuEvaluator.h File Reference\n \n \n \n \n \n \n \n@@ -83,26 +83,27 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
bufferDescriptor.h File Reference
\n+
cpuEvaluator.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include <string.h>
\n+#include "../osd/bufferDescriptor.h"
\n+#include "../osd/types.h"
\n+#include <cstddef>
\n
\n

Go to the source code of this file.

\n \n \n-\n-\n+\n \n

\n Classes

struct  BufferDescriptor
 BufferDescriptor is a struct which describes buffer elements in interleaved data buffers. Almost all Osd Evaluator APIs take BufferDescriptors along with device-specific buffer objects. More...
class  CpuEvaluator
 
\n \n \n \n \n@@ -111,13 +112,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,28 +5,27 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-bufferDescriptor.h File Reference\n+cpuEvaluator.h File Reference\n #include \"../version.h\"\n-#include \n+#include \"../osd/bufferDescriptor.h\"\n+#include \"../osd/types.h\"\n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n-struct \u00a0BufferDescriptor\n- BufferDescriptor is a struct which describes buffer elements in\n-\u00a0 interleaved data buffers. Almost all Osd Evaluator APIs take\n- BufferDescriptors along with device-specific buffer objects. More...\n+class \u00a0CpuEvaluator\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * bufferDescriptor.h\n+ * cpuEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00896.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00896.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00896 = [\n- [\"BufferDescriptor\", \"a01133.html\", \"a01133\"]\n+ [\"CpuEvaluator\", \"a01157.html\", null]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00896_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00896_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/bufferDescriptor.h Source File\n+OpenSubdiv: opensubdiv/osd/cpuEvaluator.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
bufferDescriptor.h
\n+
cpuEvaluator.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,98 +107,501 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H
\n-
26#define OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H
\n+
25#ifndef OPENSUBDIV3_OSD_CPU_EVALUATOR_H
\n+
26#define OPENSUBDIV3_OSD_CPU_EVALUATOR_H
\n
27
\n
28#include "../version.h"
\n-
29#include <string.h>
\n-
30
\n-
31namespace OpenSubdiv {
\n-
32namespace OPENSUBDIV_VERSION {
\n+
29#include "../osd/bufferDescriptor.h"
\n+
30#include "../osd/types.h"
\n+
31
\n+
32#include <cstddef>
\n
33
\n-
34namespace Osd {
\n-
35
\n+
34namespace OpenSubdiv {
\n+
35namespace OPENSUBDIV_VERSION {
\n+
36
\n+
37namespace Osd {
\n+
38
\n+
39class CpuEvaluator {
\n+
40public:
\n
46
\n-
47// example:
\n-
48// n
\n-
49// -----+----------------------------------------+-------------------------
\n-
50// | vertex 0 |
\n-
51// -----+----------------------------------------+-------------------------
\n-
52// | X Y Z R G B A Xu Yu Zu Xv Yv Zv |
\n-
53// -----+----------------------------------------+-------------------------
\n-
54// <------------- stride = 13 -------------->
\n-
55//
\n-
56// - XYZ (offset = n+0, length = 3, stride = 13)
\n-
57// - RGBA (offset = n+3, length = 4, stride = 13)
\n-
58// - uTangent (offset = n+7, length = 3, stride = 13)
\n-
59// - vTangent (offset = n+10, length = 3, stride = 13)
\n-
60//
\n-
61struct BufferDescriptor {
\n-
62
\n-
64 BufferDescriptor() : offset(0), length(0), stride(0) { }
\n-
65
\n-
67 BufferDescriptor(int o, int l, int s) : offset(o), length(l), stride(s) { }
\n-
68
\n-
70 int GetLocalOffset() const {
\n-
71 return stride > 0 ? offset % stride : 0;
\n-
72 }
\n-
73
\n-
75 bool IsValid() const {
\n-
76 return ((length > 0) &&
\n-
77 (length <= stride - GetLocalOffset()));
\n-
78 }
\n-
79
\n-
81 void Reset() {
\n-
82 offset = length = stride = 0;
\n-
83 }
\n+
71 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
72 static bool EvalStencils(
\n+
73 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
74 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
75 STENCIL_TABLE const *stencilTable,
\n+
76 const CpuEvaluator *instance = NULL,
\n+
77 void * deviceContext = NULL) {
\n+
78
\n+
79 (void)instance; // unused
\n+
80 (void)deviceContext; // unused
\n+
81
\n+
82 if (stencilTable->GetNumStencils() == 0)
\n+
83 return false;
\n
84
\n-
86 bool operator == (BufferDescriptor const &other) const {
\n-
87 return (offset == other.offset &&
\n-
88 length == other.length &&
\n-
89 stride == other.stride);
\n-
90 }
\n-
91
\n-
93 bool operator != (BufferDescriptor const &other) const {
\n-
94 return !(this->operator==(other));
\n-
95 }
\n-
96
\n-
98 int offset;
\n-
100 int length;
\n-
102 int stride;
\n-
103};
\n-
104
\n-
105} // end namespace Osd
\n-
106
\n-
107} // end namespace OPENSUBDIV_VERSION
\n-
108using namespace OPENSUBDIV_VERSION;
\n-
109
\n-
110} // end namespace OpenSubdiv
\n-
111
\n-
112#endif // OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H
\n+
85 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
86 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
87 &stencilTable->GetSizes()[0],
\n+
88 &stencilTable->GetOffsets()[0],
\n+
89 &stencilTable->GetControlIndices()[0],
\n+
90 &stencilTable->GetWeights()[0],
\n+
91 /*start = */ 0,
\n+
92 /*end = */ stencilTable->GetNumStencils());
\n+
93 }
\n+
94
\n+
121 static bool EvalStencils(
\n+
122 const float *src, BufferDescriptor const &srcDesc,
\n+
123 float *dst, BufferDescriptor const &dstDesc,
\n+
124 const int * sizes,
\n+
125 const int * offsets,
\n+
126 const int * indices,
\n+
127 const float * weights,
\n+
128 int start, int end);
\n+
129
\n+
167 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
168 static bool EvalStencils(
\n+
169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
171 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
172 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
173 STENCIL_TABLE const *stencilTable,
\n+
174 const CpuEvaluator *instance = NULL,
\n+
175 void * deviceContext = NULL) {
\n+
176
\n+
177 (void)instance; // unused
\n+
178 (void)deviceContext; // unused
\n+
179
\n+
180 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
181 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
182 duBuffer->BindCpuBuffer(), duDesc,
\n+
183 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
184 &stencilTable->GetSizes()[0],
\n+
185 &stencilTable->GetOffsets()[0],
\n+
186 &stencilTable->GetControlIndices()[0],
\n+
187 &stencilTable->GetWeights()[0],
\n+
188 &stencilTable->GetDuWeights()[0],
\n+
189 &stencilTable->GetDvWeights()[0],
\n+
190 /*start = */ 0,
\n+
191 /*end = */ stencilTable->GetNumStencils());
\n+
192 }
\n+
193
\n+
234 static bool EvalStencils(
\n+
235 const float *src, BufferDescriptor const &srcDesc,
\n+
236 float *dst, BufferDescriptor const &dstDesc,
\n+
237 float *du, BufferDescriptor const &duDesc,
\n+
238 float *dv, BufferDescriptor const &dvDesc,
\n+
239 const int * sizes,
\n+
240 const int * offsets,
\n+
241 const int * indices,
\n+
242 const float * weights,
\n+
243 const float * duWeights,
\n+
244 const float * dvWeights,
\n+
245 int start, int end);
\n+
246
\n+
302 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n+
303 static bool EvalStencils(
\n+
304 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
305 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
306 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
307 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
308 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
309 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
310 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
311 STENCIL_TABLE const *stencilTable,
\n+
312 const CpuEvaluator *instance = NULL,
\n+
313 void * deviceContext = NULL) {
\n+
314
\n+
315 (void)instance; // unused
\n+
316 (void)deviceContext; // unused
\n+
317
\n+
318 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
319 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
320 duBuffer->BindCpuBuffer(), duDesc,
\n+
321 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
322 duuBuffer->BindCpuBuffer(), duuDesc,
\n+
323 duvBuffer->BindCpuBuffer(), duvDesc,
\n+
324 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n+
325 &stencilTable->GetSizes()[0],
\n+
326 &stencilTable->GetOffsets()[0],
\n+
327 &stencilTable->GetControlIndices()[0],
\n+
328 &stencilTable->GetWeights()[0],
\n+
329 &stencilTable->GetDuWeights()[0],
\n+
330 &stencilTable->GetDvWeights()[0],
\n+
331 &stencilTable->GetDuuWeights()[0],
\n+
332 &stencilTable->GetDuvWeights()[0],
\n+
333 &stencilTable->GetDvvWeights()[0],
\n+
334 /*start = */ 0,
\n+
335 /*end = */ stencilTable->GetNumStencils());
\n+
336 }
\n+
337
\n+
399 static bool EvalStencils(
\n+
400 const float *src, BufferDescriptor const &srcDesc,
\n+
401 float *dst, BufferDescriptor const &dstDesc,
\n+
402 float *du, BufferDescriptor const &duDesc,
\n+
403 float *dv, BufferDescriptor const &dvDesc,
\n+
404 float *duu, BufferDescriptor const &duuDesc,
\n+
405 float *duv, BufferDescriptor const &duvDesc,
\n+
406 float *dvv, BufferDescriptor const &dvvDesc,
\n+
407 const int * sizes,
\n+
408 const int * offsets,
\n+
409 const int * indices,
\n+
410 const float * weights,
\n+
411 const float * duWeights,
\n+
412 const float * dvWeights,
\n+
413 const float * duuWeights,
\n+
414 const float * duvWeights,
\n+
415 const float * dvvWeights,
\n+
416 int start, int end);
\n+
417
\n+
423
\n+
452 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
453 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
454 static bool EvalPatches(
\n+
455 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
456 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
457 int numPatchCoords,
\n+
458 PATCHCOORD_BUFFER *patchCoords,
\n+
459 PATCH_TABLE *patchTable,
\n+
460 CpuEvaluator const *instance = NULL,
\n+
461 void * deviceContext = NULL) {
\n+
462
\n+
463 (void)instance; // unused
\n+
464 (void)deviceContext; // unused
\n+
465
\n+
466 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
467 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
468 numPatchCoords,
\n+
469 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
470 patchTable->GetPatchArrayBuffer(),
\n+
471 patchTable->GetPatchIndexBuffer(),
\n+
472 patchTable->GetPatchParamBuffer());
\n+
473 }
\n+
474
\n+
515 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
516 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
517 static bool EvalPatches(
\n+
518 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
519 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
520 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
521 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
522 int numPatchCoords,
\n+
523 PATCHCOORD_BUFFER *patchCoords,
\n+
524 PATCH_TABLE *patchTable,
\n+
525 CpuEvaluator const *instance = NULL,
\n+
526 void * deviceContext = NULL) {
\n+
527
\n+
528 (void)instance; // unused
\n+
529 (void)deviceContext; // unused
\n+
530
\n+
531 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.
\n+
532 // ideally all buffer classes should have templated by datatype
\n+
533 // so that downcast isn't needed there.
\n+
534 // (e.g. Osd::CpuBuffer<PatchCoord> )
\n+
535 //
\n+
536 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
537 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
538 duBuffer->BindCpuBuffer(), duDesc,
\n+
539 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
540 numPatchCoords,
\n+
541 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
542 patchTable->GetPatchArrayBuffer(),
\n+
543 patchTable->GetPatchIndexBuffer(),
\n+
544 patchTable->GetPatchParamBuffer());
\n+
545 }
\n+
546
\n+
605 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
606 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
607 static bool EvalPatches(
\n+
608 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
609 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
610 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
611 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
612 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
613 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
614 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
615 int numPatchCoords,
\n+
616 PATCHCOORD_BUFFER *patchCoords,
\n+
617 PATCH_TABLE *patchTable,
\n+
618 CpuEvaluator const *instance = NULL,
\n+
619 void * deviceContext = NULL) {
\n+
620
\n+
621 (void)instance; // unused
\n+
622 (void)deviceContext; // unused
\n+
623
\n+
624 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.
\n+
625 // ideally all buffer classes should have templated by datatype
\n+
626 // so that downcast isn't needed there.
\n+
627 // (e.g. Osd::CpuBuffer<PatchCoord> )
\n+
628 //
\n+
629 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
630 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
631 duBuffer->BindCpuBuffer(), duDesc,
\n+
632 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
633 duuBuffer->BindCpuBuffer(), duuDesc,
\n+
634 duvBuffer->BindCpuBuffer(), duvDesc,
\n+
635 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n+
636 numPatchCoords,
\n+
637 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
638 patchTable->GetPatchArrayBuffer(),
\n+
639 patchTable->GetPatchIndexBuffer(),
\n+
640 patchTable->GetPatchParamBuffer());
\n+
641 }
\n+
642
\n+
670 static bool EvalPatches(
\n+
671 const float *src, BufferDescriptor const &srcDesc,
\n+
672 float *dst, BufferDescriptor const &dstDesc,
\n+
673 int numPatchCoords,
\n+
674 const PatchCoord *patchCoords,
\n+
675 const PatchArray *patchArrays,
\n+
676 const int *patchIndexBuffer,
\n+
677 const PatchParam *patchParamBuffer);
\n+
678
\n+
716 static bool EvalPatches(
\n+
717 const float *src, BufferDescriptor const &srcDesc,
\n+
718 float *dst, BufferDescriptor const &dstDesc,
\n+
719 float *du, BufferDescriptor const &duDesc,
\n+
720 float *dv, BufferDescriptor const &dvDesc,
\n+
721 int numPatchCoords,
\n+
722 PatchCoord const *patchCoords,
\n+
723 PatchArray const *patchArrays,
\n+
724 const int *patchIndexBuffer,
\n+
725 PatchParam const *patchParamBuffer);
\n+
726
\n+
779 static bool EvalPatches(
\n+
780 const float *src, BufferDescriptor const &srcDesc,
\n+
781 float *dst, BufferDescriptor const &dstDesc,
\n+
782 float *du, BufferDescriptor const &duDesc,
\n+
783 float *dv, BufferDescriptor const &dvDesc,
\n+
784 float *duu, BufferDescriptor const &duuDesc,
\n+
785 float *duv, BufferDescriptor const &duvDesc,
\n+
786 float *dvv, BufferDescriptor const &dvvDesc,
\n+
787 int numPatchCoords,
\n+
788 PatchCoord const *patchCoords,
\n+
789 PatchArray const *patchArrays,
\n+
790 const int *patchIndexBuffer,
\n+
791 PatchParam const *patchParamBuffer);
\n+
792
\n+
821 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
822 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
823 static bool EvalPatchesVarying(
\n+
824 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
825 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
826 int numPatchCoords,
\n+
827 PATCHCOORD_BUFFER *patchCoords,
\n+
828 PATCH_TABLE *patchTable,
\n+
829 CpuEvaluator const *instance = NULL,
\n+
830 void * deviceContext = NULL) {
\n+
831
\n+
832 (void)instance; // unused
\n+
833 (void)deviceContext; // unused
\n+
834
\n+
835 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
836 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
837 numPatchCoords,
\n+
838 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
839 patchTable->GetVaryingPatchArrayBuffer(),
\n+
840 patchTable->GetVaryingPatchIndexBuffer(),
\n+
841 patchTable->GetPatchParamBuffer());
\n+
842 }
\n+
843
\n+
884 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
885 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
886 static bool EvalPatchesVarying(
\n+
887 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
888 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
889 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
890 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
891 int numPatchCoords,
\n+
892 PATCHCOORD_BUFFER *patchCoords,
\n+
893 PATCH_TABLE *patchTable,
\n+
894 CpuEvaluator const *instance = NULL,
\n+
895 void * deviceContext = NULL) {
\n+
896
\n+
897 (void)instance; // unused
\n+
898 (void)deviceContext; // unused
\n+
899
\n+
900 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
901 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
902 duBuffer->BindCpuBuffer(), duDesc,
\n+
903 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
904 numPatchCoords,
\n+
905 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
906 patchTable->GetVaryingPatchArrayBuffer(),
\n+
907 patchTable->GetVaryingPatchIndexBuffer(),
\n+
908 patchTable->GetPatchParamBuffer());
\n+
909 }
\n+
910
\n+
969 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
970 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
971 static bool EvalPatchesVarying(
\n+
972 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
973 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
974 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
975 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
976 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
977 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
978 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
979 int numPatchCoords,
\n+
980 PATCHCOORD_BUFFER *patchCoords,
\n+
981 PATCH_TABLE *patchTable,
\n+
982 CpuEvaluator const *instance = NULL,
\n+
983 void * deviceContext = NULL) {
\n+
984
\n+
985 (void)instance; // unused
\n+
986 (void)deviceContext; // unused
\n+
987
\n+
988 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
989 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
990 duBuffer->BindCpuBuffer(), duDesc,
\n+
991 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
992 duuBuffer->BindCpuBuffer(), duuDesc,
\n+
993 duvBuffer->BindCpuBuffer(), duvDesc,
\n+
994 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n+
995 numPatchCoords,
\n+
996 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
997 patchTable->GetVaryingPatchArrayBuffer(),
\n+
998 patchTable->GetVaryingPatchIndexBuffer(),
\n+
999 patchTable->GetPatchParamBuffer());
\n+
1000 }
\n+
1001
\n+
1032 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1033 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1034 static bool EvalPatchesFaceVarying(
\n+
1035 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1036 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1037 int numPatchCoords,
\n+
1038 PATCHCOORD_BUFFER *patchCoords,
\n+
1039 PATCH_TABLE *patchTable,
\n+
1040 int fvarChannel,
\n+
1041 CpuEvaluator const *instance = NULL,
\n+
1042 void * deviceContext = NULL) {
\n+
1043
\n+
1044 (void)instance; // unused
\n+
1045 (void)deviceContext; // unused
\n+
1046
\n+
1047 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
1048 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
1049 numPatchCoords,
\n+
1050 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
1051 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
1052 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1053 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
1054 }
\n+
1055
\n+
1098 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1099 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1100 static bool EvalPatchesFaceVarying(
\n+
1101 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1102 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1103 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1104 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1105 int numPatchCoords,
\n+
1106 PATCHCOORD_BUFFER *patchCoords,
\n+
1107 PATCH_TABLE *patchTable,
\n+
1108 int fvarChannel,
\n+
1109 CpuEvaluator const *instance = NULL,
\n+
1110 void * deviceContext = NULL) {
\n+
1111
\n+
1112 (void)instance; // unused
\n+
1113 (void)deviceContext; // unused
\n+
1114
\n+
1115 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
1116 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
1117 duBuffer->BindCpuBuffer(), duDesc,
\n+
1118 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
1119 numPatchCoords,
\n+
1120 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
1121 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
1122 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1123 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
1124 }
\n+
1125
\n+
1186 template <typename SRC_BUFFER, typename DST_BUFFER,
\n+
1187 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n+
1188 static bool EvalPatchesFaceVarying(
\n+
1189 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n+
1190 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n+
1191 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n+
1192 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n+
1193 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n+
1194 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n+
1195 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n+
1196 int numPatchCoords,
\n+
1197 PATCHCOORD_BUFFER *patchCoords,
\n+
1198 PATCH_TABLE *patchTable,
\n+
1199 int fvarChannel,
\n+
1200 CpuEvaluator const *instance = NULL,
\n+
1201 void * deviceContext = NULL) {
\n+
1202
\n+
1203 (void)instance; // unused
\n+
1204 (void)deviceContext; // unused
\n+
1205
\n+
1206 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n+
1207 dstBuffer->BindCpuBuffer(), dstDesc,
\n+
1208 duBuffer->BindCpuBuffer(), duDesc,
\n+
1209 dvBuffer->BindCpuBuffer(), dvDesc,
\n+
1210 duuBuffer->BindCpuBuffer(), duuDesc,
\n+
1211 duvBuffer->BindCpuBuffer(), duvDesc,
\n+
1212 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n+
1213 numPatchCoords,
\n+
1214 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n+
1215 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n+
1216 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n+
1217 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n+
1218 }
\n+
1219
\n+
1225
\n+
1227 static void Synchronize(void * /*deviceContext = NULL*/) {
\n+
1228 // nothing.
\n+
1229 }
\n+
1230};
\n+
1231
\n+
1232
\n+
1233} // end namespace Osd
\n+
1234
\n+
1235} // end namespace OPENSUBDIV_VERSION
\n+
1236using namespace OPENSUBDIV_VERSION;
\n+
1237
\n+
1238} // end namespace OpenSubdiv
\n+
1239
\n+
1240
\n+
1241#endif // OPENSUBDIV3_OSD_CPU_EVALUATOR_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::Reset
void Reset()
Resets the descriptor to default.
Definition: bufferDescriptor.h:81
\n-\n-\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::operator!=
bool operator!=(BufferDescriptor const &other) const
True if the descriptors are not identical.
Definition: bufferDescriptor.h:93
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::operator==
bool operator==(BufferDescriptor const &other) const
True if the descriptors are identical.
Definition: bufferDescriptor.h:86
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::IsValid
bool IsValid() const
True if the descriptor values are internally consistent.
Definition: bufferDescriptor.h:75
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::GetLocalOffset
int GetLocalOffset() const
Returns the relative offset within a stride.
Definition: bufferDescriptor.h:70
\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:1034
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:1100
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, int start, int end)
Static eval stencils function which takes raw CPU pointers for input and output.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:1188
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:971
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw CPU pointers for input and output.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, const float *duuWeights, const float *duvWeights, const float *dvvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw CPU pointers for input and output.
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer, PatchParam const *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:454
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: cpuEvaluator.h:517
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer, PatchParam const *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, const CpuEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: cpuEvaluator.h:303
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrays, const int *patchIndexBuffer, const PatchParam *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:886
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:823
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::Synchronize
static void Synchronize(void *)
synchronize all asynchronous computation invoked on this device.
Definition: cpuEvaluator.h:1227
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, const CpuEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function. This function has a same signature as other device kernels hav...
Definition: cpuEvaluator.h:72
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: cpuEvaluator.h:607
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const CpuEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: cpuEvaluator.h:168
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord
Coordinates set on a patch table.
Definition: types.h:42
\n+\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-bufferDescriptor.h\n+cpuEvaluator.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,125 +30,657 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H\n- 26#define OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H\n+ 25#ifndef OPENSUBDIV3_OSD_CPU_EVALUATOR_H\n+ 26#define OPENSUBDIV3_OSD_CPU_EVALUATOR_H\n 27\n 28#include \"../version.h\"\n- 29#include \n- 30\n- 31namespace OpenSubdiv {\n- 32namespace OPENSUBDIV_VERSION {\n+ 29#include \"../osd/bufferDescriptor.h\"\n+ 30#include \"../osd/types.h\"\n+ 31\n+ 32#include \n 33\n-34namespace Osd {\n- 35\n+ 34namespace OpenSubdiv {\n+ 35namespace OPENSUBDIV_VERSION {\n+ 36\n+ 37namespace Osd {\n+ 38\n+39class CpuEvaluator {\n+ 40public:\n 46\n- 47// example:\n- 48// n\n- 49// -----+----------------------------------------+-------------------------\n- 50// | vertex 0 |\n- 51// -----+----------------------------------------+-------------------------\n- 52// | X Y Z R G B A Xu Yu Zu Xv Yv Zv |\n- 53// -----+----------------------------------------+-------------------------\n- 54// <------------- stride = 13 -------------->\n- 55//\n- 56// - XYZ (offset = n+0, length = 3, stride = 13)\n- 57// - RGBA (offset = n+3, length = 4, stride = 13)\n- 58// - uTangent (offset = n+7, length = 3, stride = 13)\n- 59// - vTangent (offset = n+10, length = 3, stride = 13)\n- 60//\n-61struct BufferDescriptor {\n- 62\n-64 BufferDescriptor() : offset(0), length(0), stride(0) { }\n- 65\n-67 BufferDescriptor(int o, int l, int s) : offset(o), length(l), stride(s) { }\n- 68\n-70 int GetLocalOffset() const {\n- 71 return stride > 0 ? offset % stride : 0;\n- 72 }\n- 73\n-75 bool IsValid() const {\n- 76 return ((length > 0) &&\n- 77 (length <= stride - GetLocalOffset()));\n- 78 }\n- 79\n-81 void Reset() {\n- 82 offset = length = stride = 0;\n- 83 }\n+ 71 template \n+72 static bool EvalStencils(\n+ 73 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 74 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 75 STENCIL_TABLE const *stencilTable,\n+ 76 const CpuEvaluator *instance = NULL,\n+ 77 void * deviceContext = NULL) {\n+ 78\n+ 79 (void)instance; // unused\n+ 80 (void)deviceContext; // unused\n+ 81\n+ 82 if (stencilTable->GetNumStencils() == 0)\n+ 83 return false;\n 84\n-86 bool operator_==(BufferDescriptor const &other) const {\n- 87 return (offset == other.offset &&\n- 88 length == other.length &&\n- 89 stride == other.stride);\n- 90 }\n- 91\n-93 bool operator_!=(BufferDescriptor const &other) const {\n- 94 return !(this->operator==(other));\n- 95 }\n- 96\n-98 int offset;\n-100 int length;\n-102 int stride;\n- 103};\n- 104\n- 105} // end namespace Osd\n- 106\n- 107} // end namespace OPENSUBDIV_VERSION\n- 108using namespace OPENSUBDIV_VERSION;\n- 109\n- 110} // end namespace OpenSubdiv\n- 111\n- 112#endif // OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H\n+ 85 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 86 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 87 &stencilTable->GetSizes()[0],\n+ 88 &stencilTable->GetOffsets()[0],\n+ 89 &stencilTable->GetControlIndices()[0],\n+ 90 &stencilTable->GetWeights()[0],\n+ 91 /*start = */ 0,\n+ 92 /*end = */ stencilTable->GetNumStencils());\n+ 93 }\n+ 94\n+121 static bool EvalStencils(\n+ 122 const float *src, BufferDescriptor const &srcDesc,\n+ 123 float *dst, BufferDescriptor const &dstDesc,\n+ 124 const int * sizes,\n+ 125 const int * offsets,\n+ 126 const int * indices,\n+ 127 const float * weights,\n+ 128 int start, int end);\n+ 129\n+ 167 template \n+168 static bool EvalStencils(\n+ 169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 171 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 172 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 173 STENCIL_TABLE const *stencilTable,\n+ 174 const CpuEvaluator *instance = NULL,\n+ 175 void * deviceContext = NULL) {\n+ 176\n+ 177 (void)instance; // unused\n+ 178 (void)deviceContext; // unused\n+ 179\n+ 180 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 181 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 182 duBuffer->BindCpuBuffer(), duDesc,\n+ 183 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 184 &stencilTable->GetSizes()[0],\n+ 185 &stencilTable->GetOffsets()[0],\n+ 186 &stencilTable->GetControlIndices()[0],\n+ 187 &stencilTable->GetWeights()[0],\n+ 188 &stencilTable->GetDuWeights()[0],\n+ 189 &stencilTable->GetDvWeights()[0],\n+ 190 /*start = */ 0,\n+ 191 /*end = */ stencilTable->GetNumStencils());\n+ 192 }\n+ 193\n+234 static bool EvalStencils(\n+ 235 const float *src, BufferDescriptor const &srcDesc,\n+ 236 float *dst, BufferDescriptor const &dstDesc,\n+ 237 float *du, BufferDescriptor const &duDesc,\n+ 238 float *dv, BufferDescriptor const &dvDesc,\n+ 239 const int * sizes,\n+ 240 const int * offsets,\n+ 241 const int * indices,\n+ 242 const float * weights,\n+ 243 const float * duWeights,\n+ 244 const float * dvWeights,\n+ 245 int start, int end);\n+ 246\n+ 302 template \n+303 static bool EvalStencils(\n+ 304 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 305 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 306 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 307 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 308 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 309 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 310 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 311 STENCIL_TABLE const *stencilTable,\n+ 312 const CpuEvaluator *instance = NULL,\n+ 313 void * deviceContext = NULL) {\n+ 314\n+ 315 (void)instance; // unused\n+ 316 (void)deviceContext; // unused\n+ 317\n+ 318 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 319 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 320 duBuffer->BindCpuBuffer(), duDesc,\n+ 321 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 322 duuBuffer->BindCpuBuffer(), duuDesc,\n+ 323 duvBuffer->BindCpuBuffer(), duvDesc,\n+ 324 dvvBuffer->BindCpuBuffer(), dvvDesc,\n+ 325 &stencilTable->GetSizes()[0],\n+ 326 &stencilTable->GetOffsets()[0],\n+ 327 &stencilTable->GetControlIndices()[0],\n+ 328 &stencilTable->GetWeights()[0],\n+ 329 &stencilTable->GetDuWeights()[0],\n+ 330 &stencilTable->GetDvWeights()[0],\n+ 331 &stencilTable->GetDuuWeights()[0],\n+ 332 &stencilTable->GetDuvWeights()[0],\n+ 333 &stencilTable->GetDvvWeights()[0],\n+ 334 /*start = */ 0,\n+ 335 /*end = */ stencilTable->GetNumStencils());\n+ 336 }\n+ 337\n+399 static bool EvalStencils(\n+ 400 const float *src, BufferDescriptor const &srcDesc,\n+ 401 float *dst, BufferDescriptor const &dstDesc,\n+ 402 float *du, BufferDescriptor const &duDesc,\n+ 403 float *dv, BufferDescriptor const &dvDesc,\n+ 404 float *duu, BufferDescriptor const &duuDesc,\n+ 405 float *duv, BufferDescriptor const &duvDesc,\n+ 406 float *dvv, BufferDescriptor const &dvvDesc,\n+ 407 const int * sizes,\n+ 408 const int * offsets,\n+ 409 const int * indices,\n+ 410 const float * weights,\n+ 411 const float * duWeights,\n+ 412 const float * dvWeights,\n+ 413 const float * duuWeights,\n+ 414 const float * duvWeights,\n+ 415 const float * dvvWeights,\n+ 416 int start, int end);\n+ 417\n+ 423\n+ 452 template \n+454 static bool EvalPatches(\n+ 455 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 456 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 457 int numPatchCoords,\n+ 458 PATCHCOORD_BUFFER *patchCoords,\n+ 459 PATCH_TABLE *patchTable,\n+ 460 CpuEvaluator const *instance = NULL,\n+ 461 void * deviceContext = NULL) {\n+ 462\n+ 463 (void)instance; // unused\n+ 464 (void)deviceContext; // unused\n+ 465\n+ 466 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 467 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 468 numPatchCoords,\n+ 469 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 470 patchTable->GetPatchArrayBuffer(),\n+ 471 patchTable->GetPatchIndexBuffer(),\n+ 472 patchTable->GetPatchParamBuffer());\n+ 473 }\n+ 474\n+ 515 template \n+517 static bool EvalPatches(\n+ 518 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 519 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 520 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 521 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 522 int numPatchCoords,\n+ 523 PATCHCOORD_BUFFER *patchCoords,\n+ 524 PATCH_TABLE *patchTable,\n+ 525 CpuEvaluator const *instance = NULL,\n+ 526 void * deviceContext = NULL) {\n+ 527\n+ 528 (void)instance; // unused\n+ 529 (void)deviceContext; // unused\n+ 530\n+ 531 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.\n+ 532 // ideally all buffer classes should have templated by datatype\n+ 533 // so that downcast isn't needed there.\n+ 534 // (e.g. Osd::CpuBuffer )\n+ 535 //\n+ 536 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 537 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 538 duBuffer->BindCpuBuffer(), duDesc,\n+ 539 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 540 numPatchCoords,\n+ 541 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 542 patchTable->GetPatchArrayBuffer(),\n+ 543 patchTable->GetPatchIndexBuffer(),\n+ 544 patchTable->GetPatchParamBuffer());\n+ 545 }\n+ 546\n+ 605 template \n+607 static bool EvalPatches(\n+ 608 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 609 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 610 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 611 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 612 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 613 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 614 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 615 int numPatchCoords,\n+ 616 PATCHCOORD_BUFFER *patchCoords,\n+ 617 PATCH_TABLE *patchTable,\n+ 618 CpuEvaluator const *instance = NULL,\n+ 619 void * deviceContext = NULL) {\n+ 620\n+ 621 (void)instance; // unused\n+ 622 (void)deviceContext; // unused\n+ 623\n+ 624 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.\n+ 625 // ideally all buffer classes should have templated by datatype\n+ 626 // so that downcast isn't needed there.\n+ 627 // (e.g. Osd::CpuBuffer )\n+ 628 //\n+ 629 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 630 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 631 duBuffer->BindCpuBuffer(), duDesc,\n+ 632 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 633 duuBuffer->BindCpuBuffer(), duuDesc,\n+ 634 duvBuffer->BindCpuBuffer(), duvDesc,\n+ 635 dvvBuffer->BindCpuBuffer(), dvvDesc,\n+ 636 numPatchCoords,\n+ 637 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 638 patchTable->GetPatchArrayBuffer(),\n+ 639 patchTable->GetPatchIndexBuffer(),\n+ 640 patchTable->GetPatchParamBuffer());\n+ 641 }\n+ 642\n+670 static bool EvalPatches(\n+ 671 const float *src, BufferDescriptor const &srcDesc,\n+ 672 float *dst, BufferDescriptor const &dstDesc,\n+ 673 int numPatchCoords,\n+ 674 const PatchCoord *patchCoords,\n+ 675 const PatchArray *patchArrays,\n+ 676 const int *patchIndexBuffer,\n+ 677 const PatchParam *patchParamBuffer);\n+ 678\n+716 static bool EvalPatches(\n+ 717 const float *src, BufferDescriptor const &srcDesc,\n+ 718 float *dst, BufferDescriptor const &dstDesc,\n+ 719 float *du, BufferDescriptor const &duDesc,\n+ 720 float *dv, BufferDescriptor const &dvDesc,\n+ 721 int numPatchCoords,\n+ 722 PatchCoord const *patchCoords,\n+ 723 PatchArray const *patchArrays,\n+ 724 const int *patchIndexBuffer,\n+ 725 PatchParam const *patchParamBuffer);\n+ 726\n+779 static bool EvalPatches(\n+ 780 const float *src, BufferDescriptor const &srcDesc,\n+ 781 float *dst, BufferDescriptor const &dstDesc,\n+ 782 float *du, BufferDescriptor const &duDesc,\n+ 783 float *dv, BufferDescriptor const &dvDesc,\n+ 784 float *duu, BufferDescriptor const &duuDesc,\n+ 785 float *duv, BufferDescriptor const &duvDesc,\n+ 786 float *dvv, BufferDescriptor const &dvvDesc,\n+ 787 int numPatchCoords,\n+ 788 PatchCoord const *patchCoords,\n+ 789 PatchArray const *patchArrays,\n+ 790 const int *patchIndexBuffer,\n+ 791 PatchParam const *patchParamBuffer);\n+ 792\n+ 821 template \n+823 static bool EvalPatchesVarying(\n+ 824 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 825 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 826 int numPatchCoords,\n+ 827 PATCHCOORD_BUFFER *patchCoords,\n+ 828 PATCH_TABLE *patchTable,\n+ 829 CpuEvaluator const *instance = NULL,\n+ 830 void * deviceContext = NULL) {\n+ 831\n+ 832 (void)instance; // unused\n+ 833 (void)deviceContext; // unused\n+ 834\n+ 835 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 836 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 837 numPatchCoords,\n+ 838 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 839 patchTable->GetVaryingPatchArrayBuffer(),\n+ 840 patchTable->GetVaryingPatchIndexBuffer(),\n+ 841 patchTable->GetPatchParamBuffer());\n+ 842 }\n+ 843\n+ 884 template \n+886 static bool EvalPatchesVarying(\n+ 887 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 888 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 889 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 890 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 891 int numPatchCoords,\n+ 892 PATCHCOORD_BUFFER *patchCoords,\n+ 893 PATCH_TABLE *patchTable,\n+ 894 CpuEvaluator const *instance = NULL,\n+ 895 void * deviceContext = NULL) {\n+ 896\n+ 897 (void)instance; // unused\n+ 898 (void)deviceContext; // unused\n+ 899\n+ 900 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 901 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 902 duBuffer->BindCpuBuffer(), duDesc,\n+ 903 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 904 numPatchCoords,\n+ 905 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 906 patchTable->GetVaryingPatchArrayBuffer(),\n+ 907 patchTable->GetVaryingPatchIndexBuffer(),\n+ 908 patchTable->GetPatchParamBuffer());\n+ 909 }\n+ 910\n+ 969 template \n+971 static bool EvalPatchesVarying(\n+ 972 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 973 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 974 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 975 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 976 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 977 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 978 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 979 int numPatchCoords,\n+ 980 PATCHCOORD_BUFFER *patchCoords,\n+ 981 PATCH_TABLE *patchTable,\n+ 982 CpuEvaluator const *instance = NULL,\n+ 983 void * deviceContext = NULL) {\n+ 984\n+ 985 (void)instance; // unused\n+ 986 (void)deviceContext; // unused\n+ 987\n+ 988 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 989 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 990 duBuffer->BindCpuBuffer(), duDesc,\n+ 991 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 992 duuBuffer->BindCpuBuffer(), duuDesc,\n+ 993 duvBuffer->BindCpuBuffer(), duvDesc,\n+ 994 dvvBuffer->BindCpuBuffer(), dvvDesc,\n+ 995 numPatchCoords,\n+ 996 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 997 patchTable->GetVaryingPatchArrayBuffer(),\n+ 998 patchTable->GetVaryingPatchIndexBuffer(),\n+ 999 patchTable->GetPatchParamBuffer());\n+ 1000 }\n+ 1001\n+ 1032 template \n+1034 static bool EvalPatchesFaceVarying(\n+ 1035 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1036 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1037 int numPatchCoords,\n+ 1038 PATCHCOORD_BUFFER *patchCoords,\n+ 1039 PATCH_TABLE *patchTable,\n+ 1040 int fvarChannel,\n+ 1041 CpuEvaluator const *instance = NULL,\n+ 1042 void * deviceContext = NULL) {\n+ 1043\n+ 1044 (void)instance; // unused\n+ 1045 (void)deviceContext; // unused\n+ 1046\n+ 1047 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 1048 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 1049 numPatchCoords,\n+ 1050 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 1051 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 1052 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1053 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 1054 }\n+ 1055\n+ 1098 template \n+1100 static bool EvalPatchesFaceVarying(\n+ 1101 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1102 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1103 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1104 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1105 int numPatchCoords,\n+ 1106 PATCHCOORD_BUFFER *patchCoords,\n+ 1107 PATCH_TABLE *patchTable,\n+ 1108 int fvarChannel,\n+ 1109 CpuEvaluator const *instance = NULL,\n+ 1110 void * deviceContext = NULL) {\n+ 1111\n+ 1112 (void)instance; // unused\n+ 1113 (void)deviceContext; // unused\n+ 1114\n+ 1115 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 1116 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 1117 duBuffer->BindCpuBuffer(), duDesc,\n+ 1118 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 1119 numPatchCoords,\n+ 1120 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 1121 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 1122 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1123 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 1124 }\n+ 1125\n+ 1186 template \n+1188 static bool EvalPatchesFaceVarying(\n+ 1189 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+ 1190 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+ 1191 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n+ 1192 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n+ 1193 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n+ 1194 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n+ 1195 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n+ 1196 int numPatchCoords,\n+ 1197 PATCHCOORD_BUFFER *patchCoords,\n+ 1198 PATCH_TABLE *patchTable,\n+ 1199 int fvarChannel,\n+ 1200 CpuEvaluator const *instance = NULL,\n+ 1201 void * deviceContext = NULL) {\n+ 1202\n+ 1203 (void)instance; // unused\n+ 1204 (void)deviceContext; // unused\n+ 1205\n+ 1206 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n+ 1207 dstBuffer->BindCpuBuffer(), dstDesc,\n+ 1208 duBuffer->BindCpuBuffer(), duDesc,\n+ 1209 dvBuffer->BindCpuBuffer(), dvDesc,\n+ 1210 duuBuffer->BindCpuBuffer(), duuDesc,\n+ 1211 duvBuffer->BindCpuBuffer(), duvDesc,\n+ 1212 dvvBuffer->BindCpuBuffer(), dvvDesc,\n+ 1213 numPatchCoords,\n+ 1214 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n+ 1215 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n+ 1216 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n+ 1217 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n+ 1218 }\n+ 1219\n+ 1225\n+1227 static void Synchronize(void * /*deviceContext = NULL*/) {\n+ 1228 // nothing.\n+ 1229 }\n+ 1230};\n+ 1231\n+ 1232\n+ 1233} // end namespace Osd\n+ 1234\n+ 1235} // end namespace OPENSUBDIV_VERSION\n+ 1236using namespace OPENSUBDIV_VERSION;\n+ 1237\n+ 1238} // end namespace OpenSubdiv\n+ 1239\n+ 1240\n+ 1241#endif // OPENSUBDIV3_OSD_CPU_EVALUATOR_H\n OpenSubdiv\n Definition: limits.h:32\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n BufferDescriptor is a struct which describes buffer elements in interleaved\n data buffers....\n Definition: bufferDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::stride\n-int stride\n-stride to the next element\n-Definition: bufferDescriptor.h:102\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::Reset\n-void Reset()\n-Resets the descriptor to default.\n-Definition: bufferDescriptor.h:81\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::BufferDescriptor\n-BufferDescriptor()\n-Default Constructor.\n-Definition: bufferDescriptor.h:64\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::BufferDescriptor\n-BufferDescriptor(int o, int l, int s)\n-Constructor.\n-Definition: bufferDescriptor.h:67\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::length\n-int length\n-number or length of the data\n-Definition: bufferDescriptor.h:100\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::operator!=\n-bool operator!=(BufferDescriptor const &other) const\n-True if the descriptors are not identical.\n-Definition: bufferDescriptor.h:93\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::operator==\n-bool operator==(BufferDescriptor const &other) const\n-True if the descriptors are identical.\n-Definition: bufferDescriptor.h:86\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::IsValid\n-bool IsValid() const\n-True if the descriptor values are internally consistent.\n-Definition: bufferDescriptor.h:75\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::GetLocalOffset\n-int GetLocalOffset() const\n-Returns the relative offset within a stride.\n-Definition: bufferDescriptor.h:70\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::offset\n-int offset\n-offset to desired element data\n-Definition: bufferDescriptor.h:98\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator\n+Definition: cpuEvaluator.h:39\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n+fvarChannel, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cpuEvaluator.h:1034\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CpuEvaluator const\n+*instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cpuEvaluator.h:1100\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n+static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int\n+*offsets, const int *indices, const float *weights, int start, int end)\n+Static eval stencils function which takes raw CPU pointers for input and\n+output.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying\n+static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n+const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n+DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n+CpuEvaluator const *instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cpuEvaluator.h:1188\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const\n+*instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cpuEvaluator.h:971\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n+static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int\n+*offsets, const int *indices, const float *weights, const float *duWeights,\n+const float *dvWeights, int start, int end)\n+Static eval stencils function with derivatives, which takes raw CPU pointers\n+for input and output.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n+static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n+BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n+float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int\n+*offsets, const int *indices, const float *weights, const float *duWeights,\n+const float *dvWeights, const float *duuWeights, const float *duvWeights, const\n+float *dvvWeights, int start, int end)\n+Static eval stencils function with derivatives, which takes raw CPU pointers\n+for input and output.\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n+static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n+BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n+float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord\n+const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer,\n+PatchParam const *patchParamBuffer)\n+Static limit eval function. It takes an array of PatchCoord and evaluate limit\n+values on given PatchT...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n+PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const\n+*instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cpuEvaluator.h:454\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n+*patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: cpuEvaluator.h:517\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n+static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n+&duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords,\n+PatchCoord const *patchCoords, PatchArray const *patchArrays, const int\n+*patchIndexBuffer, PatchParam const *patchParamBuffer)\n+Static limit eval function. It takes an array of PatchCoord and evaluate limit\n+values on given PatchT...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n+&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n+*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n+const CpuEvaluator *instance=NULL, void *deviceContext=NULL)\n+Generic static eval stencils function with derivatives. This function has a\n+same signature as other d...\n+Definition: cpuEvaluator.h:303\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n+static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n+float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const\n+PatchCoord *patchCoords, const PatchArray *patchArrays, const int\n+*patchIndexBuffer, const PatchParam *patchParamBuffer)\n+Static limit eval function. It takes an array of PatchCoord and evaluate limit\n+values on given PatchT...\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void\n+*deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cpuEvaluator.h:886\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying\n+static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n+numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n+CpuEvaluator const *instance=NULL, void *deviceContext=NULL)\n+Generic limit eval function. This function has a same signature as other device\n+kernels have so that ...\n+Definition: cpuEvaluator.h:823\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::Synchronize\n+static void Synchronize(void *)\n+synchronize all asynchronous computation invoked on this device.\n+Definition: cpuEvaluator.h:1227\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n+const *stencilTable, const CpuEvaluator *instance=NULL, void\n+*deviceContext=NULL)\n+Generic static eval stencils function. This function has a same signature as\n+other device kernels hav...\n+Definition: cpuEvaluator.h:72\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n+static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n+DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n+BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n+&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n+*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n+BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n+*patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void\n+*deviceContext=NULL)\n+Generic limit eval function with derivatives. This function has a same\n+signature as other device kern...\n+Definition: cpuEvaluator.h:607\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n+static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n+&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n+*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n+BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const\n+CpuEvaluator *instance=NULL, void *deviceContext=NULL)\n+Generic static eval stencils function with derivatives. This function has a\n+same signature as other d...\n+Definition: cpuEvaluator.h:168\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord\n+Coordinates set on a patch table.\n+Definition: types.h:42\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArray\n+Definition: types.h:66\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParam\n+Definition: types.h:127\n * opensubdiv\n * osd\n- * bufferDescriptor.h\n+ * cpuEvaluator.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00899.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00899.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cpuEvaluator.h File Reference\n+OpenSubdiv: opensubdiv/osd/bufferDescriptor.h File Reference\n \n \n \n \n \n \n \n@@ -83,27 +83,26 @@\n \n \n \n
\n
\n Classes |\n Namespaces
\n-
cpuEvaluator.h File Reference
\n+
bufferDescriptor.h File Reference
\n
\n
\n
#include "../version.h"
\n-#include "../osd/bufferDescriptor.h"
\n-#include "../osd/types.h"
\n-#include <cstddef>
\n+#include <string.h>
\n
\n

Go to the source code of this file.

\n \n \n-\n+\n+\n \n

\n Classes

class  CpuEvaluator
struct  BufferDescriptor
 BufferDescriptor is a struct which describes buffer elements in interleaved data buffers. Almost all Osd Evaluator APIs take BufferDescriptors along with device-specific buffer objects. More...
 
\n \n \n \n \n@@ -112,13 +111,13 @@\n \n

\n Namespaces

namespace  OpenSubdiv
 
namespace  OpenSubdiv::OPENSUBDIV_VERSION
 
\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -5,27 +5,28 @@\n \n \n \n [Logo] OpenSubdiv\n \n \n Classes | Namespaces\n-cpuEvaluator.h File Reference\n+bufferDescriptor.h File Reference\n #include \"../version.h\"\n-#include \"../osd/bufferDescriptor.h\"\n-#include \"../osd/types.h\"\n-#include \n+#include \n Go_to_the_source_code_of_this_file.\n Classes\n-class \u00a0CpuEvaluator\n+struct \u00a0BufferDescriptor\n+ BufferDescriptor is a struct which describes buffer elements in\n+\u00a0 interleaved data buffers. Almost all Osd Evaluator APIs take\n+ BufferDescriptors along with device-specific buffer objects. More...\n \u00a0\n Namespaces\n namespace \u00a0OpenSubdiv\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION\n \u00a0\n namespace \u00a0OpenSubdiv::OPENSUBDIV_VERSION::Osd\n \u00a0\n * opensubdiv\n * osd\n- * cpuEvaluator.h\n+ * bufferDescriptor.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00899.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00899.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var a00899 = [\n- [\"CpuEvaluator\", \"a01157.html\", null]\n+ [\"BufferDescriptor\", \"a01133.html\", \"a01133\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00899_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00899_source.html", "unified_diff": "@@ -1,15 +1,15 @@\n \n \n \n \n \n \n \n-OpenSubdiv: opensubdiv/osd/cpuEvaluator.h Source File\n+OpenSubdiv: opensubdiv/osd/bufferDescriptor.h Source File\n \n \n \n \n \n \n \n@@ -80,15 +80,15 @@\n
\n \n
\n \n
\n-
cpuEvaluator.h
\n+
bufferDescriptor.h
\n
\n
\n Go to the documentation of this file.
1//
\n
2// Copyright 2015 Pixar
\n
3//
\n
4// Licensed under the Apache License, Version 2.0 (the "Apache License")
\n
5// with the following modification; you may not use this file except in
\n@@ -107,501 +107,98 @@\n
18// Unless required by applicable law or agreed to in writing, software
\n
19// distributed under the Apache License with the above modification is
\n
20// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
\n
21// KIND, either express or implied. See the Apache License for the specific
\n
22// language governing permissions and limitations under the Apache License.
\n
23//
\n
24
\n-
25#ifndef OPENSUBDIV3_OSD_CPU_EVALUATOR_H
\n-
26#define OPENSUBDIV3_OSD_CPU_EVALUATOR_H
\n+
25#ifndef OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H
\n+
26#define OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H
\n
27
\n
28#include "../version.h"
\n-
29#include "../osd/bufferDescriptor.h"
\n-
30#include "../osd/types.h"
\n-
31
\n-
32#include <cstddef>
\n+
29#include <string.h>
\n+
30
\n+
31namespace OpenSubdiv {
\n+
32namespace OPENSUBDIV_VERSION {
\n
33
\n-
34namespace OpenSubdiv {
\n-
35namespace OPENSUBDIV_VERSION {
\n-
36
\n-
37namespace Osd {
\n-
38
\n-
39class CpuEvaluator {
\n-
40public:
\n+
34namespace Osd {
\n+
35
\n
46
\n-
71 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
72 static bool EvalStencils(
\n-
73 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
74 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
75 STENCIL_TABLE const *stencilTable,
\n-
76 const CpuEvaluator *instance = NULL,
\n-
77 void * deviceContext = NULL) {
\n-
78
\n-
79 (void)instance; // unused
\n-
80 (void)deviceContext; // unused
\n-
81
\n-
82 if (stencilTable->GetNumStencils() == 0)
\n-
83 return false;
\n+
47// example:
\n+
48// n
\n+
49// -----+----------------------------------------+-------------------------
\n+
50// | vertex 0 |
\n+
51// -----+----------------------------------------+-------------------------
\n+
52// | X Y Z R G B A Xu Yu Zu Xv Yv Zv |
\n+
53// -----+----------------------------------------+-------------------------
\n+
54// <------------- stride = 13 -------------->
\n+
55//
\n+
56// - XYZ (offset = n+0, length = 3, stride = 13)
\n+
57// - RGBA (offset = n+3, length = 4, stride = 13)
\n+
58// - uTangent (offset = n+7, length = 3, stride = 13)
\n+
59// - vTangent (offset = n+10, length = 3, stride = 13)
\n+
60//
\n+
61struct BufferDescriptor {
\n+
62
\n+
64 BufferDescriptor() : offset(0), length(0), stride(0) { }
\n+
65
\n+
67 BufferDescriptor(int o, int l, int s) : offset(o), length(l), stride(s) { }
\n+
68
\n+
70 int GetLocalOffset() const {
\n+
71 return stride > 0 ? offset % stride : 0;
\n+
72 }
\n+
73
\n+
75 bool IsValid() const {
\n+
76 return ((length > 0) &&
\n+
77 (length <= stride - GetLocalOffset()));
\n+
78 }
\n+
79
\n+
81 void Reset() {
\n+
82 offset = length = stride = 0;
\n+
83 }
\n
84
\n-
85 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
86 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
87 &stencilTable->GetSizes()[0],
\n-
88 &stencilTable->GetOffsets()[0],
\n-
89 &stencilTable->GetControlIndices()[0],
\n-
90 &stencilTable->GetWeights()[0],
\n-
91 /*start = */ 0,
\n-
92 /*end = */ stencilTable->GetNumStencils());
\n-
93 }
\n-
94
\n-
121 static bool EvalStencils(
\n-
122 const float *src, BufferDescriptor const &srcDesc,
\n-
123 float *dst, BufferDescriptor const &dstDesc,
\n-
124 const int * sizes,
\n-
125 const int * offsets,
\n-
126 const int * indices,
\n-
127 const float * weights,
\n-
128 int start, int end);
\n-
129
\n-
167 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
168 static bool EvalStencils(
\n-
169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
171 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
172 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
173 STENCIL_TABLE const *stencilTable,
\n-
174 const CpuEvaluator *instance = NULL,
\n-
175 void * deviceContext = NULL) {
\n-
176
\n-
177 (void)instance; // unused
\n-
178 (void)deviceContext; // unused
\n-
179
\n-
180 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
181 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
182 duBuffer->BindCpuBuffer(), duDesc,
\n-
183 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
184 &stencilTable->GetSizes()[0],
\n-
185 &stencilTable->GetOffsets()[0],
\n-
186 &stencilTable->GetControlIndices()[0],
\n-
187 &stencilTable->GetWeights()[0],
\n-
188 &stencilTable->GetDuWeights()[0],
\n-
189 &stencilTable->GetDvWeights()[0],
\n-
190 /*start = */ 0,
\n-
191 /*end = */ stencilTable->GetNumStencils());
\n-
192 }
\n-
193
\n-
234 static bool EvalStencils(
\n-
235 const float *src, BufferDescriptor const &srcDesc,
\n-
236 float *dst, BufferDescriptor const &dstDesc,
\n-
237 float *du, BufferDescriptor const &duDesc,
\n-
238 float *dv, BufferDescriptor const &dvDesc,
\n-
239 const int * sizes,
\n-
240 const int * offsets,
\n-
241 const int * indices,
\n-
242 const float * weights,
\n-
243 const float * duWeights,
\n-
244 const float * dvWeights,
\n-
245 int start, int end);
\n-
246
\n-
302 template <typename SRC_BUFFER, typename DST_BUFFER, typename STENCIL_TABLE>
\n-
303 static bool EvalStencils(
\n-
304 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
305 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
306 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
307 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
308 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
309 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
310 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
311 STENCIL_TABLE const *stencilTable,
\n-
312 const CpuEvaluator *instance = NULL,
\n-
313 void * deviceContext = NULL) {
\n-
314
\n-
315 (void)instance; // unused
\n-
316 (void)deviceContext; // unused
\n-
317
\n-
318 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
319 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
320 duBuffer->BindCpuBuffer(), duDesc,
\n-
321 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
322 duuBuffer->BindCpuBuffer(), duuDesc,
\n-
323 duvBuffer->BindCpuBuffer(), duvDesc,
\n-
324 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n-
325 &stencilTable->GetSizes()[0],
\n-
326 &stencilTable->GetOffsets()[0],
\n-
327 &stencilTable->GetControlIndices()[0],
\n-
328 &stencilTable->GetWeights()[0],
\n-
329 &stencilTable->GetDuWeights()[0],
\n-
330 &stencilTable->GetDvWeights()[0],
\n-
331 &stencilTable->GetDuuWeights()[0],
\n-
332 &stencilTable->GetDuvWeights()[0],
\n-
333 &stencilTable->GetDvvWeights()[0],
\n-
334 /*start = */ 0,
\n-
335 /*end = */ stencilTable->GetNumStencils());
\n-
336 }
\n-
337
\n-
399 static bool EvalStencils(
\n-
400 const float *src, BufferDescriptor const &srcDesc,
\n-
401 float *dst, BufferDescriptor const &dstDesc,
\n-
402 float *du, BufferDescriptor const &duDesc,
\n-
403 float *dv, BufferDescriptor const &dvDesc,
\n-
404 float *duu, BufferDescriptor const &duuDesc,
\n-
405 float *duv, BufferDescriptor const &duvDesc,
\n-
406 float *dvv, BufferDescriptor const &dvvDesc,
\n-
407 const int * sizes,
\n-
408 const int * offsets,
\n-
409 const int * indices,
\n-
410 const float * weights,
\n-
411 const float * duWeights,
\n-
412 const float * dvWeights,
\n-
413 const float * duuWeights,
\n-
414 const float * duvWeights,
\n-
415 const float * dvvWeights,
\n-
416 int start, int end);
\n-
417
\n-
423
\n-
452 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
453 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
454 static bool EvalPatches(
\n-
455 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
456 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
457 int numPatchCoords,
\n-
458 PATCHCOORD_BUFFER *patchCoords,
\n-
459 PATCH_TABLE *patchTable,
\n-
460 CpuEvaluator const *instance = NULL,
\n-
461 void * deviceContext = NULL) {
\n-
462
\n-
463 (void)instance; // unused
\n-
464 (void)deviceContext; // unused
\n-
465
\n-
466 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
467 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
468 numPatchCoords,
\n-
469 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
470 patchTable->GetPatchArrayBuffer(),
\n-
471 patchTable->GetPatchIndexBuffer(),
\n-
472 patchTable->GetPatchParamBuffer());
\n-
473 }
\n-
474
\n-
515 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
516 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
517 static bool EvalPatches(
\n-
518 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
519 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
520 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
521 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
522 int numPatchCoords,
\n-
523 PATCHCOORD_BUFFER *patchCoords,
\n-
524 PATCH_TABLE *patchTable,
\n-
525 CpuEvaluator const *instance = NULL,
\n-
526 void * deviceContext = NULL) {
\n-
527
\n-
528 (void)instance; // unused
\n-
529 (void)deviceContext; // unused
\n-
530
\n-
531 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.
\n-
532 // ideally all buffer classes should have templated by datatype
\n-
533 // so that downcast isn't needed there.
\n-
534 // (e.g. Osd::CpuBuffer<PatchCoord> )
\n-
535 //
\n-
536 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
537 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
538 duBuffer->BindCpuBuffer(), duDesc,
\n-
539 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
540 numPatchCoords,
\n-
541 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
542 patchTable->GetPatchArrayBuffer(),
\n-
543 patchTable->GetPatchIndexBuffer(),
\n-
544 patchTable->GetPatchParamBuffer());
\n-
545 }
\n-
546
\n-
605 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
606 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
607 static bool EvalPatches(
\n-
608 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
609 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
610 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
611 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
612 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
613 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
614 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
615 int numPatchCoords,
\n-
616 PATCHCOORD_BUFFER *patchCoords,
\n-
617 PATCH_TABLE *patchTable,
\n-
618 CpuEvaluator const *instance = NULL,
\n-
619 void * deviceContext = NULL) {
\n-
620
\n-
621 (void)instance; // unused
\n-
622 (void)deviceContext; // unused
\n-
623
\n-
624 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.
\n-
625 // ideally all buffer classes should have templated by datatype
\n-
626 // so that downcast isn't needed there.
\n-
627 // (e.g. Osd::CpuBuffer<PatchCoord> )
\n-
628 //
\n-
629 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
630 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
631 duBuffer->BindCpuBuffer(), duDesc,
\n-
632 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
633 duuBuffer->BindCpuBuffer(), duuDesc,
\n-
634 duvBuffer->BindCpuBuffer(), duvDesc,
\n-
635 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n-
636 numPatchCoords,
\n-
637 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
638 patchTable->GetPatchArrayBuffer(),
\n-
639 patchTable->GetPatchIndexBuffer(),
\n-
640 patchTable->GetPatchParamBuffer());
\n-
641 }
\n-
642
\n-
670 static bool EvalPatches(
\n-
671 const float *src, BufferDescriptor const &srcDesc,
\n-
672 float *dst, BufferDescriptor const &dstDesc,
\n-
673 int numPatchCoords,
\n-
674 const PatchCoord *patchCoords,
\n-
675 const PatchArray *patchArrays,
\n-
676 const int *patchIndexBuffer,
\n-
677 const PatchParam *patchParamBuffer);
\n-
678
\n-
716 static bool EvalPatches(
\n-
717 const float *src, BufferDescriptor const &srcDesc,
\n-
718 float *dst, BufferDescriptor const &dstDesc,
\n-
719 float *du, BufferDescriptor const &duDesc,
\n-
720 float *dv, BufferDescriptor const &dvDesc,
\n-
721 int numPatchCoords,
\n-
722 PatchCoord const *patchCoords,
\n-
723 PatchArray const *patchArrays,
\n-
724 const int *patchIndexBuffer,
\n-
725 PatchParam const *patchParamBuffer);
\n-
726
\n-
779 static bool EvalPatches(
\n-
780 const float *src, BufferDescriptor const &srcDesc,
\n-
781 float *dst, BufferDescriptor const &dstDesc,
\n-
782 float *du, BufferDescriptor const &duDesc,
\n-
783 float *dv, BufferDescriptor const &dvDesc,
\n-
784 float *duu, BufferDescriptor const &duuDesc,
\n-
785 float *duv, BufferDescriptor const &duvDesc,
\n-
786 float *dvv, BufferDescriptor const &dvvDesc,
\n-
787 int numPatchCoords,
\n-
788 PatchCoord const *patchCoords,
\n-
789 PatchArray const *patchArrays,
\n-
790 const int *patchIndexBuffer,
\n-
791 PatchParam const *patchParamBuffer);
\n-
792
\n-
821 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
822 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
823 static bool EvalPatchesVarying(
\n-
824 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
825 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
826 int numPatchCoords,
\n-
827 PATCHCOORD_BUFFER *patchCoords,
\n-
828 PATCH_TABLE *patchTable,
\n-
829 CpuEvaluator const *instance = NULL,
\n-
830 void * deviceContext = NULL) {
\n-
831
\n-
832 (void)instance; // unused
\n-
833 (void)deviceContext; // unused
\n-
834
\n-
835 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
836 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
837 numPatchCoords,
\n-
838 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
839 patchTable->GetVaryingPatchArrayBuffer(),
\n-
840 patchTable->GetVaryingPatchIndexBuffer(),
\n-
841 patchTable->GetPatchParamBuffer());
\n-
842 }
\n-
843
\n-
884 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
885 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
886 static bool EvalPatchesVarying(
\n-
887 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
888 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
889 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
890 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
891 int numPatchCoords,
\n-
892 PATCHCOORD_BUFFER *patchCoords,
\n-
893 PATCH_TABLE *patchTable,
\n-
894 CpuEvaluator const *instance = NULL,
\n-
895 void * deviceContext = NULL) {
\n-
896
\n-
897 (void)instance; // unused
\n-
898 (void)deviceContext; // unused
\n-
899
\n-
900 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
901 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
902 duBuffer->BindCpuBuffer(), duDesc,
\n-
903 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
904 numPatchCoords,
\n-
905 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
906 patchTable->GetVaryingPatchArrayBuffer(),
\n-
907 patchTable->GetVaryingPatchIndexBuffer(),
\n-
908 patchTable->GetPatchParamBuffer());
\n-
909 }
\n-
910
\n-
969 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
970 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
971 static bool EvalPatchesVarying(
\n-
972 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
973 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
974 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
975 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
976 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
977 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
978 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
979 int numPatchCoords,
\n-
980 PATCHCOORD_BUFFER *patchCoords,
\n-
981 PATCH_TABLE *patchTable,
\n-
982 CpuEvaluator const *instance = NULL,
\n-
983 void * deviceContext = NULL) {
\n-
984
\n-
985 (void)instance; // unused
\n-
986 (void)deviceContext; // unused
\n-
987
\n-
988 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
989 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
990 duBuffer->BindCpuBuffer(), duDesc,
\n-
991 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
992 duuBuffer->BindCpuBuffer(), duuDesc,
\n-
993 duvBuffer->BindCpuBuffer(), duvDesc,
\n-
994 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n-
995 numPatchCoords,
\n-
996 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
997 patchTable->GetVaryingPatchArrayBuffer(),
\n-
998 patchTable->GetVaryingPatchIndexBuffer(),
\n-
999 patchTable->GetPatchParamBuffer());
\n-
1000 }
\n-
1001
\n-
1032 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1033 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1034 static bool EvalPatchesFaceVarying(
\n-
1035 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1036 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1037 int numPatchCoords,
\n-
1038 PATCHCOORD_BUFFER *patchCoords,
\n-
1039 PATCH_TABLE *patchTable,
\n-
1040 int fvarChannel,
\n-
1041 CpuEvaluator const *instance = NULL,
\n-
1042 void * deviceContext = NULL) {
\n-
1043
\n-
1044 (void)instance; // unused
\n-
1045 (void)deviceContext; // unused
\n-
1046
\n-
1047 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
1048 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
1049 numPatchCoords,
\n-
1050 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
1051 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
1052 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1053 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
1054 }
\n-
1055
\n-
1098 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1099 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1100 static bool EvalPatchesFaceVarying(
\n-
1101 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1102 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1103 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1104 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1105 int numPatchCoords,
\n-
1106 PATCHCOORD_BUFFER *patchCoords,
\n-
1107 PATCH_TABLE *patchTable,
\n-
1108 int fvarChannel,
\n-
1109 CpuEvaluator const *instance = NULL,
\n-
1110 void * deviceContext = NULL) {
\n-
1111
\n-
1112 (void)instance; // unused
\n-
1113 (void)deviceContext; // unused
\n-
1114
\n-
1115 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
1116 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
1117 duBuffer->BindCpuBuffer(), duDesc,
\n-
1118 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
1119 numPatchCoords,
\n-
1120 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
1121 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
1122 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1123 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
1124 }
\n-
1125
\n-
1186 template <typename SRC_BUFFER, typename DST_BUFFER,
\n-
1187 typename PATCHCOORD_BUFFER, typename PATCH_TABLE>
\n-
1188 static bool EvalPatchesFaceVarying(
\n-
1189 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,
\n-
1190 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,
\n-
1191 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,
\n-
1192 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,
\n-
1193 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,
\n-
1194 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,
\n-
1195 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,
\n-
1196 int numPatchCoords,
\n-
1197 PATCHCOORD_BUFFER *patchCoords,
\n-
1198 PATCH_TABLE *patchTable,
\n-
1199 int fvarChannel,
\n-
1200 CpuEvaluator const *instance = NULL,
\n-
1201 void * deviceContext = NULL) {
\n-
1202
\n-
1203 (void)instance; // unused
\n-
1204 (void)deviceContext; // unused
\n-
1205
\n-
1206 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,
\n-
1207 dstBuffer->BindCpuBuffer(), dstDesc,
\n-
1208 duBuffer->BindCpuBuffer(), duDesc,
\n-
1209 dvBuffer->BindCpuBuffer(), dvDesc,
\n-
1210 duuBuffer->BindCpuBuffer(), duuDesc,
\n-
1211 duvBuffer->BindCpuBuffer(), duvDesc,
\n-
1212 dvvBuffer->BindCpuBuffer(), dvvDesc,
\n-
1213 numPatchCoords,
\n-
1214 (const PatchCoord*)patchCoords->BindCpuBuffer(),
\n-
1215 patchTable->GetFVarPatchArrayBuffer(fvarChannel),
\n-
1216 patchTable->GetFVarPatchIndexBuffer(fvarChannel),
\n-
1217 patchTable->GetFVarPatchParamBuffer(fvarChannel));
\n-
1218 }
\n-
1219
\n-
1225
\n-
1227 static void Synchronize(void * /*deviceContext = NULL*/) {
\n-
1228 // nothing.
\n-
1229 }
\n-
1230};
\n-
1231
\n-
1232
\n-
1233} // end namespace Osd
\n-
1234
\n-
1235} // end namespace OPENSUBDIV_VERSION
\n-
1236using namespace OPENSUBDIV_VERSION;
\n-
1237
\n-
1238} // end namespace OpenSubdiv
\n-
1239
\n-
1240
\n-
1241#endif // OPENSUBDIV3_OSD_CPU_EVALUATOR_H
\n+
86 bool operator == (BufferDescriptor const &other) const {
\n+
87 return (offset == other.offset &&
\n+
88 length == other.length &&
\n+
89 stride == other.stride);
\n+
90 }
\n+
91
\n+
93 bool operator != (BufferDescriptor const &other) const {
\n+
94 return !(this->operator==(other));
\n+
95 }
\n+
96
\n+
98 int offset;
\n+
100 int length;
\n+
102 int stride;
\n+
103};
\n+
104
\n+
105} // end namespace Osd
\n+
106
\n+
107} // end namespace OPENSUBDIV_VERSION
\n+
108using namespace OPENSUBDIV_VERSION;
\n+
109
\n+
110} // end namespace OpenSubdiv
\n+
111
\n+
112#endif // OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H
\n \n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:1034
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:1100
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, int start, int end)
Static eval stencils function which takes raw CPU pointers for input and output.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying
static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:1188
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:971
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw CPU pointers for input and output.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int *offsets, const int *indices, const float *weights, const float *duWeights, const float *dvWeights, const float *duuWeights, const float *duvWeights, const float *dvvWeights, int start, int end)
Static eval stencils function with derivatives, which takes raw CPU pointers for input and output.
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu, BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc, float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer, PatchParam const *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:454
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: cpuEvaluator.h:517
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const &duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords, PatchCoord const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer, PatchParam const *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable, const CpuEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: cpuEvaluator.h:303
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrays, const int *patchIndexBuffer, const PatchParam *patchParamBuffer)
Static limit eval function. It takes an array of PatchCoord and evaluate limit values on given PatchT...
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:886
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying
static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function. This function has a same signature as other device kernels have so that ...
Definition: cpuEvaluator.h:823
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::Synchronize
static void Synchronize(void *)
synchronize all asynchronous computation invoked on this device.
Definition: cpuEvaluator.h:1227
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, const CpuEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function. This function has a same signature as other device kernels hav...
Definition: cpuEvaluator.h:72
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches
static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)
Generic limit eval function with derivatives. This function has a same signature as other device kern...
Definition: cpuEvaluator.h:607
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils
static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const CpuEvaluator *instance=NULL, void *deviceContext=NULL)
Generic static eval stencils function with derivatives. This function has a same signature as other d...
Definition: cpuEvaluator.h:168
\n-
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord
Coordinates set on a patch table.
Definition: types.h:42
\n-\n-\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor
BufferDescriptor is a struct which describes buffer elements in interleaved data buffers....
Definition: bufferDescriptor.h:61
\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::Reset
void Reset()
Resets the descriptor to default.
Definition: bufferDescriptor.h:81
\n+\n+\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::operator!=
bool operator!=(BufferDescriptor const &other) const
True if the descriptors are not identical.
Definition: bufferDescriptor.h:93
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::operator==
bool operator==(BufferDescriptor const &other) const
True if the descriptors are identical.
Definition: bufferDescriptor.h:86
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::IsValid
bool IsValid() const
True if the descriptor values are internally consistent.
Definition: bufferDescriptor.h:75
\n+
OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::GetLocalOffset
int GetLocalOffset() const
Returns the relative offset within a stride.
Definition: bufferDescriptor.h:70
\n+\n
\n \n \n
\n \n
\n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -4,15 +4,15 @@\n \n \n \n \n [Logo] OpenSubdiv\n \n \n-cpuEvaluator.h\n+bufferDescriptor.h\n Go_to_the_documentation_of_this_file.\n 1//\n 2// Copyright 2015 Pixar\n 3//\n 4// Licensed under the Apache License, Version 2.0 (the \"Apache License\")\n 5// with the following modification; you may not use this file except in\n 6// compliance with the Apache License and the following modification to it:\n@@ -30,657 +30,125 @@\n 18// Unless required by applicable law or agreed to in writing, software\n 19// distributed under the Apache License with the above modification is\n 20// distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n 21// KIND, either express or implied. See the Apache License for the specific\n 22// language governing permissions and limitations under the Apache License.\n 23//\n 24\n- 25#ifndef OPENSUBDIV3_OSD_CPU_EVALUATOR_H\n- 26#define OPENSUBDIV3_OSD_CPU_EVALUATOR_H\n+ 25#ifndef OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H\n+ 26#define OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H\n 27\n 28#include \"../version.h\"\n- 29#include \"../osd/bufferDescriptor.h\"\n- 30#include \"../osd/types.h\"\n- 31\n- 32#include \n+ 29#include \n+ 30\n+ 31namespace OpenSubdiv {\n+ 32namespace OPENSUBDIV_VERSION {\n 33\n- 34namespace OpenSubdiv {\n- 35namespace OPENSUBDIV_VERSION {\n- 36\n- 37namespace Osd {\n- 38\n-39class CpuEvaluator {\n- 40public:\n+34namespace Osd {\n+ 35\n 46\n- 71 template \n-72 static bool EvalStencils(\n- 73 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 74 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 75 STENCIL_TABLE const *stencilTable,\n- 76 const CpuEvaluator *instance = NULL,\n- 77 void * deviceContext = NULL) {\n- 78\n- 79 (void)instance; // unused\n- 80 (void)deviceContext; // unused\n- 81\n- 82 if (stencilTable->GetNumStencils() == 0)\n- 83 return false;\n+ 47// example:\n+ 48// n\n+ 49// -----+----------------------------------------+-------------------------\n+ 50// | vertex 0 |\n+ 51// -----+----------------------------------------+-------------------------\n+ 52// | X Y Z R G B A Xu Yu Zu Xv Yv Zv |\n+ 53// -----+----------------------------------------+-------------------------\n+ 54// <------------- stride = 13 -------------->\n+ 55//\n+ 56// - XYZ (offset = n+0, length = 3, stride = 13)\n+ 57// - RGBA (offset = n+3, length = 4, stride = 13)\n+ 58// - uTangent (offset = n+7, length = 3, stride = 13)\n+ 59// - vTangent (offset = n+10, length = 3, stride = 13)\n+ 60//\n+61struct BufferDescriptor {\n+ 62\n+64 BufferDescriptor() : offset(0), length(0), stride(0) { }\n+ 65\n+67 BufferDescriptor(int o, int l, int s) : offset(o), length(l), stride(s) { }\n+ 68\n+70 int GetLocalOffset() const {\n+ 71 return stride > 0 ? offset % stride : 0;\n+ 72 }\n+ 73\n+75 bool IsValid() const {\n+ 76 return ((length > 0) &&\n+ 77 (length <= stride - GetLocalOffset()));\n+ 78 }\n+ 79\n+81 void Reset() {\n+ 82 offset = length = stride = 0;\n+ 83 }\n 84\n- 85 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n- 86 dstBuffer->BindCpuBuffer(), dstDesc,\n- 87 &stencilTable->GetSizes()[0],\n- 88 &stencilTable->GetOffsets()[0],\n- 89 &stencilTable->GetControlIndices()[0],\n- 90 &stencilTable->GetWeights()[0],\n- 91 /*start = */ 0,\n- 92 /*end = */ stencilTable->GetNumStencils());\n- 93 }\n- 94\n-121 static bool EvalStencils(\n- 122 const float *src, BufferDescriptor const &srcDesc,\n- 123 float *dst, BufferDescriptor const &dstDesc,\n- 124 const int * sizes,\n- 125 const int * offsets,\n- 126 const int * indices,\n- 127 const float * weights,\n- 128 int start, int end);\n- 129\n- 167 template \n-168 static bool EvalStencils(\n- 169 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 170 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 171 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 172 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 173 STENCIL_TABLE const *stencilTable,\n- 174 const CpuEvaluator *instance = NULL,\n- 175 void * deviceContext = NULL) {\n- 176\n- 177 (void)instance; // unused\n- 178 (void)deviceContext; // unused\n- 179\n- 180 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n- 181 dstBuffer->BindCpuBuffer(), dstDesc,\n- 182 duBuffer->BindCpuBuffer(), duDesc,\n- 183 dvBuffer->BindCpuBuffer(), dvDesc,\n- 184 &stencilTable->GetSizes()[0],\n- 185 &stencilTable->GetOffsets()[0],\n- 186 &stencilTable->GetControlIndices()[0],\n- 187 &stencilTable->GetWeights()[0],\n- 188 &stencilTable->GetDuWeights()[0],\n- 189 &stencilTable->GetDvWeights()[0],\n- 190 /*start = */ 0,\n- 191 /*end = */ stencilTable->GetNumStencils());\n- 192 }\n- 193\n-234 static bool EvalStencils(\n- 235 const float *src, BufferDescriptor const &srcDesc,\n- 236 float *dst, BufferDescriptor const &dstDesc,\n- 237 float *du, BufferDescriptor const &duDesc,\n- 238 float *dv, BufferDescriptor const &dvDesc,\n- 239 const int * sizes,\n- 240 const int * offsets,\n- 241 const int * indices,\n- 242 const float * weights,\n- 243 const float * duWeights,\n- 244 const float * dvWeights,\n- 245 int start, int end);\n- 246\n- 302 template \n-303 static bool EvalStencils(\n- 304 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 305 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 306 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 307 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 308 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 309 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 310 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 311 STENCIL_TABLE const *stencilTable,\n- 312 const CpuEvaluator *instance = NULL,\n- 313 void * deviceContext = NULL) {\n- 314\n- 315 (void)instance; // unused\n- 316 (void)deviceContext; // unused\n- 317\n- 318 return EvalStencils(srcBuffer->BindCpuBuffer(), srcDesc,\n- 319 dstBuffer->BindCpuBuffer(), dstDesc,\n- 320 duBuffer->BindCpuBuffer(), duDesc,\n- 321 dvBuffer->BindCpuBuffer(), dvDesc,\n- 322 duuBuffer->BindCpuBuffer(), duuDesc,\n- 323 duvBuffer->BindCpuBuffer(), duvDesc,\n- 324 dvvBuffer->BindCpuBuffer(), dvvDesc,\n- 325 &stencilTable->GetSizes()[0],\n- 326 &stencilTable->GetOffsets()[0],\n- 327 &stencilTable->GetControlIndices()[0],\n- 328 &stencilTable->GetWeights()[0],\n- 329 &stencilTable->GetDuWeights()[0],\n- 330 &stencilTable->GetDvWeights()[0],\n- 331 &stencilTable->GetDuuWeights()[0],\n- 332 &stencilTable->GetDuvWeights()[0],\n- 333 &stencilTable->GetDvvWeights()[0],\n- 334 /*start = */ 0,\n- 335 /*end = */ stencilTable->GetNumStencils());\n- 336 }\n- 337\n-399 static bool EvalStencils(\n- 400 const float *src, BufferDescriptor const &srcDesc,\n- 401 float *dst, BufferDescriptor const &dstDesc,\n- 402 float *du, BufferDescriptor const &duDesc,\n- 403 float *dv, BufferDescriptor const &dvDesc,\n- 404 float *duu, BufferDescriptor const &duuDesc,\n- 405 float *duv, BufferDescriptor const &duvDesc,\n- 406 float *dvv, BufferDescriptor const &dvvDesc,\n- 407 const int * sizes,\n- 408 const int * offsets,\n- 409 const int * indices,\n- 410 const float * weights,\n- 411 const float * duWeights,\n- 412 const float * dvWeights,\n- 413 const float * duuWeights,\n- 414 const float * duvWeights,\n- 415 const float * dvvWeights,\n- 416 int start, int end);\n- 417\n- 423\n- 452 template \n-454 static bool EvalPatches(\n- 455 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 456 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 457 int numPatchCoords,\n- 458 PATCHCOORD_BUFFER *patchCoords,\n- 459 PATCH_TABLE *patchTable,\n- 460 CpuEvaluator const *instance = NULL,\n- 461 void * deviceContext = NULL) {\n- 462\n- 463 (void)instance; // unused\n- 464 (void)deviceContext; // unused\n- 465\n- 466 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 467 dstBuffer->BindCpuBuffer(), dstDesc,\n- 468 numPatchCoords,\n- 469 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 470 patchTable->GetPatchArrayBuffer(),\n- 471 patchTable->GetPatchIndexBuffer(),\n- 472 patchTable->GetPatchParamBuffer());\n- 473 }\n- 474\n- 515 template \n-517 static bool EvalPatches(\n- 518 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 519 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 520 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 521 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 522 int numPatchCoords,\n- 523 PATCHCOORD_BUFFER *patchCoords,\n- 524 PATCH_TABLE *patchTable,\n- 525 CpuEvaluator const *instance = NULL,\n- 526 void * deviceContext = NULL) {\n- 527\n- 528 (void)instance; // unused\n- 529 (void)deviceContext; // unused\n- 530\n- 531 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.\n- 532 // ideally all buffer classes should have templated by datatype\n- 533 // so that downcast isn't needed there.\n- 534 // (e.g. Osd::CpuBuffer )\n- 535 //\n- 536 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 537 dstBuffer->BindCpuBuffer(), dstDesc,\n- 538 duBuffer->BindCpuBuffer(), duDesc,\n- 539 dvBuffer->BindCpuBuffer(), dvDesc,\n- 540 numPatchCoords,\n- 541 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 542 patchTable->GetPatchArrayBuffer(),\n- 543 patchTable->GetPatchIndexBuffer(),\n- 544 patchTable->GetPatchParamBuffer());\n- 545 }\n- 546\n- 605 template \n-607 static bool EvalPatches(\n- 608 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 609 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 610 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 611 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 612 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 613 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 614 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 615 int numPatchCoords,\n- 616 PATCHCOORD_BUFFER *patchCoords,\n- 617 PATCH_TABLE *patchTable,\n- 618 CpuEvaluator const *instance = NULL,\n- 619 void * deviceContext = NULL) {\n- 620\n- 621 (void)instance; // unused\n- 622 (void)deviceContext; // unused\n- 623\n- 624 // XXX: PatchCoords is somewhat abusing vertex primvar buffer interop.\n- 625 // ideally all buffer classes should have templated by datatype\n- 626 // so that downcast isn't needed there.\n- 627 // (e.g. Osd::CpuBuffer )\n- 628 //\n- 629 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 630 dstBuffer->BindCpuBuffer(), dstDesc,\n- 631 duBuffer->BindCpuBuffer(), duDesc,\n- 632 dvBuffer->BindCpuBuffer(), dvDesc,\n- 633 duuBuffer->BindCpuBuffer(), duuDesc,\n- 634 duvBuffer->BindCpuBuffer(), duvDesc,\n- 635 dvvBuffer->BindCpuBuffer(), dvvDesc,\n- 636 numPatchCoords,\n- 637 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 638 patchTable->GetPatchArrayBuffer(),\n- 639 patchTable->GetPatchIndexBuffer(),\n- 640 patchTable->GetPatchParamBuffer());\n- 641 }\n- 642\n-670 static bool EvalPatches(\n- 671 const float *src, BufferDescriptor const &srcDesc,\n- 672 float *dst, BufferDescriptor const &dstDesc,\n- 673 int numPatchCoords,\n- 674 const PatchCoord *patchCoords,\n- 675 const PatchArray *patchArrays,\n- 676 const int *patchIndexBuffer,\n- 677 const PatchParam *patchParamBuffer);\n- 678\n-716 static bool EvalPatches(\n- 717 const float *src, BufferDescriptor const &srcDesc,\n- 718 float *dst, BufferDescriptor const &dstDesc,\n- 719 float *du, BufferDescriptor const &duDesc,\n- 720 float *dv, BufferDescriptor const &dvDesc,\n- 721 int numPatchCoords,\n- 722 PatchCoord const *patchCoords,\n- 723 PatchArray const *patchArrays,\n- 724 const int *patchIndexBuffer,\n- 725 PatchParam const *patchParamBuffer);\n- 726\n-779 static bool EvalPatches(\n- 780 const float *src, BufferDescriptor const &srcDesc,\n- 781 float *dst, BufferDescriptor const &dstDesc,\n- 782 float *du, BufferDescriptor const &duDesc,\n- 783 float *dv, BufferDescriptor const &dvDesc,\n- 784 float *duu, BufferDescriptor const &duuDesc,\n- 785 float *duv, BufferDescriptor const &duvDesc,\n- 786 float *dvv, BufferDescriptor const &dvvDesc,\n- 787 int numPatchCoords,\n- 788 PatchCoord const *patchCoords,\n- 789 PatchArray const *patchArrays,\n- 790 const int *patchIndexBuffer,\n- 791 PatchParam const *patchParamBuffer);\n- 792\n- 821 template \n-823 static bool EvalPatchesVarying(\n- 824 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 825 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 826 int numPatchCoords,\n- 827 PATCHCOORD_BUFFER *patchCoords,\n- 828 PATCH_TABLE *patchTable,\n- 829 CpuEvaluator const *instance = NULL,\n- 830 void * deviceContext = NULL) {\n- 831\n- 832 (void)instance; // unused\n- 833 (void)deviceContext; // unused\n- 834\n- 835 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 836 dstBuffer->BindCpuBuffer(), dstDesc,\n- 837 numPatchCoords,\n- 838 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 839 patchTable->GetVaryingPatchArrayBuffer(),\n- 840 patchTable->GetVaryingPatchIndexBuffer(),\n- 841 patchTable->GetPatchParamBuffer());\n- 842 }\n- 843\n- 884 template \n-886 static bool EvalPatchesVarying(\n- 887 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 888 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 889 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 890 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 891 int numPatchCoords,\n- 892 PATCHCOORD_BUFFER *patchCoords,\n- 893 PATCH_TABLE *patchTable,\n- 894 CpuEvaluator const *instance = NULL,\n- 895 void * deviceContext = NULL) {\n- 896\n- 897 (void)instance; // unused\n- 898 (void)deviceContext; // unused\n- 899\n- 900 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 901 dstBuffer->BindCpuBuffer(), dstDesc,\n- 902 duBuffer->BindCpuBuffer(), duDesc,\n- 903 dvBuffer->BindCpuBuffer(), dvDesc,\n- 904 numPatchCoords,\n- 905 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 906 patchTable->GetVaryingPatchArrayBuffer(),\n- 907 patchTable->GetVaryingPatchIndexBuffer(),\n- 908 patchTable->GetPatchParamBuffer());\n- 909 }\n- 910\n- 969 template \n-971 static bool EvalPatchesVarying(\n- 972 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 973 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 974 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 975 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 976 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 977 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 978 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 979 int numPatchCoords,\n- 980 PATCHCOORD_BUFFER *patchCoords,\n- 981 PATCH_TABLE *patchTable,\n- 982 CpuEvaluator const *instance = NULL,\n- 983 void * deviceContext = NULL) {\n- 984\n- 985 (void)instance; // unused\n- 986 (void)deviceContext; // unused\n- 987\n- 988 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 989 dstBuffer->BindCpuBuffer(), dstDesc,\n- 990 duBuffer->BindCpuBuffer(), duDesc,\n- 991 dvBuffer->BindCpuBuffer(), dvDesc,\n- 992 duuBuffer->BindCpuBuffer(), duuDesc,\n- 993 duvBuffer->BindCpuBuffer(), duvDesc,\n- 994 dvvBuffer->BindCpuBuffer(), dvvDesc,\n- 995 numPatchCoords,\n- 996 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 997 patchTable->GetVaryingPatchArrayBuffer(),\n- 998 patchTable->GetVaryingPatchIndexBuffer(),\n- 999 patchTable->GetPatchParamBuffer());\n- 1000 }\n- 1001\n- 1032 template \n-1034 static bool EvalPatchesFaceVarying(\n- 1035 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1036 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1037 int numPatchCoords,\n- 1038 PATCHCOORD_BUFFER *patchCoords,\n- 1039 PATCH_TABLE *patchTable,\n- 1040 int fvarChannel,\n- 1041 CpuEvaluator const *instance = NULL,\n- 1042 void * deviceContext = NULL) {\n- 1043\n- 1044 (void)instance; // unused\n- 1045 (void)deviceContext; // unused\n- 1046\n- 1047 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 1048 dstBuffer->BindCpuBuffer(), dstDesc,\n- 1049 numPatchCoords,\n- 1050 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 1051 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 1052 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1053 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 1054 }\n- 1055\n- 1098 template \n-1100 static bool EvalPatchesFaceVarying(\n- 1101 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1102 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1103 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1104 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1105 int numPatchCoords,\n- 1106 PATCHCOORD_BUFFER *patchCoords,\n- 1107 PATCH_TABLE *patchTable,\n- 1108 int fvarChannel,\n- 1109 CpuEvaluator const *instance = NULL,\n- 1110 void * deviceContext = NULL) {\n- 1111\n- 1112 (void)instance; // unused\n- 1113 (void)deviceContext; // unused\n- 1114\n- 1115 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 1116 dstBuffer->BindCpuBuffer(), dstDesc,\n- 1117 duBuffer->BindCpuBuffer(), duDesc,\n- 1118 dvBuffer->BindCpuBuffer(), dvDesc,\n- 1119 numPatchCoords,\n- 1120 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 1121 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 1122 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1123 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 1124 }\n- 1125\n- 1186 template \n-1188 static bool EvalPatchesFaceVarying(\n- 1189 SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n- 1190 DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n- 1191 DST_BUFFER *duBuffer, BufferDescriptor const &duDesc,\n- 1192 DST_BUFFER *dvBuffer, BufferDescriptor const &dvDesc,\n- 1193 DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc,\n- 1194 DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc,\n- 1195 DST_BUFFER *dvvBuffer, BufferDescriptor const &dvvDesc,\n- 1196 int numPatchCoords,\n- 1197 PATCHCOORD_BUFFER *patchCoords,\n- 1198 PATCH_TABLE *patchTable,\n- 1199 int fvarChannel,\n- 1200 CpuEvaluator const *instance = NULL,\n- 1201 void * deviceContext = NULL) {\n- 1202\n- 1203 (void)instance; // unused\n- 1204 (void)deviceContext; // unused\n- 1205\n- 1206 return EvalPatches(srcBuffer->BindCpuBuffer(), srcDesc,\n- 1207 dstBuffer->BindCpuBuffer(), dstDesc,\n- 1208 duBuffer->BindCpuBuffer(), duDesc,\n- 1209 dvBuffer->BindCpuBuffer(), dvDesc,\n- 1210 duuBuffer->BindCpuBuffer(), duuDesc,\n- 1211 duvBuffer->BindCpuBuffer(), duvDesc,\n- 1212 dvvBuffer->BindCpuBuffer(), dvvDesc,\n- 1213 numPatchCoords,\n- 1214 (const PatchCoord*)patchCoords->BindCpuBuffer(),\n- 1215 patchTable->GetFVarPatchArrayBuffer(fvarChannel),\n- 1216 patchTable->GetFVarPatchIndexBuffer(fvarChannel),\n- 1217 patchTable->GetFVarPatchParamBuffer(fvarChannel));\n- 1218 }\n- 1219\n- 1225\n-1227 static void Synchronize(void * /*deviceContext = NULL*/) {\n- 1228 // nothing.\n- 1229 }\n- 1230};\n- 1231\n- 1232\n- 1233} // end namespace Osd\n- 1234\n- 1235} // end namespace OPENSUBDIV_VERSION\n- 1236using namespace OPENSUBDIV_VERSION;\n- 1237\n- 1238} // end namespace OpenSubdiv\n- 1239\n- 1240\n- 1241#endif // OPENSUBDIV3_OSD_CPU_EVALUATOR_H\n+86 bool operator_==(BufferDescriptor const &other) const {\n+ 87 return (offset == other.offset &&\n+ 88 length == other.length &&\n+ 89 stride == other.stride);\n+ 90 }\n+ 91\n+93 bool operator_!=(BufferDescriptor const &other) const {\n+ 94 return !(this->operator==(other));\n+ 95 }\n+ 96\n+98 int offset;\n+100 int length;\n+102 int stride;\n+ 103};\n+ 104\n+ 105} // end namespace Osd\n+ 106\n+ 107} // end namespace OPENSUBDIV_VERSION\n+ 108using namespace OPENSUBDIV_VERSION;\n+ 109\n+ 110} // end namespace OpenSubdiv\n+ 111\n+ 112#endif // OPENSUBDIV3_OSD_BUFFER_DESCRIPTOR_H\n OpenSubdiv\n Definition: limits.h:32\n OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor\n BufferDescriptor is a struct which describes buffer elements in interleaved\n data buffers....\n Definition: bufferDescriptor.h:61\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator\n-Definition: cpuEvaluator.h:39\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int\n-fvarChannel, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cpuEvaluator.h:1034\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, int fvarChannel, CpuEvaluator const\n-*instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cpuEvaluator.h:1100\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n-static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, const int *sizes, const int\n-*offsets, const int *indices, const float *weights, int start, int end)\n-Static eval stencils function which takes raw CPU pointers for input and\n-output.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesFaceVarying\n-static bool EvalPatchesFaceVarying(SRC_BUFFER *srcBuffer, BufferDescriptor\n-const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc,\n-DST_BUFFER *duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, int fvarChannel,\n-CpuEvaluator const *instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cpuEvaluator.h:1188\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const\n-*instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cpuEvaluator.h:971\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n-static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, const int *sizes, const int\n-*offsets, const int *indices, const float *weights, const float *duWeights,\n-const float *dvWeights, int start, int end)\n-Static eval stencils function with derivatives, which takes raw CPU pointers\n-for input and output.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n-static bool EvalStencils(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n-BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n-float *dvv, BufferDescriptor const &dvvDesc, const int *sizes, const int\n-*offsets, const int *indices, const float *weights, const float *duWeights,\n-const float *dvWeights, const float *duuWeights, const float *duvWeights, const\n-float *dvvWeights, int start, int end)\n-Static eval stencils function with derivatives, which takes raw CPU pointers\n-for input and output.\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n-static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, float *duu,\n-BufferDescriptor const &duuDesc, float *duv, BufferDescriptor const &duvDesc,\n-float *dvv, BufferDescriptor const &dvvDesc, int numPatchCoords, PatchCoord\n-const *patchCoords, PatchArray const *patchArrays, const int *patchIndexBuffer,\n-PatchParam const *patchParamBuffer)\n-Static limit eval function. It takes an array of PatchCoord and evaluate limit\n-values on given PatchT...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int numPatchCoords,\n-PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const\n-*instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cpuEvaluator.h:454\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, int numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE\n-*patchTable, CpuEvaluator const *instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: cpuEvaluator.h:517\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n-static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, float *du, BufferDescriptor const\n-&duDesc, float *dv, BufferDescriptor const &dvDesc, int numPatchCoords,\n-PatchCoord const *patchCoords, PatchArray const *patchArrays, const int\n-*patchIndexBuffer, PatchParam const *patchParamBuffer)\n-Static limit eval function. It takes an array of PatchCoord and evaluate limit\n-values on given PatchT...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const\n-&duuDesc, DST_BUFFER *duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER\n-*dvvBuffer, BufferDescriptor const &dvvDesc, STENCIL_TABLE const *stencilTable,\n-const CpuEvaluator *instance=NULL, void *deviceContext=NULL)\n-Generic static eval stencils function with derivatives. This function has a\n-same signature as other d...\n-Definition: cpuEvaluator.h:303\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n-static bool EvalPatches(const float *src, BufferDescriptor const &srcDesc,\n-float *dst, BufferDescriptor const &dstDesc, int numPatchCoords, const\n-PatchCoord *patchCoords, const PatchArray *patchArrays, const int\n-*patchIndexBuffer, const PatchParam *patchParamBuffer)\n-Static limit eval function. It takes an array of PatchCoord and evaluate limit\n-values on given PatchT...\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void\n-*deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cpuEvaluator.h:886\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatchesVarying\n-static bool EvalPatchesVarying(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, int\n-numPatchCoords, PATCHCOORD_BUFFER *patchCoords, PATCH_TABLE *patchTable,\n-CpuEvaluator const *instance=NULL, void *deviceContext=NULL)\n-Generic limit eval function. This function has a same signature as other device\n-kernels have so that ...\n-Definition: cpuEvaluator.h:823\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::Synchronize\n-static void Synchronize(void *)\n-synchronize all asynchronous computation invoked on this device.\n-Definition: cpuEvaluator.h:1227\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE\n-const *stencilTable, const CpuEvaluator *instance=NULL, void\n-*deviceContext=NULL)\n-Generic static eval stencils function. This function has a same signature as\n-other device kernels hav...\n-Definition: cpuEvaluator.h:72\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalPatches\n-static bool EvalPatches(SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc,\n-DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER *duBuffer,\n-BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer, BufferDescriptor const\n-&dvDesc, DST_BUFFER *duuBuffer, BufferDescriptor const &duuDesc, DST_BUFFER\n-*duvBuffer, BufferDescriptor const &duvDesc, DST_BUFFER *dvvBuffer,\n-BufferDescriptor const &dvvDesc, int numPatchCoords, PATCHCOORD_BUFFER\n-*patchCoords, PATCH_TABLE *patchTable, CpuEvaluator const *instance=NULL, void\n-*deviceContext=NULL)\n-Generic limit eval function with derivatives. This function has a same\n-signature as other device kern...\n-Definition: cpuEvaluator.h:607\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuEvaluator::EvalStencils\n-static bool EvalStencils(SRC_BUFFER *srcBuffer, BufferDescriptor const\n-&srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, DST_BUFFER\n-*duBuffer, BufferDescriptor const &duDesc, DST_BUFFER *dvBuffer,\n-BufferDescriptor const &dvDesc, STENCIL_TABLE const *stencilTable, const\n-CpuEvaluator *instance=NULL, void *deviceContext=NULL)\n-Generic static eval stencils function with derivatives. This function has a\n-same signature as other d...\n-Definition: cpuEvaluator.h:168\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord\n-Coordinates set on a patch table.\n-Definition: types.h:42\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArray\n-Definition: types.h:66\n-OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParam\n-Definition: types.h:127\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::stride\n+int stride\n+stride to the next element\n+Definition: bufferDescriptor.h:102\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::Reset\n+void Reset()\n+Resets the descriptor to default.\n+Definition: bufferDescriptor.h:81\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::BufferDescriptor\n+BufferDescriptor()\n+Default Constructor.\n+Definition: bufferDescriptor.h:64\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::BufferDescriptor\n+BufferDescriptor(int o, int l, int s)\n+Constructor.\n+Definition: bufferDescriptor.h:67\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::length\n+int length\n+number or length of the data\n+Definition: bufferDescriptor.h:100\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::operator!=\n+bool operator!=(BufferDescriptor const &other) const\n+True if the descriptors are not identical.\n+Definition: bufferDescriptor.h:93\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::operator==\n+bool operator==(BufferDescriptor const &other) const\n+True if the descriptors are identical.\n+Definition: bufferDescriptor.h:86\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::IsValid\n+bool IsValid() const\n+True if the descriptor values are internally consistent.\n+Definition: bufferDescriptor.h:75\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::GetLocalOffset\n+int GetLocalOffset() const\n+Returns the relative offset within a stride.\n+Definition: bufferDescriptor.h:70\n+OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::offset\n+int offset\n+offset to desired element data\n+Definition: bufferDescriptor.h:98\n * opensubdiv\n * osd\n- * cpuEvaluator.h\n+ * bufferDescriptor.h\n * Generated on Sun Dec 11 2022 09:52:28 for OpenSubdiv by [doxygen] 1.9.4\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00908.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00908.html", "unified_diff": "@@ -273,15 +273,15 @@\n \n \n \n \n
typedef Vtr::ConstArray<PatchDescriptor> ConstPatchDescriptorArray
\n
\n \n-

Definition at line 146 of file patchDescriptor.h.

\n+

Definition at line 146 of file patchDescriptor.h.

\n \n
\n \n \n

◆ ConstPatchParamArray

\n \n
\n@@ -289,15 +289,15 @@\n \n \n \n \n
typedef Vtr::ConstArray<PatchParam> ConstPatchParamArray
\n
\n \n-

Definition at line 246 of file patchParam.h.

\n+

Definition at line 246 of file patchParam.h.

\n \n
\n \n \n

◆ ErrorCallbackFunc

\n \n
\n@@ -387,15 +387,15 @@\n \n \n \n \n
typedef Vtr::Array<PatchParam> PatchParamArray
\n
\n \n-

Definition at line 245 of file patchParam.h.

\n+

Definition at line 245 of file patchParam.h.

\n \n
\n \n \n

◆ PatchParamTable

\n \n
\n@@ -403,15 +403,15 @@\n \n \n \n \n
typedef std::vector<PatchParam> PatchParamTable
\n
\n \n-

Definition at line 243 of file patchParam.h.

\n+

Definition at line 243 of file patchParam.h.

\n \n
\n \n \n

◆ WarningCallbackFunc

\n \n
\n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00912.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00912.html", "unified_diff": "@@ -284,15 +284,15 @@\n \n \n \n \n
typedef MeshInterface<D3D11PatchTable> D3D11MeshInterface
\n
\n \n-

Definition at line 38 of file d3d11Mesh.h.

\n+

Definition at line 38 of file d3d11Mesh.h.

\n \n
\n \n \n

◆ GLMeshInterface

\n \n
\n@@ -300,15 +300,15 @@\n \n \n \n \n
typedef MeshInterface<GLPatchTable> GLMeshInterface
\n
\n \n-

Definition at line 38 of file glMesh.h.

\n+

Definition at line 38 of file glMesh.h.

\n \n
\n \n \n

◆ MeshBitset

\n \n
\n@@ -332,15 +332,15 @@\n \n \n \n \n
typedef MeshInterface<MTLPatchTable> MTLMeshInterface
\n
\n \n-

Definition at line 38 of file mtlMesh.h.

\n+

Definition at line 38 of file mtlMesh.h.

\n \n
\n \n \n

◆ PatchArrayVector

\n \n
\n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00925.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00925.html", "unified_diff": "@@ -89,15 +89,15 @@\n
RefinerSurfaceFactoryBase Class Reference
\n
\n
\n \n

Intermediate subclass of SurfaceFactory with Far::TopologyRefiner as the mesh. \n More...

\n \n-

#include <refinerSurfaceFactory.h>

\n+

#include <refinerSurfaceFactory.h>

\n
\n Inheritance diagram for RefinerSurfaceFactoryBase:
\n
\n
\n \"\"/\n \n \"SurfaceFactory\"\n@@ -229,15 +229,15 @@\n

\n

Detailed Description

\n

Intermediate subclass of SurfaceFactory with Far::TopologyRefiner as the mesh.

\n

RefinerSurfaceFactoryBase is an intermediate subclass of SurfaceFactory using Far::TopologyRefiner as the connected mesh representation.

\n

The SurfaceFactoryMeshAdapter interface for TopologyRefiner is provided in full, along with some public extensions specific to TopologyRefiner.

\n

Additional caching expectations of SurfaceFactory are NOT specified here. These are deferred to subclasses to implement different behaviors of the factory's internal caching. A template for such subclasses is additionally provided – allowing clients desiring a thread-safe cache to simply declare a subclass for a preferred thread-safe type.

\n \n-

Definition at line 58 of file refinerSurfaceFactory.h.

\n+

Definition at line 58 of file refinerSurfaceFactory.h.

\n

Constructor & Destructor Documentation

\n \n

◆ RefinerSurfaceFactoryBase()

\n \n
\n
\n \n@@ -310,15 +310,15 @@\n inline\n \n
\n
\n \n

Return the instance of the mesh.

\n \n-

Definition at line 79 of file refinerSurfaceFactory.h.

\n+

Definition at line 79 of file refinerSurfaceFactory.h.

\n \n
\n
\n \n

◆ GetNumFaces()

\n \n
\n@@ -339,15 +339,15 @@\n inline \n \n \n
\n \n

Return the number of faces.

\n \n-

Definition at line 82 of file refinerSurfaceFactory.h.

\n+

Definition at line 82 of file refinerSurfaceFactory.h.

\n \n
\n
\n \n

◆ GetNumFVarChannels()

\n \n
\n@@ -368,20 +368,20 @@\n inline \n \n \n
\n \n

Return the number of face-varying channels.

\n \n-

Definition at line 85 of file refinerSurfaceFactory.h.

\n+

Definition at line 85 of file refinerSurfaceFactory.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Bfr
  • RefinerSurfaceFactoryBase
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00929.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00929.html", "unified_diff": "@@ -90,15 +90,15 @@\n
    RefinerSurfaceFactory< CACHE_TYPE > Class Template Reference
    \n
\n
\n \n

Template for concrete subclasses of RefinerSurfaceFactoryBase. \n More...

\n \n-

#include <refinerSurfaceFactory.h>

\n+

#include <refinerSurfaceFactory.h>

\n
\n Inheritance diagram for RefinerSurfaceFactory< CACHE_TYPE >:
\n
\n
\n \"\"/\n \n \"RefinerSurfaceFactoryBase\"\n@@ -234,15 +234,15 @@\n \n \n
CACHE_TYPEA subclass of SurfaceFactoryCache
\n \n \n

Note a default template parameter uses the base SurfaceFactoryCache for convenience, but which is not thread-safe.

\n \n-

Definition at line 157 of file refinerSurfaceFactory.h.

\n+

Definition at line 157 of file refinerSurfaceFactory.h.

\n

Constructor & Destructor Documentation

\n \n

◆ RefinerSurfaceFactory()

\n \n
\n
\n \n@@ -270,15 +270,15 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 159 of file refinerSurfaceFactory.h.

\n+

Definition at line 159 of file refinerSurfaceFactory.h.

\n \n
\n
\n \n

◆ ~RefinerSurfaceFactory()

\n \n
\n@@ -300,15 +300,15 @@\n \n \n
\n \n
\n
\n
The documentation for this class was generated from the following file:\n
\n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Bfr
  • RefinerSurfaceFactory
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00933.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00933.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    Surface< REAL > Class Template Reference
    \n
\n
\n \n

Encapsulates the limit surface for a face of a mesh. \n More...

\n \n-

#include <surface.h>

\n+

#include <surface.h>

\n \n \n \n \n \n

\n Classes

struct  PointDescriptor
 Simple struct defining the size and stride of points in arrays. More...
 
\n@@ -221,15 +221,15 @@\n \n
REALFloating point precision (float or double only)
\n \n \n

Instances of Surface are created or initialized by a subclass of the SurfaceFactory. Since existing instances can be re-initialized, they should be tested for validity after such re-initialization.

\n

All Surfaces are assigned a Parameterization based on the subdivision scheme and the size of the face, which can then be used for evaluation and tessellation of the surface.

\n \n-

Definition at line 59 of file surface.h.

\n+

Definition at line 59 of file surface.h.

\n

Member Typedef Documentation

\n \n

◆ Index

\n \n
\n
\n \n@@ -237,15 +237,15 @@\n \n \n
typedef int Index
\n
\n \n

Integer type representing a mesh index.

\n \n-

Definition at line 72 of file surface.h.

\n+

Definition at line 72 of file surface.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ Surface() [1/2]

\n \n@@ -506,15 +506,15 @@\n inline \n \n \n
\n \n

Clear a previously initialized Surface.

\n \n-

Definition at line 89 of file surface.h.

\n+

Definition at line 89 of file surface.h.

\n \n
\n \n \n

◆ ComputePatchPoints()

\n \n
\n@@ -554,15 +554,15 @@\n \n \n \n
patchPointsArray of patch point data to be modified
patchPointDescThe size and stride of patch point data
\n \n \n \n-

Definition at line 331 of file surface.h.

\n+

Definition at line 331 of file surface.h.

\n \n
\n \n \n

◆ Evaluate() [1/3]

\n \n
\n@@ -606,15 +606,15 @@\n inline \n \n \n
\n \n

Evaluation of position.

\n \n-

Definition at line 372 of file surface.h.

\n+

Definition at line 372 of file surface.h.

\n \n
\n \n \n

◆ Evaluate() [2/3]

\n \n
\n@@ -670,15 +670,15 @@\n inline \n \n \n
\n \n

Overload of evaluation for 1st derivatives.

\n \n-

Definition at line 382 of file surface.h.

\n+

Definition at line 382 of file surface.h.

\n \n
\n \n \n

◆ Evaluate() [3/3]

\n \n
\n@@ -752,15 +752,15 @@\n inline \n \n \n
\n \n

Overload of evaluation for 2nd derivatives.

\n \n-

Definition at line 392 of file surface.h.

\n+

Definition at line 392 of file surface.h.

\n \n
\n \n \n

◆ EvaluateStencil() [1/3]

\n \n
\n@@ -792,15 +792,15 @@\n inline \n \n \n
\n \n

Evaluation of the limit stencil for position.

\n \n-

Definition at line 416 of file surface.h.

\n+

Definition at line 416 of file surface.h.

\n \n
\n \n \n

◆ EvaluateStencil() [2/3]

\n \n
\n@@ -844,15 +844,15 @@\n inline \n \n \n
\n \n

Overload of limit stencil evaluation for 1st derivatives.

\n \n-

Definition at line 423 of file surface.h.

\n+

Definition at line 423 of file surface.h.

\n \n
\n \n \n

◆ EvaluateStencil() [3/3]

\n \n
\n@@ -914,15 +914,15 @@\n inline \n \n \n
\n \n

Overload of limit stencil evaluation for 2nd derivatives.

\n \n-

Definition at line 431 of file surface.h.

\n+

Definition at line 431 of file surface.h.

\n \n
\n \n \n

◆ GatherControlPoints()

\n \n
\n@@ -1020,15 +1020,15 @@\n inline \n \n \n
\n \n

Return the size of the face.

\n \n-

Definition at line 109 of file surface.h.

\n+

Definition at line 109 of file surface.h.

\n \n
\n \n \n

◆ GetNumControlPoints()

\n \n
\n@@ -1049,15 +1049,15 @@\n inline \n \n \n
\n \n

Return the number of control points affecting the Surface.

\n \n-

Definition at line 137 of file surface.h.

\n+

Definition at line 137 of file surface.h.

\n \n
\n \n \n

◆ GetNumPatchPoints()

\n \n
\n@@ -1097,15 +1097,15 @@\n inline \n \n \n
\n \n

Return the Parameterization.

\n \n-

Definition at line 106 of file surface.h.

\n+

Definition at line 106 of file surface.h.

\n \n
\n \n \n

◆ IsLinear()

\n \n
\n@@ -1126,15 +1126,15 @@\n inline \n \n \n
\n \n

Return if the Surface is linear.

\n \n-

Definition at line 115 of file surface.h.

\n+

Definition at line 115 of file surface.h.

\n \n
\n \n \n

◆ IsRegular()

\n \n
\n@@ -1155,15 +1155,15 @@\n inline \n \n \n
\n \n

Return if the Surface is a single regular patch.

\n \n-

Definition at line 112 of file surface.h.

\n+

Definition at line 112 of file surface.h.

\n \n
\n \n \n

◆ IsValid()

\n \n
\n@@ -1184,15 +1184,15 @@\n inline \n \n \n
\n \n

Return true if successfully initialized.

\n \n-

Definition at line 86 of file surface.h.

\n+

Definition at line 86 of file surface.h.

\n \n
\n \n \n

◆ operator=()

\n \n
\n@@ -1273,15 +1273,15 @@\n patchPointsOutput array of patch point data \n patchPointDescThe size and stride of patch point data\n \n \n \n

Note that this method requires the mesh data be in a contiguous array. If a large data set is fragmented into blocks or pages, this method cannot be used. The control points will need to be gathered explicitly as the subset of patch points, after which the method to compute the remaining patch points can be used.

\n \n-

Definition at line 345 of file surface.h.

\n+

Definition at line 345 of file surface.h.

\n \n
\n \n

Friends And Related Function Documentation

\n \n

◆ SurfaceFactory

\n \n@@ -1298,20 +1298,20 @@\n \n \n friend \n \n \n
\n \n-

Definition at line 315 of file surface.h.

\n+

Definition at line 315 of file surface.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Bfr
  • Surface
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00937.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00937.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    Surface< REAL >::PointDescriptor Struct Reference
    \n
\n
\n \n

Simple struct defining the size and stride of points in arrays. \n More...

\n \n-

#include <surface.h>

\n+

#include <surface.h>

\n \n \n \n \n \n \n@@ -113,15 +113,15 @@\n \n \n

\n Public Member Functions

 PointDescriptor ()
 
 PointDescriptor (int n)
 
int stride
 
\n

Detailed Description

\n
template<typename REAL>
\n struct OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface< REAL >::PointDescriptor

Simple struct defining the size and stride of points in arrays.

\n \n-

Definition at line 63 of file surface.h.

\n+

Definition at line 63 of file surface.h.

\n

Constructor & Destructor Documentation

\n \n

◆ PointDescriptor() [1/3]

\n \n
\n
\n \n@@ -138,15 +138,15 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 64 of file surface.h.

\n+

Definition at line 64 of file surface.h.

\n \n
\n
\n \n

◆ PointDescriptor() [2/3]

\n \n
\n@@ -166,15 +166,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 65 of file surface.h.

\n+

Definition at line 65 of file surface.h.

\n \n
\n
\n \n

◆ PointDescriptor() [3/3]

\n \n
\n@@ -204,15 +204,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 66 of file surface.h.

\n+

Definition at line 66 of file surface.h.

\n \n
\n \n

Member Data Documentation

\n \n

◆ size

\n \n@@ -221,15 +221,15 @@\n \n \n \n \n
int size
\n
\n \n-

Definition at line 68 of file surface.h.

\n+

Definition at line 68 of file surface.h.

\n \n
\n \n \n

◆ stride

\n \n
\n@@ -237,20 +237,20 @@\n \n \n \n \n
int stride
\n
\n \n-

Definition at line 68 of file surface.h.

\n+

Definition at line 68 of file surface.h.

\n \n
\n \n
The documentation for this struct was generated from the following file:\n \n \n \n
\n
\n
\n \n

Describes the type of a patch. \n More...

\n \n-

#include <patchDescriptor.h>

\n+

#include <patchDescriptor.h>

\n \n \n \n \n

\n Public Types

enum  Type {
\n   NON_PATCH = 0\n ,
\n   POINTS\n@@ -185,15 +185,15 @@\n
 Returns a vector of all the legal patch descriptors for the given adaptive subdivision scheme. More...
 
\n

Detailed Description

\n

Describes the type of a patch.

\n

Uniquely identifies all the different types of patches

\n \n-

Definition at line 44 of file patchDescriptor.h.

\n+

Definition at line 44 of file patchDescriptor.h.

\n

Member Enumeration Documentation

\n \n

◆ Type

\n \n
\n
\n \n@@ -219,15 +219,15 @@\n \n \n \n \n \n
GREGORY 
GREGORY_BOUNDARY 
GREGORY_BASIS 
GREGORY_TRIANGLE 
\n \n-

Definition at line 48 of file patchDescriptor.h.

\n+

Definition at line 48 of file patchDescriptor.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ PatchDescriptor() [1/3]

\n \n@@ -249,15 +249,15 @@\n inline \n \n \n
\n \n

Default constructor.

\n \n-

Definition at line 69 of file patchDescriptor.h.

\n+

Definition at line 69 of file patchDescriptor.h.

\n \n
\n \n \n

◆ PatchDescriptor() [2/3]

\n \n
\n@@ -279,15 +279,15 @@\n inline \n \n \n
\n \n

Constructor.

\n \n-

Definition at line 73 of file patchDescriptor.h.

\n+

Definition at line 73 of file patchDescriptor.h.

\n \n
\n \n \n

◆ PatchDescriptor() [3/3]

\n \n
\n@@ -309,15 +309,15 @@\n inline \n \n \n
\n \n

Copy Constructor.

\n \n-

Definition at line 77 of file patchDescriptor.h.

\n+

Definition at line 77 of file patchDescriptor.h.

\n \n
\n \n

Member Function Documentation

\n \n

◆ GetAdaptivePatchDescriptors()

\n \n@@ -367,15 +367,15 @@\n inlinestatic \n \n \n
\n \n

Number of control vertices of Gregory patch basis (20)

\n \n-

Definition at line 126 of file patchDescriptor.h.

\n+

Definition at line 126 of file patchDescriptor.h.

\n \n
\n \n \n

◆ GetGregoryPatchSize()

\n \n
\n@@ -396,15 +396,15 @@\n inlinestatic \n \n \n
\n \n

Number of control vertices of Gregory (and Gregory Boundary) Patches in table.

\n \n-

Definition at line 123 of file patchDescriptor.h.

\n+

Definition at line 123 of file patchDescriptor.h.

\n \n
\n \n \n

◆ GetNumControlVertices() [1/2]

\n \n
\n@@ -425,15 +425,15 @@\n inline \n \n \n
\n \n

Returns the number of control vertices expected for a patch of the type described.

\n \n-

Definition at line 110 of file patchDescriptor.h.

\n+

Definition at line 110 of file patchDescriptor.h.

\n \n
\n \n \n

◆ GetNumControlVertices() [2/2]

\n \n
\n@@ -455,15 +455,15 @@\n inlinestatic \n \n \n
\n \n

Returns the number of control vertices expected for a patch of the type described.

\n \n-

Definition at line 150 of file patchDescriptor.h.

\n+

Definition at line 150 of file patchDescriptor.h.

\n \n
\n \n \n

◆ GetNumFVarControlVertices() [1/2]

\n \n
\n@@ -485,15 +485,15 @@\n \n \n
\n \n

Deprecated.

\n
See also
PatchDescriptor::GetNumControlVertices
\n \n-

Definition at line 115 of file patchDescriptor.h.

\n+

Definition at line 115 of file patchDescriptor.h.

\n \n
\n \n \n

◆ GetNumFVarControlVertices() [2/2]

\n \n
\n@@ -516,15 +516,15 @@\n \n \n
\n \n

Deprecated.

\n
See also
PatchDescriptor::GetNumControlVertices
\n \n-

Definition at line 168 of file patchDescriptor.h.

\n+

Definition at line 168 of file patchDescriptor.h.

\n \n
\n \n \n

◆ GetRegularPatchSize()

\n \n
\n@@ -545,15 +545,15 @@\n inlinestatic \n \n \n
\n \n

Number of control vertices of Regular Patches in table.

\n \n-

Definition at line 120 of file patchDescriptor.h.

\n+

Definition at line 120 of file patchDescriptor.h.

\n \n
\n \n \n

◆ GetType()

\n \n
\n@@ -574,15 +574,15 @@\n inline \n \n \n
\n \n

Returns the type of the patch.

\n \n-

Definition at line 87 of file patchDescriptor.h.

\n+

Definition at line 87 of file patchDescriptor.h.

\n \n
\n \n \n

◆ IsAdaptive() [1/2]

\n \n
\n@@ -603,15 +603,15 @@\n inline \n \n \n
\n \n

Returns true if the type is an adaptive patch.

\n \n-

Definition at line 97 of file patchDescriptor.h.

\n+

Definition at line 97 of file patchDescriptor.h.

\n \n
\n \n \n

◆ IsAdaptive() [2/2]

\n \n
\n@@ -633,15 +633,15 @@\n inlinestatic \n \n \n
\n \n

Returns true if the type is an adaptive (non-linear) patch.

\n \n-

Definition at line 92 of file patchDescriptor.h.

\n+

Definition at line 92 of file patchDescriptor.h.

\n \n
\n \n \n

◆ operator<()

\n \n
\n@@ -663,15 +663,15 @@\n inline \n \n \n
\n \n

Allows ordering of patches by type.

\n \n-

Definition at line 173 of file patchDescriptor.h.

\n+

Definition at line 173 of file patchDescriptor.h.

\n \n
\n \n \n

◆ operator=()

\n \n
\n@@ -693,15 +693,15 @@\n inline \n \n \n
\n \n

Assignment operator.

\n \n-

Definition at line 81 of file patchDescriptor.h.

\n+

Definition at line 81 of file patchDescriptor.h.

\n \n
\n \n \n

◆ operator==()

\n \n
\n@@ -723,15 +723,15 @@\n inline \n \n \n
\n \n

True if the descriptors are identical.

\n \n-

Definition at line 180 of file patchDescriptor.h.

\n+

Definition at line 180 of file patchDescriptor.h.

\n \n
\n \n \n

◆ print()

\n \n
\n@@ -745,15 +745,15 @@\n \n \n
\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • PatchDescriptor
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a00989.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a00989.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    PatchParam Struct Reference
    \n
\n
\n \n

Patch parameterization. \n More...

\n \n-

#include <patchParam.h>

\n+

#include <patchParam.h>

\n
\n Inheritance diagram for PatchParam:
\n
\n
\n \"\"/\n \n \"PatchParam\"\n@@ -242,15 +242,15 @@\n | \\ | \\ | | \\ | \\\n | L1 \\ | L1 \\ | | L1 \\ | L1 \\\n | \\ | \\ | | \\ | \\\n |0,0 \\ |1,0 \\ | | \\ | \\\n +---------------+---------------+ +---------------+---------------+\n (0,0) (1,0) (0,0,1) (1,0,0)\n \n-

Definition at line 152 of file patchParam.h.

\n+

Definition at line 152 of file patchParam.h.

\n

Member Function Documentation

\n \n

◆ Clear()

\n \n
\n
\n \n@@ -269,15 +269,15 @@\n inline\n \n
\n
\n \n

Resets everything to 0.

\n \n-

Definition at line 174 of file patchParam.h.

\n+

Definition at line 174 of file patchParam.h.

\n \n
\n
\n \n

◆ GetBoundary()

\n \n
\n@@ -298,15 +298,15 @@\n inline \n \n \n
\n \n

Returns the boundary edge encoding for the patch.

\n \n-

Definition at line 191 of file patchParam.h.

\n+

Definition at line 191 of file patchParam.h.

\n \n
\n
\n \n

◆ GetDepth()

\n \n
\n@@ -327,15 +327,15 @@\n inline \n \n \n
\n \n

Returns the level of subdivision of the patch.

\n \n-

Definition at line 197 of file patchParam.h.

\n+

Definition at line 197 of file patchParam.h.

\n \n
\n
\n \n

◆ GetFaceId()

\n \n
\n@@ -356,15 +356,15 @@\n inline \n \n \n
\n \n

Returns the faceid.

\n \n-

Definition at line 177 of file patchParam.h.

\n+

Definition at line 177 of file patchParam.h.

\n \n
\n \n \n

◆ GetParamFraction()

\n \n
\n@@ -385,15 +385,15 @@\n inline \n \n \n
\n \n

Returns the fraction of unit parametric space covered by this face.

\n \n-

Definition at line 265 of file patchParam.h.

\n+

Definition at line 265 of file patchParam.h.

\n \n
\n \n \n

◆ GetTransition()

\n \n
\n@@ -414,15 +414,15 @@\n inline \n \n \n
\n \n

Returns the transition edge encoding for the patch.

\n \n-

Definition at line 188 of file patchParam.h.

\n+

Definition at line 188 of file patchParam.h.

\n \n
\n \n \n

◆ GetU()

\n \n
\n@@ -443,15 +443,15 @@\n inline \n \n \n
\n \n

Returns the log2 value of the u parameter at the first corner of the patch.

\n \n-

Definition at line 181 of file patchParam.h.

\n+

Definition at line 181 of file patchParam.h.

\n \n
\n \n \n

◆ GetV()

\n \n
\n@@ -472,15 +472,15 @@\n inline \n \n \n
\n \n

Returns the log2 value of the v parameter at the first corner of the patch.

\n \n-

Definition at line 185 of file patchParam.h.

\n+

Definition at line 185 of file patchParam.h.

\n \n
\n \n \n

◆ IsRegular()

\n \n
\n@@ -501,15 +501,15 @@\n inline \n \n \n
\n \n

Returns whether the patch is regular.

\n \n-

Definition at line 228 of file patchParam.h.

\n+

Definition at line 228 of file patchParam.h.

\n \n
\n \n \n

◆ IsTriangleRotated()

\n \n
\n@@ -530,15 +530,15 @@\n inline \n \n \n
\n \n

Returns if a triangular patch is parametrically rotated 180 degrees.

\n \n-

Definition at line 290 of file patchParam.h.

\n+

Definition at line 290 of file patchParam.h.

\n \n
\n \n \n

◆ NonQuadRoot()

\n \n
\n@@ -559,15 +559,15 @@\n inline \n \n \n
\n \n

True if the parent base face is a non-quad.

\n \n-

Definition at line 194 of file patchParam.h.

\n+

Definition at line 194 of file patchParam.h.

\n \n
\n \n \n

◆ Normalize()

\n \n
\n@@ -606,15 +606,15 @@\n \n \n \n
uu parameter
vv parameter
\n \n \n \n-

Definition at line 271 of file patchParam.h.

\n+

Definition at line 271 of file patchParam.h.

\n \n
\n \n \n

◆ NormalizeTriangle()

\n \n
\n@@ -644,15 +644,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 297 of file patchParam.h.

\n+

Definition at line 297 of file patchParam.h.

\n \n
\n \n \n

◆ Set()

\n \n
\n@@ -733,15 +733,15 @@\n boundary5-bits identifying boundary edges (and verts for tris) \n transition4-bits identifying transition edges\n regularwhether the patch is regular \n \n \n \n \n-

Definition at line 249 of file patchParam.h.

\n+

Definition at line 249 of file patchParam.h.

\n \n
\n \n \n

◆ Unnormalize()

\n \n
\n@@ -780,15 +780,15 @@\n \n \n \n
uu parameter
vv parameter
\n \n \n \n-

Definition at line 281 of file patchParam.h.

\n+

Definition at line 281 of file patchParam.h.

\n \n
\n \n \n

◆ UnnormalizeTriangle()

\n \n
\n@@ -818,15 +818,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 312 of file patchParam.h.

\n+

Definition at line 312 of file patchParam.h.

\n \n
\n \n

Member Data Documentation

\n \n

◆ field0

\n \n@@ -835,15 +835,15 @@\n \n \n \n \n
unsigned int field0
\n
\n \n-

Definition at line 230 of file patchParam.h.

\n+

Definition at line 230 of file patchParam.h.

\n \n
\n \n \n

◆ field1

\n \n
\n@@ -851,20 +851,20 @@\n \n \n \n \n
unsigned int field1
\n
\n \n-

Definition at line 231 of file patchParam.h.

\n+

Definition at line 231 of file patchParam.h.

\n \n
\n \n
The documentation for this struct was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • PatchParam
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01021.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01021.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    PrimvarRefinerReal< REAL > Class Template Reference
    \n
\n
\n \n

Applies refinement operations to generic primvar data. \n More...

\n \n-

#include <primvarRefiner.h>

\n+

#include <primvarRefiner.h>

\n \n \n \n \n \n \n@@ -152,15 +152,15 @@\n \n \n

\n Public Member Functions

 PrimvarRefinerReal (TopologyRefiner const &refiner)
 
 ~PrimvarRefinerReal ()
 
void LimitFaceVarying (T const &src, U &dst, int channel=0) const
 
\n

Detailed Description

\n
template<typename REAL>
\n class OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal< REAL >

Applies refinement operations to generic primvar data.

\n \n-

Definition at line 56 of file primvarRefiner.h.

\n+

Definition at line 56 of file primvarRefiner.h.

\n

Constructor & Destructor Documentation

\n \n

◆ PrimvarRefinerReal()

\n \n
\n
\n \n@@ -178,15 +178,15 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 59 of file primvarRefiner.h.

\n+

Definition at line 59 of file primvarRefiner.h.

\n \n
\n
\n \n

◆ ~PrimvarRefinerReal()

\n \n
\n@@ -205,15 +205,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 60 of file primvarRefiner.h.

\n+

Definition at line 60 of file primvarRefiner.h.

\n \n
\n
\n

Member Function Documentation

\n \n

◆ GetTopologyRefiner()

\n \n@@ -233,15 +233,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 62 of file primvarRefiner.h.

\n+

Definition at line 62 of file primvarRefiner.h.

\n \n
\n \n \n

◆ Interpolate()

\n \n
\n@@ -288,15 +288,15 @@\n levelThe refinement level\n srcSource primvar buffer (templating control vertex data)\n dstDestination primvar buffer (templating refined vertex data) \n \n \n \n \n-

Definition at line 267 of file primvarRefiner.h.

\n+

Definition at line 267 of file primvarRefiner.h.

\n \n
\n \n \n

◆ InterpolateFaceUniform()

\n \n
\n@@ -344,15 +344,15 @@\n levelThe refinement level\n srcSource primvar buffer\n dstDestination primvar buffer \n \n \n \n \n-

Definition at line 394 of file primvarRefiner.h.

\n+

Definition at line 394 of file primvarRefiner.h.

\n \n
\n \n \n

◆ InterpolateFaceVarying()

\n \n
\n@@ -398,15 +398,15 @@\n \n
\n \n

Apply face-varying interpolation weights to a primvar buffer associated with a particular face-varying channel.

\n

Unlike vertex and varying primvar buffers, there is not a 1-to-1 correspondence between vertices and face-varying values – typically there are more face-varying values than vertices. Each face-varying channel is also independent in how its values relate to the vertices.

\n

The destination buffer must allocate an array of data for all the refined values, i.e. at least refiner.GetLevel(level).GetNumFVarValues(channel).

\n \n-

Definition at line 293 of file primvarRefiner.h.

\n+

Definition at line 293 of file primvarRefiner.h.

\n \n
\n \n \n

◆ InterpolateVarying()

\n \n
\n@@ -454,15 +454,15 @@\n levelThe refinement level\n srcSource primvar buffer (templating control vertex data)\n dstDestination primvar buffer (templating refined vertex data) \n \n \n \n \n-

Definition at line 412 of file primvarRefiner.h.

\n+

Definition at line 412 of file primvarRefiner.h.

\n \n
\n \n \n

◆ Limit() [1/2]

\n \n
\n@@ -502,15 +502,15 @@\n \n \n \n
srcSource primvar buffer (refined data) for last level
dstPosDestination primvar buffer (data at the limit)
\n \n \n \n-

Definition at line 319 of file primvarRefiner.h.

\n+

Definition at line 319 of file primvarRefiner.h.

\n \n
\n \n \n

◆ Limit() [2/2]

\n \n
\n@@ -552,15 +552,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 344 of file primvarRefiner.h.

\n+

Definition at line 344 of file primvarRefiner.h.

\n \n
\n \n \n

◆ LimitFaceVarying()

\n \n
\n@@ -596,20 +596,20 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 369 of file primvarRefiner.h.

\n+

Definition at line 369 of file primvarRefiner.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
\n \n-

#include <primvarRefiner.h>

\n+

#include <primvarRefiner.h>

\n
\n Inheritance diagram for PrimvarRefiner:
\n
\n
\n \"\"/\n \n \"PrimvarRefinerReal<\n@@ -131,15 +131,15 @@\n void Limit (T const &src, U &dstPos, U1 &dstTan1, U2 &dstTan2) const\n  \n void LimitFaceVarying (T const &src, U &dst, int channel=0) const\n  \n \n

Detailed Description

\n
\n-

Definition at line 1265 of file primvarRefiner.h.

\n+

Definition at line 1265 of file primvarRefiner.h.

\n

Constructor & Destructor Documentation

\n \n

◆ PrimvarRefiner()

\n \n
\n
\n \n@@ -157,20 +157,20 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 1267 of file primvarRefiner.h.

\n+

Definition at line 1267 of file primvarRefiner.h.

\n \n
\n
\n
The documentation for this class was generated from the following file:\n
\n
\n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • PrimvarRefiner
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01033.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01033.html", "unified_diff": "@@ -90,15 +90,15 @@\n
    PtexIndices Class Reference
    \n
\n
\n \n

Object used to compute and query ptex face indices. \n More...

\n \n-

#include <ptexIndices.h>

\n+

#include <ptexIndices.h>

\n \n \n \n \n \n \n@@ -114,15 +114,15 @@\n \n \n

\n Public Member Functions

 PtexIndices (TopologyRefiner const &refiner)
 Constructor. More...
 
 ~PtexIndices ()
 Returns ptex face adjacency information for a given coarse face. More...
 
\n

Detailed Description

\n

Object used to compute and query ptex face indices.

\n

Given a refiner, constructing a PtexIndices object builds the mapping from coarse faces to ptex ids. Once built, the object can be used to query the mapping.

\n \n-

Definition at line 46 of file ptexIndices.h.

\n+

Definition at line 46 of file ptexIndices.h.

\n

Constructor & Destructor Documentation

\n \n

◆ PtexIndices()

\n \n
\n
\n \n@@ -255,15 +255,15 @@\n \n

Returns the number of ptex faces in the mesh.

\n

Ptex

\n \n \n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • PtexIndices
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01037.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01037.html", "unified_diff": "@@ -92,15 +92,15 @@\n
    StencilTableFactoryReal< REAL > Class Template Reference
    \n
\n
\n \n

A specialized factory for StencilTable. \n More...

\n \n-

#include <stencilTableFactory.h>

\n+

#include <stencilTableFactory.h>

\n
\n \n \n \n

\n Classes

struct  Options
 
\n \n \n

\n@@ -133,15 +133,15 @@\n

 Utility function for stencil splicing for local point face-varying stencils. More...
 
\n

Detailed Description

\n
template<typename REAL>
\n class OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal< REAL >

A specialized factory for StencilTable.

\n \n-

Definition at line 51 of file stencilTableFactory.h.

\n+

Definition at line 51 of file stencilTableFactory.h.

\n

Member Enumeration Documentation

\n \n

◆ Mode

\n \n
\n
\n \n@@ -155,15 +155,15 @@\n \n \n \n
INTERPOLATE_VARYING 

varying primvar stencils

\n
INTERPOLATE_FACE_VARYING 

face-varying primvar stencils

\n
\n \n-

Definition at line 55 of file stencilTableFactory.h.

\n+

Definition at line 55 of file stencilTableFactory.h.

\n \n
\n
\n

Member Function Documentation

\n \n

◆ AppendLocalPointStencilTable()

\n \n@@ -342,15 +342,15 @@\n baseStencilTableInput StencilTable for refined vertices\n localPointStencilTableStencilTable for the change of basis patch points.\n factorizeIf factorize is set to true, endcap stencils will be factorized with supporting vertices from baseStencil table so that the endcap points can be computed directly from control vertices. \n \n \n \n \n-

Definition at line 148 of file stencilTableFactory.h.

\n+

Definition at line 148 of file stencilTableFactory.h.

\n \n
\n
\n \n

◆ Create() [1/2]

\n \n
\n@@ -439,16 +439,16 @@\n \n \n \n \n
\n
\n
The documentation for this class was generated from the following files:\n
\n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • StencilTableFactoryReal
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01041.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01041.html", "unified_diff": "@@ -92,15 +92,15 @@\n
    LimitStencilTableFactoryReal< REAL > Class Template Reference
    \n
\n
\n \n

A specialized factory for LimitStencilTable. \n More...

\n \n-

#include <stencilTableFactory.h>

\n+

#include <stencilTableFactory.h>

\n \n \n \n \n \n \n@@ -128,29 +128,29 @@\n

\n Classes

struct  LocationArray
 Descriptor for limit surface locations. More...
 
struct  Options
\n

Detailed Description

\n
template<typename REAL>
\n class OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal< REAL >

A specialized factory for LimitStencilTable.

\n

The LimitStencilTableFactory creates a table of limit stencils. Limit stencils can interpolate any arbitrary location on the limit surface. The stencils will be bilinear if the surface is refined uniformly, and bicubic if feature adaptive isolation is used instead.

\n

Surface locations are expressed as a combination of ptex face index and normalized (s,t) patch coordinates. The factory exposes the LocationArray struct as a container for these location descriptors.

\n \n-

Definition at line 209 of file stencilTableFactory.h.

\n+

Definition at line 209 of file stencilTableFactory.h.

\n

Member Typedef Documentation

\n \n

◆ LocationArrayVec

\n \n
\n
\n \n \n \n \n
typedef std::vector<LocationArray> LocationArrayVec
\n
\n \n-

Definition at line 244 of file stencilTableFactory.h.

\n+

Definition at line 244 of file stencilTableFactory.h.

\n \n
\n
\n

Member Enumeration Documentation

\n \n

◆ Mode

\n \n@@ -167,15 +167,15 @@\n \n INTERPOLATE_VARYING 

varying primvar stencils

\n \n INTERPOLATE_FACE_VARYING 

face-varying primvar stencils

\n \n \n \n-

Definition at line 213 of file stencilTableFactory.h.

\n+

Definition at line 213 of file stencilTableFactory.h.

\n \n
\n \n

Member Function Documentation

\n \n

◆ Create()

\n \n@@ -239,16 +239,16 @@\n \n \n \n \n \n \n
The documentation for this class was generated from the following files:\n \n \n \n
\n
\n
\n \n

Vertex stencil descriptor. \n More...

\n \n-

#include <stencilTable.h>

\n+

#include <stencilTable.h>

\n
\n Inheritance diagram for StencilReal< REAL >:
\n
\n
\n \"\"/\n \n \"LimitStencilReal<\n@@ -151,15 +151,15 @@\n  \n \n

Detailed Description

\n
template<typename REAL>
\n class OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal< REAL >

Vertex stencil descriptor.

\n

Allows access and manipulation of a single stencil in a StencilTable.

\n \n-

Definition at line 53 of file stencilTable.h.

\n+

Definition at line 53 of file stencilTable.h.

\n

Constructor & Destructor Documentation

\n \n

◆ StencilReal() [1/3]

\n \n
\n
\n \n@@ -178,15 +178,15 @@\n inline\n \n
\n
\n \n

Default constructor.

\n \n-

Definition at line 57 of file stencilTable.h.

\n+

Definition at line 57 of file stencilTable.h.

\n \n
\n
\n \n

◆ StencilReal() [2/3]

\n \n
\n@@ -232,15 +232,15 @@\n sizeTable pointer to the size of the stencil\n indicesTable pointer to the vertex indices of the stencil\n weightsTable pointer to the vertex weights of the stencil \n \n \n \n \n-

Definition at line 67 of file stencilTable.h.

\n+

Definition at line 67 of file stencilTable.h.

\n \n
\n
\n \n

◆ StencilReal() [3/3]

\n \n
\n@@ -262,15 +262,15 @@\n inline \n \n \n
\n \n

Copy constructor.

\n \n-

Definition at line 71 of file stencilTable.h.

\n+

Definition at line 71 of file stencilTable.h.

\n \n
\n
\n

Member Function Documentation

\n \n

◆ GetSize()

\n \n@@ -292,15 +292,15 @@\n inline \n \n \n
\n \n

Returns the size of the stencil.

\n \n-

Definition at line 78 of file stencilTable.h.

\n+

Definition at line 78 of file stencilTable.h.

\n \n
\n \n \n

◆ GetSizePtr()

\n \n
\n@@ -321,15 +321,15 @@\n inline \n \n \n
\n \n

Returns the size of the stencil as a pointer.

\n \n-

Definition at line 83 of file stencilTable.h.

\n+

Definition at line 83 of file stencilTable.h.

\n \n
\n \n \n

◆ GetVertexIndices()

\n \n
\n@@ -350,15 +350,15 @@\n inline \n \n \n
\n \n

Returns the control vertices' indices.

\n \n-

Definition at line 88 of file stencilTable.h.

\n+

Definition at line 88 of file stencilTable.h.

\n \n
\n \n \n

◆ GetWeights()

\n \n
\n@@ -379,15 +379,15 @@\n inline \n \n \n
\n \n

Returns the interpolation weights.

\n \n-

Definition at line 93 of file stencilTable.h.

\n+

Definition at line 93 of file stencilTable.h.

\n \n
\n \n \n

◆ Next()

\n \n
\n@@ -408,15 +408,15 @@\n inline \n \n \n
\n \n

Advance to the next stencil in the table.

\n \n-

Definition at line 98 of file stencilTable.h.

\n+

Definition at line 98 of file stencilTable.h.

\n \n
\n \n

Friends And Related Function Documentation

\n \n

◆ LimitStencilTableFactoryReal< REAL >

\n \n@@ -433,15 +433,15 @@\n \n \n friend \n \n \n
\n \n-

Definition at line 98 of file stencilTable.h.

\n+

Definition at line 98 of file stencilTable.h.

\n \n
\n \n \n

◆ StencilTableFactoryReal< REAL >

\n \n
\n@@ -457,15 +457,15 @@\n \n \n friend \n \n \n
\n \n-

Definition at line 98 of file stencilTable.h.

\n+

Definition at line 98 of file stencilTable.h.

\n \n
\n \n

Member Data Documentation

\n \n

◆ _indices

\n \n@@ -482,15 +482,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 110 of file stencilTable.h.

\n+

Definition at line 110 of file stencilTable.h.

\n \n
\n \n \n

◆ _size

\n \n
\n@@ -506,15 +506,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 109 of file stencilTable.h.

\n+

Definition at line 109 of file stencilTable.h.

\n \n
\n \n \n

◆ _weights

\n \n
\n@@ -530,20 +530,20 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 111 of file stencilTable.h.

\n+

Definition at line 111 of file stencilTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • StencilReal
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01049.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01049.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    Stencil Class Reference
    \n
\n
\n \n

Vertex stencil class wrapping the template for compatibility. \n More...

\n \n-

#include <stencilTable.h>

\n+

#include <stencilTable.h>

\n
\n Inheritance diagram for Stencil:
\n
\n
\n \"\"/\n \n \"StencilReal<\n@@ -154,15 +154,15 @@\n  \n float * _weights\n  \n \n

Detailed Description

\n

Vertex stencil class wrapping the template for compatibility.

\n \n-

Definition at line 116 of file stencilTable.h.

\n+

Definition at line 116 of file stencilTable.h.

\n

Member Typedef Documentation

\n \n

◆ BaseStencil

\n \n
\n
\n \n@@ -176,15 +176,15 @@\n \n \n \n
\n protected
\n
\n \n-

Definition at line 118 of file stencilTable.h.

\n+

Definition at line 118 of file stencilTable.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ Stencil() [1/3]

\n \n@@ -204,15 +204,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 121 of file stencilTable.h.

\n+

Definition at line 121 of file stencilTable.h.

\n \n
\n
\n \n

◆ Stencil() [2/3]

\n \n
\n@@ -232,15 +232,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 122 of file stencilTable.h.

\n+

Definition at line 122 of file stencilTable.h.

\n \n
\n
\n \n

◆ Stencil() [3/3]

\n \n
\n@@ -276,20 +276,20 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 123 of file stencilTable.h.

\n+

Definition at line 123 of file stencilTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • Stencil
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01053.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01053.html", "unified_diff": "@@ -93,15 +93,15 @@\n
    StencilTableReal< REAL > Class Template Reference
    \n
\n
\n \n

Table of subdivision stencils. \n More...

\n \n-

#include <stencilTable.h>

\n+

#include <stencilTable.h>

\n
\n Inheritance diagram for StencilTableReal< REAL >:
\n
\n
\n \"\"/\n \n \"LimitStencilTableReal<\n@@ -200,15 +200,15 @@\n \n

Detailed Description

\n
template<typename REAL>
\n class OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal< REAL >

Table of subdivision stencils.

\n

Stencils are the most direct method of evaluation of locations on the limit of a surface. Every point of a limit surface can be computed by linearly blending a collection of coarse control vertices.

\n

A stencil assigns a series of control vertex indices with a blending weight that corresponds to a unique parametric location of the limit surface. When the control vertices move in space, the limit location can be very efficiently recomputed simply by applying the blending weights to the series of coarse control vertices.

\n \n-

Definition at line 141 of file stencilTable.h.

\n+

Definition at line 141 of file stencilTable.h.

\n

Constructor & Destructor Documentation

\n \n

◆ StencilTableReal() [1/3]

\n \n
\n
\n \n@@ -291,15 +291,15 @@\n \n \n \n
\n inlinevirtual
\n
\n \n-

Definition at line 153 of file stencilTable.h.

\n+

Definition at line 153 of file stencilTable.h.

\n \n
\n
\n \n

◆ StencilTableReal() [2/3]

\n \n
\n@@ -318,15 +318,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 255 of file stencilTable.h.

\n+

Definition at line 255 of file stencilTable.h.

\n \n
\n
\n \n

◆ StencilTableReal() [3/3]

\n \n
\n@@ -346,15 +346,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 256 of file stencilTable.h.

\n+

Definition at line 256 of file stencilTable.h.

\n \n
\n
\n

Member Function Documentation

\n \n

◆ Clear()

\n \n@@ -390,15 +390,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 714 of file stencilTable.h.

\n+

Definition at line 714 of file stencilTable.h.

\n \n
\n \n \n

◆ generateOffsets()

\n \n
\n@@ -414,15 +414,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 678 of file stencilTable.h.

\n+

Definition at line 678 of file stencilTable.h.

\n \n
\n \n \n

◆ GetControlIndices()

\n \n
\n@@ -443,15 +443,15 @@\n inline \n \n \n
\n \n

Returns the indices of the control vertices.

\n \n-

Definition at line 179 of file stencilTable.h.

\n+

Definition at line 179 of file stencilTable.h.

\n \n
\n \n \n

◆ GetNumControlVertices()

\n \n
\n@@ -472,15 +472,15 @@\n inline \n \n \n
\n \n

Returns the number of control vertices indexed in the table.

\n \n-

Definition at line 161 of file stencilTable.h.

\n+

Definition at line 161 of file stencilTable.h.

\n \n
\n \n \n

◆ GetNumStencils()

\n \n
\n@@ -501,15 +501,15 @@\n inline \n \n \n
\n \n

Returns the number of stencils in the table.

\n \n-

Definition at line 156 of file stencilTable.h.

\n+

Definition at line 156 of file stencilTable.h.

\n \n
\n \n \n

◆ GetOffsets()

\n \n
\n@@ -530,15 +530,15 @@\n inline \n \n \n
\n \n

Returns the offset to a given stencil (factory may leave empty)

\n \n-

Definition at line 174 of file stencilTable.h.

\n+

Definition at line 174 of file stencilTable.h.

\n \n
\n \n \n

◆ GetSizes()

\n \n
\n@@ -559,15 +559,15 @@\n inline \n \n \n
\n \n

Returns the number of control vertices of each stencil in the table.

\n \n-

Definition at line 169 of file stencilTable.h.

\n+

Definition at line 169 of file stencilTable.h.

\n \n
\n \n \n

◆ GetStencil()

\n \n
\n@@ -589,15 +589,15 @@\n inline \n \n \n
\n \n

Returns a Stencil at index i in the table.

\n \n-

Definition at line 722 of file stencilTable.h.

\n+

Definition at line 722 of file stencilTable.h.

\n \n
\n \n \n

◆ GetWeights()

\n \n
\n@@ -618,15 +618,15 @@\n inline \n \n \n
\n \n

Returns the stencil interpolation weights.

\n \n-

Definition at line 184 of file stencilTable.h.

\n+

Definition at line 184 of file stencilTable.h.

\n \n
\n \n \n

◆ operator[]()

\n \n
\n@@ -648,15 +648,15 @@\n inline \n \n \n
\n \n

Returns the stencil at index i in the table.

\n \n-

Definition at line 734 of file stencilTable.h.

\n+

Definition at line 734 of file stencilTable.h.

\n \n
\n \n \n

◆ reserve()

\n \n
\n@@ -686,15 +686,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 698 of file stencilTable.h.

\n+

Definition at line 698 of file stencilTable.h.

\n \n
\n \n \n

◆ resize()

\n \n
\n@@ -724,15 +724,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 690 of file stencilTable.h.

\n+

Definition at line 690 of file stencilTable.h.

\n \n
\n \n \n

◆ shrinkToFit()

\n \n
\n@@ -748,15 +748,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 706 of file stencilTable.h.

\n+

Definition at line 706 of file stencilTable.h.

\n \n
\n \n \n

◆ update() [1/2]

\n \n
\n@@ -804,15 +804,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 650 of file stencilTable.h.

\n+

Definition at line 650 of file stencilTable.h.

\n \n
\n \n \n

◆ update() [2/2]

\n \n
\n@@ -872,15 +872,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 618 of file stencilTable.h.

\n+

Definition at line 618 of file stencilTable.h.

\n \n
\n \n \n

◆ UpdateValues() [1/4]

\n \n
\n@@ -934,15 +934,15 @@\n dstValuesDestination buffer for the interpolated primvar data\n startIndex of first destination value to update\n endIndex of last destination value to update \n \n \n \n \n-

Definition at line 205 of file stencilTable.h.

\n+

Definition at line 205 of file stencilTable.h.

\n \n
\n \n \n

◆ UpdateValues() [2/4]

\n \n
\n@@ -984,15 +984,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 217 of file stencilTable.h.

\n+

Definition at line 217 of file stencilTable.h.

\n \n
\n \n \n

◆ UpdateValues() [3/4]

\n \n
\n@@ -1046,15 +1046,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 210 of file stencilTable.h.

\n+

Definition at line 210 of file stencilTable.h.

\n \n
\n \n \n

◆ UpdateValues() [4/4]

\n \n
\n@@ -1108,15 +1108,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 221 of file stencilTable.h.

\n+

Definition at line 221 of file stencilTable.h.

\n \n
\n \n

Friends And Related Function Documentation

\n \n

◆ Far::PatchTableBuilder

\n \n@@ -1133,15 +1133,15 @@\n \n \n friend \n \n \n
\n \n-

Definition at line 261 of file stencilTable.h.

\n+

Definition at line 261 of file stencilTable.h.

\n \n
\n \n \n

◆ StencilTableFactoryReal< REAL >

\n \n
\n@@ -1157,15 +1157,15 @@\n \n \n friend \n \n \n
\n \n-

Definition at line 256 of file stencilTable.h.

\n+

Definition at line 256 of file stencilTable.h.

\n \n
\n \n

Member Data Documentation

\n \n

◆ _indices

\n \n@@ -1182,15 +1182,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 267 of file stencilTable.h.

\n+

Definition at line 267 of file stencilTable.h.

\n \n
\n \n \n

◆ _numControlVertices

\n \n
\n@@ -1206,15 +1206,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 263 of file stencilTable.h.

\n+

Definition at line 263 of file stencilTable.h.

\n \n
\n \n \n

◆ _offsets

\n \n
\n@@ -1230,15 +1230,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 266 of file stencilTable.h.

\n+

Definition at line 266 of file stencilTable.h.

\n \n
\n \n \n

◆ _sizes

\n \n
\n@@ -1254,15 +1254,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 265 of file stencilTable.h.

\n+

Definition at line 265 of file stencilTable.h.

\n \n
\n \n \n

◆ _weights

\n \n
\n@@ -1278,20 +1278,20 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 268 of file stencilTable.h.

\n+

Definition at line 268 of file stencilTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • StencilTableReal
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01057.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01057.html", "unified_diff": "@@ -92,15 +92,15 @@\n
    StencilTable Class Reference
    \n
\n
\n \n

Stencil table class wrapping the template for compatibility. \n More...

\n \n-

#include <stencilTable.h>

\n+

#include <stencilTable.h>

\n
\n Inheritance diagram for StencilTable:
\n
\n
\n \"\"/\n \n \"StencilTableReal<\n@@ -201,15 +201,15 @@\n  \n std::vector< float > _weights\n  \n \n

Detailed Description

\n

Stencil table class wrapping the template for compatibility.

\n \n-

Definition at line 273 of file stencilTable.h.

\n+

Definition at line 273 of file stencilTable.h.

\n

Member Typedef Documentation

\n \n

◆ BaseTable

\n \n
\n
\n \n@@ -223,15 +223,15 @@\n \n \n \n
\n protected
\n
\n \n-

Definition at line 275 of file stencilTable.h.

\n+

Definition at line 275 of file stencilTable.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ StencilTable() [1/3]

\n \n@@ -251,15 +251,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 286 of file stencilTable.h.

\n+

Definition at line 286 of file stencilTable.h.

\n \n
\n
\n \n

◆ StencilTable() [2/3]

\n \n
\n@@ -279,15 +279,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 287 of file stencilTable.h.

\n+

Definition at line 287 of file stencilTable.h.

\n \n
\n
\n \n

◆ StencilTable() [3/3]

\n \n
\n@@ -347,15 +347,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 288 of file stencilTable.h.

\n+

Definition at line 288 of file stencilTable.h.

\n \n
\n \n

Member Function Documentation

\n \n

◆ GetStencil()

\n \n@@ -376,15 +376,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 278 of file stencilTable.h.

\n+

Definition at line 278 of file stencilTable.h.

\n \n
\n \n \n

◆ operator[]()

\n \n
\n@@ -404,20 +404,20 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 281 of file stencilTable.h.

\n+

Definition at line 281 of file stencilTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • StencilTable
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01061.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01061.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    LimitStencilReal< REAL > Class Template Reference
    \n
\n
\n \n

Limit point stencil descriptor. \n More...

\n \n-

#include <stencilTable.h>

\n+

#include <stencilTable.h>

\n
\n Inheritance diagram for LimitStencilReal< REAL >:
\n
\n
\n \"\"/\n \n \"StencilReal<\n@@ -172,15 +172,15 @@\n REAL * _weights\n  \n \n

Detailed Description

\n
template<typename REAL>
\n class OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal< REAL >

Limit point stencil descriptor.

\n \n-

Definition at line 303 of file stencilTable.h.

\n+

Definition at line 303 of file stencilTable.h.

\n

Constructor & Destructor Documentation

\n \n

◆ LimitStencilReal()

\n \n
\n
\n \n@@ -259,15 +259,15 @@\n \n \n \n
duuWeightsTable pointer to the 'uu' derivative weights
duvWeightsTable pointer to the 'uv' derivative weights
dvvWeightsTable pointer to the 'vv' derivative weights
\n \n \n \n-

Definition at line 324 of file stencilTable.h.

\n+

Definition at line 324 of file stencilTable.h.

\n \n
\n
\n

Member Function Documentation

\n \n

◆ GetDuuWeights()

\n \n@@ -289,15 +289,15 @@\n inline \n \n \n
\n \n

Returns the uu derivative weights.

\n \n-

Definition at line 351 of file stencilTable.h.

\n+

Definition at line 351 of file stencilTable.h.

\n \n
\n
\n \n

◆ GetDuvWeights()

\n \n
\n@@ -318,15 +318,15 @@\n inline \n \n \n
\n \n

Returns the uv derivative weights.

\n \n-

Definition at line 356 of file stencilTable.h.

\n+

Definition at line 356 of file stencilTable.h.

\n \n
\n
\n \n

◆ GetDuWeights()

\n \n
\n@@ -347,15 +347,15 @@\n inline \n \n \n
\n \n

Returns the u derivative weights.

\n \n-

Definition at line 341 of file stencilTable.h.

\n+

Definition at line 341 of file stencilTable.h.

\n \n
\n \n \n

◆ GetDvvWeights()

\n \n
\n@@ -376,15 +376,15 @@\n inline \n \n \n
\n \n

Returns the vv derivative weights.

\n \n-

Definition at line 361 of file stencilTable.h.

\n+

Definition at line 361 of file stencilTable.h.

\n \n
\n \n \n

◆ GetDvWeights()

\n \n
\n@@ -405,15 +405,15 @@\n inline \n \n \n
\n \n

Returns the v derivative weights.

\n \n-

Definition at line 346 of file stencilTable.h.

\n+

Definition at line 346 of file stencilTable.h.

\n \n
\n \n \n

◆ Next()

\n \n
\n@@ -434,15 +434,15 @@\n inline \n \n \n
\n \n

Advance to the next stencil in the table.

\n \n-

Definition at line 366 of file stencilTable.h.

\n+

Definition at line 366 of file stencilTable.h.

\n \n
\n \n

Friends And Related Function Documentation

\n \n

◆ LimitStencilTableFactoryReal< REAL >

\n \n@@ -459,15 +459,15 @@\n \n \n friend \n \n \n
\n \n-

Definition at line 366 of file stencilTable.h.

\n+

Definition at line 366 of file stencilTable.h.

\n \n
\n \n \n

◆ StencilTableFactoryReal< REAL >

\n \n
\n@@ -483,20 +483,20 @@\n \n \n friend \n \n \n
\n \n-

Definition at line 366 of file stencilTable.h.

\n+

Definition at line 366 of file stencilTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • LimitStencilReal
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01065.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01065.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    LimitStencil Class Reference
    \n
\n
\n \n

Limit point stencil class wrapping the template for compatibility. \n More...

\n \n-

#include <stencilTable.h>

\n+

#include <stencilTable.h>

\n
\n Inheritance diagram for LimitStencil:
\n
\n
\n \"\"/\n \n \"LimitStencilReal<\n@@ -175,15 +175,15 @@\n  \n float * _weights\n  \n \n

Detailed Description

\n

Limit point stencil class wrapping the template for compatibility.

\n \n-

Definition at line 392 of file stencilTable.h.

\n+

Definition at line 392 of file stencilTable.h.

\n

Member Typedef Documentation

\n \n

◆ BaseStencil

\n \n
\n
\n \n@@ -197,15 +197,15 @@\n \n \n \n
\n protected
\n
\n \n-

Definition at line 394 of file stencilTable.h.

\n+

Definition at line 394 of file stencilTable.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ LimitStencil() [1/2]

\n \n@@ -226,15 +226,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 397 of file stencilTable.h.

\n+

Definition at line 397 of file stencilTable.h.

\n \n
\n
\n \n

◆ LimitStencil() [2/2]

\n \n
\n@@ -300,20 +300,20 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 398 of file stencilTable.h.

\n+

Definition at line 398 of file stencilTable.h.

\n \n
\n
\n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • LimitStencil
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01069.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01069.html", "unified_diff": "@@ -92,15 +92,15 @@\n
    LimitStencilTableReal< REAL > Class Template Reference
    \n
\n
\n \n

Table of limit subdivision stencils. \n More...

\n \n-

#include <stencilTable.h>

\n+

#include <stencilTable.h>

\n
\n Inheritance diagram for LimitStencilTableReal< REAL >:
\n
\n
\n \"\"/\n \n \"StencilTableReal<\n@@ -250,15 +250,15 @@\n std::vector< REAL > _weights\n  \n \n

Detailed Description

\n
template<typename REAL>
\n class OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal< REAL >

Table of limit subdivision stencils.

\n \n-

Definition at line 409 of file stencilTable.h.

\n+

Definition at line 409 of file stencilTable.h.

\n

Constructor & Destructor Documentation

\n \n

◆ LimitStencilTableReal()

\n \n
\n
\n \n@@ -393,15 +393,15 @@\n inline\n \n
\n
\n \n

Returns the 'uu' derivative stencil interpolation weights.

\n \n-

Definition at line 444 of file stencilTable.h.

\n+

Definition at line 444 of file stencilTable.h.

\n \n
\n
\n \n

◆ GetDuvWeights()

\n \n
\n@@ -422,15 +422,15 @@\n inline \n \n \n
\n \n

Returns the 'uv' derivative stencil interpolation weights.

\n \n-

Definition at line 449 of file stencilTable.h.

\n+

Definition at line 449 of file stencilTable.h.

\n \n
\n
\n \n

◆ GetDuWeights()

\n \n
\n@@ -451,15 +451,15 @@\n inline \n \n \n
\n \n

Returns the 'u' derivative stencil interpolation weights.

\n \n-

Definition at line 434 of file stencilTable.h.

\n+

Definition at line 434 of file stencilTable.h.

\n \n
\n
\n \n

◆ GetDvvWeights()

\n \n
\n@@ -480,15 +480,15 @@\n inline \n \n \n
\n \n

Returns the 'vv' derivative stencil interpolation weights.

\n \n-

Definition at line 454 of file stencilTable.h.

\n+

Definition at line 454 of file stencilTable.h.

\n \n
\n
\n \n

◆ GetDvWeights()

\n \n
\n@@ -509,15 +509,15 @@\n inline \n \n \n
\n \n

Returns the 'v' derivative stencil interpolation weights.

\n \n-

Definition at line 439 of file stencilTable.h.

\n+

Definition at line 439 of file stencilTable.h.

\n \n
\n \n \n

◆ GetLimitStencil()

\n \n
\n@@ -539,15 +539,15 @@\n inline \n \n \n
\n \n

Returns a LimitStencil at index i in the table.

\n \n-

Definition at line 749 of file stencilTable.h.

\n+

Definition at line 749 of file stencilTable.h.

\n \n
\n \n \n

◆ operator[]()

\n \n
\n@@ -569,15 +569,15 @@\n inline \n \n \n
\n \n

Returns the limit stencil at index i in the table.

\n \n-

Definition at line 782 of file stencilTable.h.

\n+

Definition at line 782 of file stencilTable.h.

\n \n
\n \n \n

◆ Update2ndDerivs() [1/4]

\n \n
\n@@ -645,15 +645,15 @@\n vvderivsDestination buffer for the interpolated 'vv' derivative primvar data\n startIndex of first destination derivative to update\n endIndex of last destination derivative to update \n \n \n \n \n-

Definition at line 528 of file stencilTable.h.

\n+

Definition at line 528 of file stencilTable.h.

\n \n
\n \n \n

◆ Update2ndDerivs() [2/4]

\n \n
\n@@ -707,15 +707,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 548 of file stencilTable.h.

\n+

Definition at line 548 of file stencilTable.h.

\n \n
\n \n \n

◆ Update2ndDerivs() [3/4]

\n \n
\n@@ -781,15 +781,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 538 of file stencilTable.h.

\n+

Definition at line 538 of file stencilTable.h.

\n \n
\n \n \n

◆ Update2ndDerivs() [4/4]

\n \n
\n@@ -855,15 +855,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 556 of file stencilTable.h.

\n+

Definition at line 556 of file stencilTable.h.

\n \n
\n \n \n

◆ UpdateDerivs() [1/4]

\n \n
\n@@ -924,15 +924,15 @@\n vderivsDestination buffer for the interpolated 'v' derivative primvar data\n startIndex of first destination derivative to update\n endIndex of last destination derivative to update \n \n \n \n \n-

Definition at line 476 of file stencilTable.h.

\n+

Definition at line 476 of file stencilTable.h.

\n \n
\n \n \n

◆ UpdateDerivs() [2/4]

\n \n
\n@@ -980,15 +980,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 493 of file stencilTable.h.

\n+

Definition at line 493 of file stencilTable.h.

\n \n
\n \n \n

◆ UpdateDerivs() [3/4]

\n \n
\n@@ -1048,15 +1048,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 484 of file stencilTable.h.

\n+

Definition at line 484 of file stencilTable.h.

\n \n
\n \n \n

◆ UpdateDerivs() [4/4]

\n \n
\n@@ -1116,15 +1116,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 500 of file stencilTable.h.

\n+

Definition at line 500 of file stencilTable.h.

\n \n
\n \n

Friends And Related Function Documentation

\n \n

◆ LimitStencilTableFactoryReal< REAL >

\n \n@@ -1141,20 +1141,20 @@\n \n \n friend \n \n \n
\n \n-

Definition at line 565 of file stencilTable.h.

\n+

Definition at line 565 of file stencilTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • LimitStencilTableReal
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01073.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01073.html", "unified_diff": "@@ -92,15 +92,15 @@\n
    LimitStencilTable Class Reference
    \n
\n
\n \n

Limit stencil table class wrapping the template for compatibility. \n More...

\n \n-

#include <stencilTable.h>

\n+

#include <stencilTable.h>

\n
\n Inheritance diagram for LimitStencilTable:
\n
\n
\n \"\"/\n \n \"LimitStencilTableReal<\n@@ -244,15 +244,15 @@\n  \n std::vector< float > _weights\n  \n \n

Detailed Description

\n

Limit stencil table class wrapping the template for compatibility.

\n \n-

Definition at line 583 of file stencilTable.h.

\n+

Definition at line 583 of file stencilTable.h.

\n

Member Typedef Documentation

\n \n

◆ BaseTable

\n \n
\n
\n \n@@ -266,15 +266,15 @@\n \n \n \n
\n protected
\n
\n \n-

Definition at line 585 of file stencilTable.h.

\n+

Definition at line 585 of file stencilTable.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ LimitStencilTable()

\n \n@@ -365,15 +365,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 596 of file stencilTable.h.

\n+

Definition at line 596 of file stencilTable.h.

\n \n
\n
\n

Member Function Documentation

\n \n

◆ GetLimitStencil()

\n \n@@ -394,15 +394,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 588 of file stencilTable.h.

\n+

Definition at line 588 of file stencilTable.h.

\n \n
\n \n \n

◆ operator[]()

\n \n
\n@@ -422,20 +422,20 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 591 of file stencilTable.h.

\n+

Definition at line 591 of file stencilTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
StencilTableFactoryReal< REAL >::Options Struct Reference
\n \n
\n \n-

#include <stencilTableFactory.h>

\n+

#include <stencilTableFactory.h>

\n \n \n \n \n

\n Public Member Functions

 Options ()
 
\n \n \n \n

\n@@ -120,15 +120,15 @@\n

 
unsigned int fvarChannel
 
\n

Detailed Description

\n
template<typename REAL>
\n struct OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal< REAL >::Options
\n-

Definition at line 61 of file stencilTableFactory.h.

\n+

Definition at line 61 of file stencilTableFactory.h.

\n

Constructor & Destructor Documentation

\n \n

◆ Options()

\n \n
\n
\n \n@@ -145,15 +145,15 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 63 of file stencilTableFactory.h.

\n+

Definition at line 63 of file stencilTableFactory.h.

\n \n
\n
\n

Member Data Documentation

\n \n

◆ factorizeIntermediateLevels

\n \n@@ -163,15 +163,15 @@\n \n unsigned int factorizeIntermediateLevels\n \n \n
\n

accumulate stencil weights from control vertices or from the stencils of the previous level

\n \n-

Definition at line 75 of file stencilTableFactory.h.

\n+

Definition at line 75 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ fvarChannel

\n \n
\n@@ -180,15 +180,15 @@\n \n unsigned int fvarChannel\n \n \n
\n

face-varying channel to use when generating face-varying stencils

\n \n-

Definition at line 79 of file stencilTableFactory.h.

\n+

Definition at line 79 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ generateControlVerts

\n \n
\n@@ -198,15 +198,15 @@\n unsigned int generateControlVerts\n \n \n
\n \n

generate stencils for control-vertices

\n \n-

Definition at line 73 of file stencilTableFactory.h.

\n+

Definition at line 73 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ generateIntermediateLevels

\n \n
\n@@ -216,15 +216,15 @@\n unsigned int generateIntermediateLevels\n \n \n
\n \n

vertices at all levels or highest only

\n \n-

Definition at line 74 of file stencilTableFactory.h.

\n+

Definition at line 74 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ generateOffsets

\n \n
\n@@ -234,15 +234,15 @@\n unsigned int generateOffsets\n \n \n
\n \n

populate optional \"_offsets\" field

\n \n-

Definition at line 72 of file stencilTableFactory.h.

\n+

Definition at line 72 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ interpolationMode

\n \n
\n@@ -252,15 +252,15 @@\n unsigned int interpolationMode\n \n \n
\n \n

interpolation mode

\n \n-

Definition at line 71 of file stencilTableFactory.h.

\n+

Definition at line 71 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ maxLevel

\n \n
\n@@ -270,20 +270,20 @@\n unsigned int maxLevel\n \n \n
\n \n

generate stencils up to 'maxLevel'

\n \n-

Definition at line 78 of file stencilTableFactory.h.

\n+

Definition at line 78 of file stencilTableFactory.h.

\n \n
\n \n
The documentation for this struct was generated from the following file:\n \n \n \n
\n
\n
LimitStencilTableFactoryReal< REAL >::Options Struct Reference
\n \n
\n \n-

#include <stencilTableFactory.h>

\n+

#include <stencilTableFactory.h>

\n \n \n \n \n

\n Public Member Functions

 Options ()
 
\n \n \n \n

\n@@ -113,15 +113,15 @@\n

unsigned int fvarChannel
 face-varying channel to use More...
 
\n

Detailed Description

\n
template<typename REAL>
\n struct OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal< REAL >::Options
\n-

Definition at line 219 of file stencilTableFactory.h.

\n+

Definition at line 219 of file stencilTableFactory.h.

\n

Constructor & Destructor Documentation

\n \n

◆ Options()

\n \n
\n
\n \n@@ -138,15 +138,15 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 221 of file stencilTableFactory.h.

\n+

Definition at line 221 of file stencilTableFactory.h.

\n \n
\n
\n

Member Data Documentation

\n \n

◆ fvarChannel

\n \n@@ -157,15 +157,15 @@\n unsigned int fvarChannel\n \n \n
\n \n

face-varying channel to use

\n \n-

Definition at line 229 of file stencilTableFactory.h.

\n+

Definition at line 229 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ generate1stDerivatives

\n \n
\n@@ -175,15 +175,15 @@\n unsigned int generate1stDerivatives\n \n \n
\n \n

Generate weights for 1st derivatives.

\n \n-

Definition at line 227 of file stencilTableFactory.h.

\n+

Definition at line 227 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ generate2ndDerivatives

\n \n
\n@@ -193,15 +193,15 @@\n unsigned int generate2ndDerivatives\n \n \n
\n \n

Generate weights for 2nd derivatives.

\n \n-

Definition at line 228 of file stencilTableFactory.h.

\n+

Definition at line 228 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ interpolationMode

\n \n
\n@@ -211,20 +211,20 @@\n unsigned int interpolationMode\n \n \n
\n \n

interpolation mode

\n \n-

Definition at line 226 of file stencilTableFactory.h.

\n+

Definition at line 226 of file stencilTableFactory.h.

\n \n
\n \n
The documentation for this struct was generated from the following file:\n \n \n \n
\n
\n
\n \n

Descriptor for limit surface locations. \n More...

\n \n-

#include <stencilTableFactory.h>

\n+

#include <stencilTableFactory.h>

\n \n \n \n \n

\n Public Member Functions

 LocationArray ()
 
\n \n \n

\n@@ -117,15 +117,15 @@\n

 array of v coordinates More...
 
\n

Detailed Description

\n
template<typename REAL>
\n struct OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal< REAL >::LocationArray

Descriptor for limit surface locations.

\n \n-

Definition at line 233 of file stencilTableFactory.h.

\n+

Definition at line 233 of file stencilTableFactory.h.

\n

Constructor & Destructor Documentation

\n \n

◆ LocationArray()

\n \n
\n
\n \n@@ -142,15 +142,15 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 235 of file stencilTableFactory.h.

\n+

Definition at line 235 of file stencilTableFactory.h.

\n \n
\n
\n

Member Data Documentation

\n \n

◆ numLocations

\n \n@@ -161,15 +161,15 @@\n int numLocations\n \n \n
\n \n

number of (u,v) coordinates in the array

\n \n-

Definition at line 238 of file stencilTableFactory.h.

\n+

Definition at line 238 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ ptexIdx

\n \n
\n@@ -179,15 +179,15 @@\n int ptexIdx\n \n \n
\n \n

ptex face index

\n \n-

Definition at line 237 of file stencilTableFactory.h.

\n+

Definition at line 237 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ s

\n \n
\n@@ -197,15 +197,15 @@\n REAL const* s\n \n \n
\n \n

array of u coordinates

\n \n-

Definition at line 240 of file stencilTableFactory.h.

\n+

Definition at line 240 of file stencilTableFactory.h.

\n \n
\n \n \n

◆ t

\n \n
\n@@ -215,20 +215,20 @@\n REAL const * t\n \n \n
\n \n

array of v coordinates

\n \n-

Definition at line 241 of file stencilTableFactory.h.

\n+

Definition at line 241 of file stencilTableFactory.h.

\n \n
\n \n
The documentation for this struct was generated from the following file:\n \n \n \n
\n
\n
\n \n

Stencil table factory class wrapping the template for compatibility. \n More...

\n \n-

#include <stencilTableFactory.h>

\n+

#include <stencilTableFactory.h>

\n
\n Inheritance diagram for StencilTableFactory:
\n
\n
\n \"\"/\n \n \"StencilTableFactoryReal<\n@@ -139,15 +139,15 @@\n \"-\"/ Public Types inherited from StencilTableFactoryReal< float >\n enum  Mode \n  \n \n

Detailed Description

\n

Stencil table factory class wrapping the template for compatibility.

\n \n-

Definition at line 287 of file stencilTableFactory.h.

\n+

Definition at line 287 of file stencilTableFactory.h.

\n

Member Function Documentation

\n \n

◆ AppendLocalPointStencilTable()

\n \n
\n
\n \n@@ -187,15 +187,15 @@\n \n \n \n
\n inlinestatic
\n
\n \n-

Definition at line 308 of file stencilTableFactory.h.

\n+

Definition at line 308 of file stencilTableFactory.h.

\n \n
\n
\n \n

◆ AppendLocalPointStencilTableFaceVarying()

\n \n
\n@@ -243,15 +243,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 334 of file stencilTableFactory.h.

\n+

Definition at line 334 of file stencilTableFactory.h.

\n \n
\n
\n \n

◆ AppendLocalPointStencilTableVarying()

\n \n
\n@@ -293,15 +293,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 321 of file stencilTableFactory.h.

\n+

Definition at line 321 of file stencilTableFactory.h.

\n \n
\n
\n \n

◆ Create() [1/2]

\n \n
\n@@ -331,15 +331,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 300 of file stencilTableFactory.h.

\n+

Definition at line 300 of file stencilTableFactory.h.

\n \n
\n
\n \n

◆ Create() [2/2]

\n \n
\n@@ -369,20 +369,20 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 293 of file stencilTableFactory.h.

\n+

Definition at line 293 of file stencilTableFactory.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • StencilTableFactory
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01093.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01093.html", "unified_diff": "@@ -90,15 +90,15 @@\n
    LimitStencilTableFactory Class Reference
    \n
\n
\n \n

Stencil table factory class wrapping the template for compatibility. \n More...

\n \n-

#include <stencilTableFactory.h>

\n+

#include <stencilTableFactory.h>

\n
\n Inheritance diagram for LimitStencilTableFactory:
\n
\n
\n \"\"/\n \n \"LimitStencilTableFactoryReal<\n@@ -121,15 +121,15 @@\n  \n typedef std::vector< LocationArray > LocationArrayVec\n  \n \n

Detailed Description

\n

Stencil table factory class wrapping the template for compatibility.

\n \n-

Definition at line 354 of file stencilTableFactory.h.

\n+

Definition at line 354 of file stencilTableFactory.h.

\n

Member Function Documentation

\n \n

◆ Create()

\n \n
\n
\n \n@@ -175,20 +175,20 @@\n \n \n \n
\n inlinestatic
\n
\n \n-

Definition at line 360 of file stencilTableFactory.h.

\n+

Definition at line 360 of file stencilTableFactory.h.

\n \n
\n
\n
The documentation for this class was generated from the following file:\n
\n
\n \n
\n
\n
\n \n

A simple reference to raw topology data for use with TopologyRefinerFactory. \n More...

\n \n-

#include <topologyDescriptor.h>

\n+

#include <topologyDescriptor.h>

\n \n \n \n \n

\n Classes

struct  FVarChannel
 
\n \n \n

\n@@ -141,15 +141,15 @@\n

FVarChannel const * fvarChannels
 
\n

Detailed Description

\n

A simple reference to raw topology data for use with TopologyRefinerFactory.

\n

TopologyDescriptor is a simple struct containing references to raw topology data used to construct a TopologyRefiner. It is not a requirement but a convenience for use with TopologyRefinerFactory when mesh topology is not available in an existing mesh data structure. It should be functionally complete and simple to use, but for more demanding situations, writing a custom Factory is usually warranted.

\n \n-

Definition at line 49 of file topologyDescriptor.h.

\n+

Definition at line 49 of file topologyDescriptor.h.

\n

Constructor & Destructor Documentation

\n \n

◆ TopologyDescriptor()

\n \n
\n
\n \n@@ -173,15 +173,15 @@\n
\n \n \n \n
Index const* cornerVertexIndices
\n
\n \n-

Definition at line 62 of file topologyDescriptor.h.

\n+

Definition at line 62 of file topologyDescriptor.h.

\n \n
\n
\n \n

◆ cornerWeights

\n \n
\n@@ -189,15 +189,15 @@\n \n \n \n \n
float const* cornerWeights
\n
\n \n-

Definition at line 63 of file topologyDescriptor.h.

\n+

Definition at line 63 of file topologyDescriptor.h.

\n \n
\n
\n \n

◆ creaseVertexIndexPairs

\n \n
\n@@ -205,15 +205,15 @@\n \n \n \n \n
Index const* creaseVertexIndexPairs
\n
\n \n-

Definition at line 58 of file topologyDescriptor.h.

\n+

Definition at line 58 of file topologyDescriptor.h.

\n \n
\n
\n \n

◆ creaseWeights

\n \n
\n@@ -221,15 +221,15 @@\n \n \n \n \n
float const* creaseWeights
\n
\n \n-

Definition at line 59 of file topologyDescriptor.h.

\n+

Definition at line 59 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ fvarChannels

\n \n
\n@@ -237,15 +237,15 @@\n \n \n \n \n
FVarChannel const* fvarChannels
\n
\n \n-

Definition at line 82 of file topologyDescriptor.h.

\n+

Definition at line 82 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ holeIndices

\n \n
\n@@ -253,15 +253,15 @@\n \n \n \n \n
Index const* holeIndices
\n
\n \n-

Definition at line 66 of file topologyDescriptor.h.

\n+

Definition at line 66 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ isLeftHanded

\n \n
\n@@ -269,15 +269,15 @@\n \n \n \n \n
bool isLeftHanded
\n
\n \n-

Definition at line 68 of file topologyDescriptor.h.

\n+

Definition at line 68 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ numCorners

\n \n
\n@@ -285,15 +285,15 @@\n \n \n \n \n
int numCorners
\n
\n \n-

Definition at line 61 of file topologyDescriptor.h.

\n+

Definition at line 61 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ numCreases

\n \n
\n@@ -301,15 +301,15 @@\n \n \n \n \n
int numCreases
\n
\n \n-

Definition at line 57 of file topologyDescriptor.h.

\n+

Definition at line 57 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ numFaces

\n \n
\n@@ -317,15 +317,15 @@\n \n \n \n \n
int numFaces
\n
\n \n-

Definition at line 52 of file topologyDescriptor.h.

\n+

Definition at line 52 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ numFVarChannels

\n \n
\n@@ -333,15 +333,15 @@\n \n \n \n \n
int numFVarChannels
\n
\n \n-

Definition at line 81 of file topologyDescriptor.h.

\n+

Definition at line 81 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ numHoles

\n \n
\n@@ -349,15 +349,15 @@\n \n \n \n \n
int numHoles
\n
\n \n-

Definition at line 65 of file topologyDescriptor.h.

\n+

Definition at line 65 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ numVertices

\n \n
\n@@ -365,15 +365,15 @@\n \n \n \n \n
int numVertices
\n
\n \n-

Definition at line 51 of file topologyDescriptor.h.

\n+

Definition at line 51 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ numVertsPerFace

\n \n
\n@@ -381,15 +381,15 @@\n \n \n \n \n
int const* numVertsPerFace
\n
\n \n-

Definition at line 54 of file topologyDescriptor.h.

\n+

Definition at line 54 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ vertIndicesPerFace

\n \n
\n@@ -397,20 +397,20 @@\n \n \n \n \n
Index const* vertIndicesPerFace
\n
\n \n-

Definition at line 55 of file topologyDescriptor.h.

\n+

Definition at line 55 of file topologyDescriptor.h.

\n \n
\n \n
The documentation for this struct was generated from the following file:\n \n \n \n
\n
\n
TopologyDescriptor::FVarChannel Struct Reference
\n \n
\n \n-

#include <topologyDescriptor.h>

\n+

#include <topologyDescriptor.h>

\n \n \n \n \n

\n Public Member Functions

 FVarChannel ()
 
\n \n \n \n \n

\n@@ -104,15 +104,15 @@\n

int numValues
 
Index const * valueIndices
 
\n

Detailed Description

\n
\n-

Definition at line 73 of file topologyDescriptor.h.

\n+

Definition at line 73 of file topologyDescriptor.h.

\n

Constructor & Destructor Documentation

\n \n

◆ FVarChannel()

\n \n
\n
\n \n@@ -129,15 +129,15 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 78 of file topologyDescriptor.h.

\n+

Definition at line 78 of file topologyDescriptor.h.

\n \n
\n
\n

Member Data Documentation

\n \n

◆ numValues

\n \n@@ -146,15 +146,15 @@\n \n \n \n \n
int numValues
\n
\n \n-

Definition at line 75 of file topologyDescriptor.h.

\n+

Definition at line 75 of file topologyDescriptor.h.

\n \n
\n \n \n

◆ valueIndices

\n \n
\n@@ -162,20 +162,20 @@\n \n \n \n \n
Index const* valueIndices
\n
\n \n-

Definition at line 76 of file topologyDescriptor.h.

\n+

Definition at line 76 of file topologyDescriptor.h.

\n \n
\n \n
The documentation for this struct was generated from the following file:\n \n \n \n
\n
\n
\n \n

An interface for accessing data in a specific level of a refined topology hierarchy. \n More...

\n \n-

#include <topologyLevel.h>

\n+

#include <topologyLevel.h>

\n \n \n \n \n \n@@ -251,15 +251,15 @@\n \n \n

\n Public Member Functions

Methods to inspect the overall inventory of components:

All three main component types are indexed locally within each level. For some topological relationships – notably face-vertices, which is often the only relationship of interest – the total number of entries is also made available.

\n
int GetNumVertices () const
 ~TopologyLevel ()
 
\n

Detailed Description

\n

An interface for accessing data in a specific level of a refined topology hierarchy.

\n

TopologyLevel provides an interface to data in a specific level of a topology hierarchy. Instances of TopologyLevel are created and owned by a TopologyRefiner, which will return const-references to them. Such references are only valid during the lifetime of the TopologyRefiner that created and returned them, and only for a given refinement, i.e. if the TopologyRefiner is re-refined, any references to TopoologyLevels are invalidated.

\n \n-

Definition at line 49 of file topologyLevel.h.

\n+

Definition at line 49 of file topologyLevel.h.

\n

Constructor & Destructor Documentation

\n \n

◆ TopologyLevel()

\n \n
\n
\n \n@@ -276,15 +276,15 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 300 of file topologyLevel.h.

\n+

Definition at line 300 of file topologyLevel.h.

\n \n
\n
\n \n

◆ ~TopologyLevel()

\n \n
\n@@ -303,15 +303,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 301 of file topologyLevel.h.

\n+

Definition at line 301 of file topologyLevel.h.

\n \n
\n
\n

Member Function Documentation

\n \n

◆ DoesEdgeFVarTopologyMatch()

\n \n@@ -344,15 +344,15 @@\n inline \n \n \n
\n \n

Return if face-varying topology across the edge only matches.

\n \n-

Definition at line 247 of file topologyLevel.h.

\n+

Definition at line 247 of file topologyLevel.h.

\n \n
\n \n \n

◆ DoesFaceFVarTopologyMatch()

\n \n
\n@@ -384,15 +384,15 @@\n inline \n \n \n
\n \n

Return if face-varying topology around a face matches.

\n \n-

Definition at line 252 of file topologyLevel.h.

\n+

Definition at line 252 of file topologyLevel.h.

\n \n
\n \n \n

◆ DoesVertexFVarTopologyMatch()

\n \n
\n@@ -424,15 +424,15 @@\n inline \n \n \n
\n \n

Return if face-varying topology around a vertex matches.

\n \n-

Definition at line 242 of file topologyLevel.h.

\n+

Definition at line 242 of file topologyLevel.h.

\n \n
\n \n \n

◆ FindEdge()

\n \n
\n@@ -464,15 +464,15 @@\n inline \n \n \n
\n \n

Identify the edge matching the given vertex pair.

\n \n-

Definition at line 120 of file topologyLevel.h.

\n+

Definition at line 120 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetEdgeChildEdges()

\n \n
\n@@ -494,15 +494,15 @@\n inline \n \n \n
\n \n

Access the child edges (in the next level) of a given edge.

\n \n-

Definition at line 268 of file topologyLevel.h.

\n+

Definition at line 268 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetEdgeChildVertex()

\n \n
\n@@ -524,15 +524,15 @@\n inline \n \n \n
\n \n

Return the child vertex (in the next level) of a given edge.

\n \n-

Definition at line 274 of file topologyLevel.h.

\n+

Definition at line 274 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetEdgeFaceLocalIndices()

\n \n
\n@@ -554,15 +554,15 @@\n inline \n \n \n
\n \n

Access the local indices of an edge with respect to its incident faces.

\n \n-

Definition at line 117 of file topologyLevel.h.

\n+

Definition at line 117 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetEdgeFaces()

\n \n
\n@@ -584,15 +584,15 @@\n inline \n \n \n
\n \n

Access the faces incident a given edge.

\n \n-

Definition at line 102 of file topologyLevel.h.

\n+

Definition at line 102 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetEdgeSharpness()

\n \n
\n@@ -614,15 +614,15 @@\n inline \n \n \n
\n \n

Return the sharpness assigned a given edge.

\n \n-

Definition at line 173 of file topologyLevel.h.

\n+

Definition at line 173 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetEdgeVertices()

\n \n
\n@@ -644,15 +644,15 @@\n inline \n \n \n
\n \n

Access the vertices incident a given edge.

\n \n-

Definition at line 99 of file topologyLevel.h.

\n+

Definition at line 99 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetFaceChildEdges()

\n \n
\n@@ -674,15 +674,15 @@\n inline \n \n \n
\n \n

Access the child edges (in the next level) of a given face.

\n \n-

Definition at line 265 of file topologyLevel.h.

\n+

Definition at line 265 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetFaceChildFaces()

\n \n
\n@@ -704,15 +704,15 @@\n inline \n \n \n
\n \n

Access the child faces (in the next level) of a given face.

\n \n-

Definition at line 262 of file topologyLevel.h.

\n+

Definition at line 262 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetFaceChildVertex()

\n \n
\n@@ -734,15 +734,15 @@\n inline \n \n \n
\n \n

Return the child vertex (in the next level) of a given face.

\n \n-

Definition at line 271 of file topologyLevel.h.

\n+

Definition at line 271 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetFaceEdges()

\n \n
\n@@ -764,15 +764,15 @@\n inline \n \n \n
\n \n

Access the edges incident a given face.

\n \n-

Definition at line 96 of file topologyLevel.h.

\n+

Definition at line 96 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetFaceFVarValues()

\n \n
\n@@ -804,15 +804,15 @@\n inline \n \n \n
\n \n

Access the face-varying values associated with a particular face.

\n \n-

Definition at line 237 of file topologyLevel.h.

\n+

Definition at line 237 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetFaceParentFace()

\n \n
\n@@ -834,15 +834,15 @@\n inline \n \n \n
\n \n

Return the parent face (in the previous level) of a given face.

\n \n-

Definition at line 280 of file topologyLevel.h.

\n+

Definition at line 280 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetFaceVertices()

\n \n
\n@@ -864,15 +864,15 @@\n inline \n \n \n
\n \n

Access the vertices incident a given face.

\n \n-

Definition at line 93 of file topologyLevel.h.

\n+

Definition at line 93 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetNumEdges()

\n \n
\n@@ -893,15 +893,15 @@\n inline \n \n \n
\n \n

Return the number of edges in this level.

\n \n-

Definition at line 68 of file topologyLevel.h.

\n+

Definition at line 68 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetNumFaces()

\n \n
\n@@ -922,15 +922,15 @@\n inline \n \n \n
\n \n

Return the number of faces in this level.

\n \n-

Definition at line 65 of file topologyLevel.h.

\n+

Definition at line 65 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetNumFaceVertices()

\n \n
\n@@ -951,15 +951,15 @@\n inline \n \n \n
\n \n

Return the total number of face-vertices, i.e. the sum of all vertices for all faces.

\n \n-

Definition at line 71 of file topologyLevel.h.

\n+

Definition at line 71 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetNumFVarChannels()

\n \n
\n@@ -980,15 +980,15 @@\n inline \n \n \n
\n \n

Return the number of face-varying channels (should be same for all levels)

\n \n-

Definition at line 230 of file topologyLevel.h.

\n+

Definition at line 230 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetNumFVarValues()

\n \n
\n@@ -1010,15 +1010,15 @@\n inline \n \n \n
\n \n

Return the total number of face-varying values in a particular channel (the upper bound of a face-varying value index)

\n \n-

Definition at line 234 of file topologyLevel.h.

\n+

Definition at line 234 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetNumVertices()

\n \n
\n@@ -1039,15 +1039,15 @@\n inline \n \n \n
\n \n

Return the number of vertices in this level.

\n \n-

Definition at line 62 of file topologyLevel.h.

\n+

Definition at line 62 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetVertexChildVertex()

\n \n
\n@@ -1069,15 +1069,15 @@\n inline \n \n \n
\n \n

Return the child vertex (in the next level) of a given vertex.

\n \n-

Definition at line 277 of file topologyLevel.h.

\n+

Definition at line 277 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetVertexEdgeLocalIndices()

\n \n
\n@@ -1099,15 +1099,15 @@\n inline \n \n \n
\n \n

Access the local indices of a vertex with respect to its incident edges.

\n \n-

Definition at line 114 of file topologyLevel.h.

\n+

Definition at line 114 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetVertexEdges()

\n \n
\n@@ -1129,15 +1129,15 @@\n inline \n \n \n
\n \n

Access the edges incident a given vertex.

\n \n-

Definition at line 108 of file topologyLevel.h.

\n+

Definition at line 108 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetVertexFaceLocalIndices()

\n \n
\n@@ -1159,15 +1159,15 @@\n inline \n \n \n
\n \n

Access the local indices of a vertex with respect to its incident faces.

\n \n-

Definition at line 111 of file topologyLevel.h.

\n+

Definition at line 111 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetVertexFaces()

\n \n
\n@@ -1189,15 +1189,15 @@\n inline \n \n \n
\n \n

Access the faces incident a given vertex.

\n \n-

Definition at line 105 of file topologyLevel.h.

\n+

Definition at line 105 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetVertexRule()

\n \n
\n@@ -1219,15 +1219,15 @@\n inline \n \n \n
\n \n

Return the subdivision rule assigned a given vertex specific to this level.

\n \n-

Definition at line 194 of file topologyLevel.h.

\n+

Definition at line 194 of file topologyLevel.h.

\n \n
\n \n \n

◆ GetVertexSharpness()

\n \n
\n@@ -1249,15 +1249,15 @@\n inline \n \n \n
\n \n

Return the sharpness assigned a given vertex.

\n \n-

Definition at line 176 of file topologyLevel.h.

\n+

Definition at line 176 of file topologyLevel.h.

\n \n
\n \n \n

◆ IsEdgeBoundary()

\n \n
\n@@ -1279,15 +1279,15 @@\n inline \n \n \n
\n \n

Return if the edge is a boundary (only one incident face)

\n \n-

Definition at line 134 of file topologyLevel.h.

\n+

Definition at line 134 of file topologyLevel.h.

\n \n
\n \n \n

◆ IsEdgeInfSharp()

\n \n
\n@@ -1309,15 +1309,15 @@\n inline \n \n \n
\n \n

Return if the edge is infinitely-sharp.

\n \n-

Definition at line 179 of file topologyLevel.h.

\n+

Definition at line 179 of file topologyLevel.h.

\n \n
\n \n \n

◆ IsEdgeNonManifold()

\n \n
\n@@ -1339,15 +1339,15 @@\n inline \n \n \n
\n \n

Return if the edge is non-manifold.

\n \n-

Definition at line 128 of file topologyLevel.h.

\n+

Definition at line 128 of file topologyLevel.h.

\n \n
\n \n \n

◆ IsEdgeSemiSharp()

\n \n
\n@@ -1369,15 +1369,15 @@\n inline \n \n \n
\n \n

Return if the edge is semi-sharp.

\n \n-

Definition at line 185 of file topologyLevel.h.

\n+

Definition at line 185 of file topologyLevel.h.

\n \n
\n \n \n

◆ IsFaceHole()

\n \n
\n@@ -1399,15 +1399,15 @@\n inline \n \n \n
\n \n

Return if a given face has been tagged as a hole.

\n \n-

Definition at line 191 of file topologyLevel.h.

\n+

Definition at line 191 of file topologyLevel.h.

\n \n
\n \n \n

◆ IsVertexBoundary()

\n \n
\n@@ -1429,15 +1429,15 @@\n inline \n \n \n
\n \n

Return if the vertex is on a boundary (at least one incident boundary edge)

\n \n-

Definition at line 137 of file topologyLevel.h.

\n+

Definition at line 137 of file topologyLevel.h.

\n \n
\n \n \n

◆ IsVertexCorner()

\n \n
\n@@ -1459,15 +1459,15 @@\n inline \n \n \n
\n \n

Return if the vertex is a corner (only one incident face)

\n \n-

Definition at line 140 of file topologyLevel.h.

\n+

Definition at line 140 of file topologyLevel.h.

\n \n
\n \n \n

◆ IsVertexInfSharp()

\n \n
\n@@ -1489,15 +1489,15 @@\n inline \n \n \n
\n \n

Return if the vertex is infinitely-sharp.

\n \n-

Definition at line 182 of file topologyLevel.h.

\n+

Definition at line 182 of file topologyLevel.h.

\n \n
\n \n \n

◆ IsVertexNonManifold()

\n \n
\n@@ -1519,15 +1519,15 @@\n inline \n \n \n
\n \n

Return if the vertex is non-manifold.

\n \n-

Definition at line 131 of file topologyLevel.h.

\n+

Definition at line 131 of file topologyLevel.h.

\n \n
\n \n \n

◆ IsVertexSemiSharp()

\n \n
\n@@ -1549,15 +1549,15 @@\n inline \n \n \n
\n \n

Return if the vertex is semi-sharp.

\n \n-

Definition at line 188 of file topologyLevel.h.

\n+

Definition at line 188 of file topologyLevel.h.

\n \n
\n \n \n

◆ IsVertexValenceRegular()

\n \n
\n@@ -1582,15 +1582,15 @@\n
\n \n

Return if the valence of the vertex is regular (must be manifold)

\n

Note that this test only determines if the valence of the vertex is regular with respect to the assigned subdivision scheme – not if the neighborhood around the vertex is regular. The latter depends on a number of factors including the incident faces of the vertex (they must all be regular) and the presence of sharpness at the vertex itself or its incident edges.

\n

The regularity of the valence is a necessary but not a sufficient condition in determining the regularity of the neighborhood. For example, while the valence of an interior vertex may be regular, its neighborhood is not if the vertex was made infinitely sharp. Conversely, a corner vertex is considered regular by its valence but its neighborhood is not if the vertex was not made infinitely sharp.

\n

Whether the valence of the vertex is regular is also a property that remains the same for the vertex in all subdivision levels. In contrast, the regularity of the region around the vertex may change as the presence of irregular faces or semi-sharp features is reduced by subdivision.

\n \n-

Definition at line 162 of file topologyLevel.h.

\n+

Definition at line 162 of file topologyLevel.h.

\n \n
\n \n \n

◆ PrintTopology()

\n \n
\n@@ -1610,15 +1610,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 287 of file topologyLevel.h.

\n+

Definition at line 287 of file topologyLevel.h.

\n \n
\n \n \n

◆ ValidateTopology()

\n \n
\n@@ -1637,15 +1637,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 286 of file topologyLevel.h.

\n+

Definition at line 286 of file topologyLevel.h.

\n \n
\n \n

Friends And Related Function Documentation

\n \n

◆ TopologyRefiner

\n \n@@ -1662,20 +1662,20 @@\n \n \n friend \n \n \n
\n \n-

Definition at line 292 of file topologyLevel.h.

\n+

Definition at line 292 of file topologyLevel.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Far
  • TopologyLevel
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01133.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01133.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    BufferDescriptor Struct Reference
    \n
\n
\n \n

BufferDescriptor is a struct which describes buffer elements in interleaved data buffers. Almost all Osd Evaluator APIs take BufferDescriptors along with device-specific buffer objects. \n More...

\n \n-

#include <bufferDescriptor.h>

\n+

#include <bufferDescriptor.h>

\n \n \n \n \n \n \n@@ -136,15 +136,15 @@\n

Detailed Description

\n

BufferDescriptor is a struct which describes buffer elements in interleaved data buffers. Almost all Osd Evaluator APIs take BufferDescriptors along with device-specific buffer objects.

\n

The offset of BufferDescriptor can also be used to express a batching offset if the data buffer is combined across multiple objects together.

\n
    \n
  • Note that each element has the same data type (float)
    • \n
\n \n-

Definition at line 61 of file bufferDescriptor.h.

\n+

Definition at line 61 of file bufferDescriptor.h.

\n

Constructor & Destructor Documentation

\n \n

◆ BufferDescriptor() [1/2]

\n \n
\n
\n

\n Public Member Functions

 BufferDescriptor ()
 Default Constructor. More...
 
 BufferDescriptor (int o, int l, int s)
\n@@ -163,15 +163,15 @@\n inline\n \n
\n
\n \n

Default Constructor.

\n \n-

Definition at line 64 of file bufferDescriptor.h.

\n+

Definition at line 64 of file bufferDescriptor.h.

\n \n
\n \n \n

◆ BufferDescriptor() [2/2]

\n \n
\n@@ -209,15 +209,15 @@\n inline \n \n \n
\n \n

Constructor.

\n \n-

Definition at line 67 of file bufferDescriptor.h.

\n+

Definition at line 67 of file bufferDescriptor.h.

\n \n
\n \n

Member Function Documentation

\n \n

◆ GetLocalOffset()

\n \n@@ -239,15 +239,15 @@\n inline \n \n \n
\n \n

Returns the relative offset within a stride.

\n \n-

Definition at line 70 of file bufferDescriptor.h.

\n+

Definition at line 70 of file bufferDescriptor.h.

\n \n
\n \n \n

◆ IsValid()

\n \n
\n@@ -268,15 +268,15 @@\n inline \n \n \n
\n \n

True if the descriptor values are internally consistent.

\n \n-

Definition at line 75 of file bufferDescriptor.h.

\n+

Definition at line 75 of file bufferDescriptor.h.

\n \n
\n \n \n

◆ operator!=()

\n \n
\n@@ -298,15 +298,15 @@\n inline \n \n \n
\n \n

True if the descriptors are not identical.

\n \n-

Definition at line 93 of file bufferDescriptor.h.

\n+

Definition at line 93 of file bufferDescriptor.h.

\n \n
\n \n \n

◆ operator==()

\n \n
\n@@ -328,15 +328,15 @@\n inline \n \n \n
\n \n

True if the descriptors are identical.

\n \n-

Definition at line 86 of file bufferDescriptor.h.

\n+

Definition at line 86 of file bufferDescriptor.h.

\n \n
\n \n \n

◆ Reset()

\n \n
\n@@ -357,15 +357,15 @@\n inline \n \n \n
\n \n

Resets the descriptor to default.

\n \n-

Definition at line 81 of file bufferDescriptor.h.

\n+

Definition at line 81 of file bufferDescriptor.h.

\n \n
\n \n

Member Data Documentation

\n \n

◆ length

\n \n@@ -376,15 +376,15 @@\n int length\n \n \n
\n \n

number or length of the data

\n \n-

Definition at line 100 of file bufferDescriptor.h.

\n+

Definition at line 100 of file bufferDescriptor.h.

\n \n
\n \n \n

◆ offset

\n \n
\n@@ -394,15 +394,15 @@\n int offset\n \n \n
\n \n

offset to desired element data

\n \n-

Definition at line 98 of file bufferDescriptor.h.

\n+

Definition at line 98 of file bufferDescriptor.h.

\n \n
\n \n \n

◆ stride

\n \n
\n@@ -412,20 +412,20 @@\n int stride\n \n \n
\n \n

stride to the next element

\n \n-

Definition at line 102 of file bufferDescriptor.h.

\n+

Definition at line 102 of file bufferDescriptor.h.

\n \n
\n \n
The documentation for this struct was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • BufferDescriptor
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01137.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01137.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    CLStencilTable Class Reference
    \n
\n
\n \n

OpenCL stencil table. \n More...

\n \n-

#include <clEvaluator.h>

\n+

#include <clEvaluator.h>

\n \n \n \n \n \n \n@@ -136,15 +136,15 @@\n \n

\n Public Member Functions

 CLStencilTable (Far::StencilTable const *stencilTable, cl_context clContext)
 
 CLStencilTable (Far::LimitStencilTable const *limitStencilTable, cl_context clContext)
 
 
\n

Detailed Description

\n

OpenCL stencil table.

\n

This class is an OpenCL buffer representation of Far::StencilTable.

\n

CLCompute consumes this table to apply stencils

\n \n-

Definition at line 52 of file clEvaluator.h.

\n+

Definition at line 52 of file clEvaluator.h.

\n

Constructor & Destructor Documentation

\n \n

◆ CLStencilTable() [1/2]

\n \n
\n
\n \n@@ -246,15 +246,15 @@\n \n \n \n
\n inlinestatic
\n
\n \n-

Definition at line 61 of file clEvaluator.h.

\n+

Definition at line 61 of file clEvaluator.h.

\n \n
\n
\n \n

◆ Create() [2/2]

\n \n
\n@@ -284,15 +284,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 55 of file clEvaluator.h.

\n+

Definition at line 55 of file clEvaluator.h.

\n \n
\n
\n \n

◆ GetDuuWeightsBuffer()

\n \n
\n@@ -311,15 +311,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 80 of file clEvaluator.h.

\n+

Definition at line 80 of file clEvaluator.h.

\n \n
\n \n \n

◆ GetDuvWeightsBuffer()

\n \n
\n@@ -338,15 +338,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 81 of file clEvaluator.h.

\n+

Definition at line 81 of file clEvaluator.h.

\n \n
\n \n \n

◆ GetDuWeightsBuffer()

\n \n
\n@@ -365,15 +365,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 78 of file clEvaluator.h.

\n+

Definition at line 78 of file clEvaluator.h.

\n \n
\n \n \n

◆ GetDvvWeightsBuffer()

\n \n
\n@@ -392,15 +392,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 82 of file clEvaluator.h.

\n+

Definition at line 82 of file clEvaluator.h.

\n \n
\n \n \n

◆ GetDvWeightsBuffer()

\n \n
\n@@ -419,15 +419,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 79 of file clEvaluator.h.

\n+

Definition at line 79 of file clEvaluator.h.

\n \n
\n \n \n

◆ GetIndicesBuffer()

\n \n
\n@@ -446,15 +446,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 76 of file clEvaluator.h.

\n+

Definition at line 76 of file clEvaluator.h.

\n \n
\n \n \n

◆ GetNumStencils()

\n \n
\n@@ -473,15 +473,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 83 of file clEvaluator.h.

\n+

Definition at line 83 of file clEvaluator.h.

\n \n
\n \n \n

◆ GetOffsetsBuffer()

\n \n
\n@@ -500,15 +500,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 75 of file clEvaluator.h.

\n+

Definition at line 75 of file clEvaluator.h.

\n \n
\n \n \n

◆ GetSizesBuffer()

\n \n
\n@@ -527,15 +527,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 74 of file clEvaluator.h.

\n+

Definition at line 74 of file clEvaluator.h.

\n \n
\n \n \n

◆ GetWeightsBuffer()

\n \n
\n@@ -554,20 +554,20 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 77 of file clEvaluator.h.

\n+

Definition at line 77 of file clEvaluator.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
CLEvaluator Class Reference
\n \n
\n \n-

#include <clEvaluator.h>

\n+

#include <clEvaluator.h>

\n \n \n \n \n

\n Public Types

typedef bool Instantiatable
 
\n \n \n \n \n

\n@@ -238,29 +238,29 @@\n

 Wait the OpenCL kernels finish. More...
 
static void Synchronize (cl_command_queue queue)
 
\n

Detailed Description

\n
\n-

Definition at line 100 of file clEvaluator.h.

\n+

Definition at line 100 of file clEvaluator.h.

\n

Member Typedef Documentation

\n \n

◆ Instantiatable

\n \n
\n
\n \n \n \n \n
typedef bool Instantiatable
\n
\n \n-

Definition at line 102 of file clEvaluator.h.

\n+

Definition at line 102 of file clEvaluator.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ CLEvaluator()

\n \n@@ -445,15 +445,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 145 of file clEvaluator.h.

\n+

Definition at line 145 of file clEvaluator.h.

\n \n
\n \n \n

◆ Create() [2/4]

\n \n
\n@@ -521,15 +521,15 @@\n inlinestatic \n \n \n
\n \n

Generic creator template.

\n \n-

Definition at line 131 of file clEvaluator.h.

\n+

Definition at line 131 of file clEvaluator.h.

\n \n
\n \n \n

◆ Create() [3/4]

\n \n
\n@@ -583,15 +583,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 116 of file clEvaluator.h.

\n+

Definition at line 116 of file clEvaluator.h.

\n \n
\n \n \n

◆ Create() [4/4]

\n \n
\n@@ -641,15 +641,15 @@\n inlinestatic \n \n \n
\n \n

Generic creator template.

\n \n-

Definition at line 106 of file clEvaluator.h.

\n+

Definition at line 106 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [1/8]

\n \n
\n@@ -1088,15 +1088,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 1125 of file clEvaluator.h.

\n+

Definition at line 1125 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [4/8]

\n \n
\n@@ -1261,15 +1261,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 1379 of file clEvaluator.h.

\n+

Definition at line 1379 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [5/8]

\n \n
\n@@ -1406,15 +1406,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 1003 of file clEvaluator.h.

\n+

Definition at line 1003 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [6/8]

\n \n
\n@@ -1537,15 +1537,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 1288 of file clEvaluator.h.

\n+

Definition at line 1288 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [7/8]

\n \n
\n@@ -1657,15 +1657,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 902 of file clEvaluator.h.

\n+

Definition at line 902 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [8/8]

\n \n
\n@@ -1760,15 +1760,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 1217 of file clEvaluator.h.

\n+

Definition at line 1217 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [1/6]

\n \n
\n@@ -1954,15 +1954,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 2437 of file clEvaluator.h.

\n+

Definition at line 2437 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [2/6]

\n \n
\n@@ -2134,15 +2134,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 2564 of file clEvaluator.h.

\n+

Definition at line 2564 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [3/6]

\n \n
\n@@ -2286,15 +2286,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 2231 of file clEvaluator.h.

\n+

Definition at line 2231 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [4/6]

\n \n
\n@@ -2424,15 +2424,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 2329 of file clEvaluator.h.

\n+

Definition at line 2329 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [5/6]

\n \n
\n@@ -2548,15 +2548,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 2061 of file clEvaluator.h.

\n+

Definition at line 2061 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [6/6]

\n \n
\n@@ -2658,15 +2658,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 2143 of file clEvaluator.h.

\n+

Definition at line 2143 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [1/6]

\n \n
\n@@ -2845,15 +2845,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 1854 of file clEvaluator.h.

\n+

Definition at line 1854 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [2/6]

\n \n
\n@@ -3018,15 +3018,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 1978 of file clEvaluator.h.

\n+

Definition at line 1978 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [3/6]

\n \n
\n@@ -3163,15 +3163,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 1653 of file clEvaluator.h.

\n+

Definition at line 1653 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [4/6]

\n \n
\n@@ -3294,15 +3294,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 1749 of file clEvaluator.h.

\n+

Definition at line 1749 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [5/6]

\n \n
\n@@ -3411,15 +3411,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 1489 of file clEvaluator.h.

\n+

Definition at line 1489 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [6/6]

\n \n
\n@@ -3514,15 +3514,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 1568 of file clEvaluator.h.

\n+

Definition at line 1568 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [1/9]

\n \n
\n@@ -4157,15 +4157,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 421 of file clEvaluator.h.

\n+

Definition at line 421 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [5/9]

\n \n
\n@@ -4316,15 +4316,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 646 of file clEvaluator.h.

\n+

Definition at line 646 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [6/9]

\n \n
\n@@ -4447,15 +4447,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 309 of file clEvaluator.h.

\n+

Definition at line 309 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [7/9]

\n \n
\n@@ -4564,15 +4564,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 561 of file clEvaluator.h.

\n+

Definition at line 561 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [8/9]

\n \n
\n@@ -4670,15 +4670,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 219 of file clEvaluator.h.

\n+

Definition at line 219 of file clEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [9/9]

\n \n
\n@@ -4759,15 +4759,15 @@\n numStartEventsthe number of events in the array pointed to by startEvents.\n startEventspoints to an array of cl_event which will determine when it is safe for the OpenCL device to begin work or NULL if it can begin immediately.\n endEventpointer to a cl_event which will receive a copy of the cl_event which indicates when all work for this call has completed. This cl_event has an incremented reference count and should be released via clReleaseEvent(). NULL if not required. \n \n \n \n \n-

Definition at line 500 of file clEvaluator.h.

\n+

Definition at line 500 of file clEvaluator.h.

\n \n
\n \n \n

◆ Synchronize() [1/2]

\n \n
\n@@ -4815,20 +4815,20 @@\n inlinestatic \n \n \n
\n \n

Wait the OpenCL kernels finish.

\n \n-

Definition at line 2613 of file clEvaluator.h.

\n+

Definition at line 2613 of file clEvaluator.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • CLEvaluator
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01145.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01145.html", "unified_diff": "@@ -93,15 +93,15 @@\n
    CLPatchTable Class Reference
    \n
\n
\n \n

CL patch table. \n More...

\n \n-

#include <clPatchTable.h>

\n+

#include <clPatchTable.h>

\n
\n Inheritance diagram for CLPatchTable:
\n
\n
\n \"\"/\n
\n \n@@ -174,15 +174,15 @@\n \n
 
\n

Detailed Description

\n

CL patch table.

\n

This class is a CL buffer representation of Far::PatchTable.

\n

CLEvaluator consumes this table to evaluate on the patches.

\n \n-

Definition at line 52 of file clPatchTable.h.

\n+

Definition at line 52 of file clPatchTable.h.

\n

Constructor & Destructor Documentation

\n \n

◆ ~CLPatchTable()

\n \n
\n
\n \n@@ -329,15 +329,15 @@\n \n \n \n
\n inlinestatic
\n
\n \n-

Definition at line 59 of file clPatchTable.h.

\n+

Definition at line 59 of file clPatchTable.h.

\n \n
\n
\n \n

◆ GetFVarPatchArrayBuffer()

\n \n
\n@@ -359,15 +359,15 @@\n inline \n \n \n
\n \n

Returns the CL memory of the array of Osd::PatchArray buffer.

\n \n-

Definition at line 86 of file clPatchTable.h.

\n+

Definition at line 86 of file clPatchTable.h.

\n \n
\n
\n \n

◆ GetFVarPatchIndexBuffer()

\n \n
\n@@ -389,15 +389,15 @@\n inline \n \n \n
\n \n

Returns the CL memory of the face-varying control vertices.

\n \n-

Definition at line 89 of file clPatchTable.h.

\n+

Definition at line 89 of file clPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchParamBuffer()

\n \n
\n@@ -419,15 +419,15 @@\n inline \n \n \n
\n \n

Returns the CL memory of the array of Osd::PatchParam buffer.

\n \n-

Definition at line 92 of file clPatchTable.h.

\n+

Definition at line 92 of file clPatchTable.h.

\n \n
\n \n \n

◆ GetNumFVarChannels()

\n \n
\n@@ -448,15 +448,15 @@\n inline \n \n \n
\n \n

Returns the number of face-varying channel buffers.

\n \n-

Definition at line 83 of file clPatchTable.h.

\n+

Definition at line 83 of file clPatchTable.h.

\n \n
\n \n \n

◆ GetPatchArrayBuffer()

\n \n
\n@@ -477,15 +477,15 @@\n inline \n \n \n
\n \n

Returns the CL memory of the array of Osd::PatchArray buffer.

\n \n-

Definition at line 68 of file clPatchTable.h.

\n+

Definition at line 68 of file clPatchTable.h.

\n \n
\n \n \n

◆ GetPatchIndexBuffer()

\n \n
\n@@ -506,15 +506,15 @@\n inline \n \n \n
\n \n

Returns the CL memory of the patch control vertices.

\n \n-

Definition at line 71 of file clPatchTable.h.

\n+

Definition at line 71 of file clPatchTable.h.

\n \n
\n \n \n

◆ GetPatchParamBuffer()

\n \n
\n@@ -535,15 +535,15 @@\n inline \n \n \n
\n \n

Returns the CL memory of the array of Osd::PatchParam buffer.

\n \n-

Definition at line 74 of file clPatchTable.h.

\n+

Definition at line 74 of file clPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchArrayBuffer()

\n \n
\n@@ -564,15 +564,15 @@\n inline \n \n \n
\n \n

Returns the CL memory of the array of Osd::PatchArray buffer.

\n \n-

Definition at line 77 of file clPatchTable.h.

\n+

Definition at line 77 of file clPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchIndexBuffer()

\n \n
\n@@ -593,15 +593,15 @@\n inline \n \n \n
\n \n

Returns the CL memory of the varying control vertices.

\n \n-

Definition at line 80 of file clPatchTable.h.

\n+

Definition at line 80 of file clPatchTable.h.

\n \n
\n \n

Member Data Documentation

\n \n

◆ _fvarIndexBuffers

\n \n@@ -618,15 +618,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 107 of file clPatchTable.h.

\n+

Definition at line 107 of file clPatchTable.h.

\n \n
\n \n \n

◆ _fvarParamBuffers

\n \n
\n@@ -642,15 +642,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 108 of file clPatchTable.h.

\n+

Definition at line 108 of file clPatchTable.h.

\n \n
\n \n \n

◆ _fvarPatchArrays

\n \n
\n@@ -666,15 +666,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 106 of file clPatchTable.h.

\n+

Definition at line 106 of file clPatchTable.h.

\n \n
\n \n \n

◆ _indexBuffer

\n \n
\n@@ -690,15 +690,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 100 of file clPatchTable.h.

\n+

Definition at line 100 of file clPatchTable.h.

\n \n
\n \n \n

◆ _patchArrays

\n \n
\n@@ -714,15 +714,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 99 of file clPatchTable.h.

\n+

Definition at line 99 of file clPatchTable.h.

\n \n
\n \n \n

◆ _patchParamBuffer

\n \n
\n@@ -738,15 +738,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 101 of file clPatchTable.h.

\n+

Definition at line 101 of file clPatchTable.h.

\n \n
\n \n \n

◆ _varyingIndexBuffer

\n \n
\n@@ -762,15 +762,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 104 of file clPatchTable.h.

\n+

Definition at line 104 of file clPatchTable.h.

\n \n
\n \n \n

◆ _varyingPatchArrays

\n \n
\n@@ -786,20 +786,20 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 103 of file clPatchTable.h.

\n+

Definition at line 103 of file clPatchTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • CLPatchTable
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01153.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01153.html", "unified_diff": "@@ -92,15 +92,15 @@\n
    CpuD3D11VertexBuffer Class Reference
    \n
\n
\n \n

Concrete vertex buffer class for Cpu subdivision and DirectX drawing. \n More...

\n \n-

#include <cpuD3D11VertexBuffer.h>

\n+

#include <cpuD3D11VertexBuffer.h>

\n \n \n \n \n \n \n@@ -136,15 +136,15 @@\n \n

\n Public Member Functions

virtual ~CpuD3D11VertexBuffer ()
 Destructor. More...
 
void UpdateData (const float *src, int startVertex, int numVertices, void *deviceContext=NULL)
 
\n

Detailed Description

\n

Concrete vertex buffer class for Cpu subdivision and DirectX drawing.

\n

CpuD3D11VertexBuffer implements CpuVertexBufferInterface and D3D11VertexBufferInterface.

\n

An instance of this buffer class can be passed to CpuEvaluator.

\n \n-

Definition at line 48 of file cpuD3D11VertexBuffer.h.

\n+

Definition at line 48 of file cpuD3D11VertexBuffer.h.

\n

Constructor & Destructor Documentation

\n \n

◆ ~CpuD3D11VertexBuffer()

\n \n
\n
\n \n@@ -295,15 +295,15 @@\n inline\n \n
\n
\n \n

Returns the D3D11 buffer object (for Osd::Mesh interface)

\n \n-

Definition at line 75 of file cpuD3D11VertexBuffer.h.

\n+

Definition at line 75 of file cpuD3D11VertexBuffer.h.

\n \n
\n
\n \n

◆ Create()

\n \n
\n@@ -423,15 +423,15 @@\n \n
\n

This method is meant to be used in client code in order to provide coarse vertices data to Osd.

\n \n
\n
\n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
\n \n-

#include <cpuEvaluator.h>

\n+

#include <cpuEvaluator.h>

\n \n \n \n \n \n \n@@ -163,15 +163,15 @@\n \n \n \n \n

\n Static Public Member Functions

template<typename SRC_BUFFER , typename DST_BUFFER , typename STENCIL_TABLE >
static bool EvalStencils (SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, const CpuEvaluator *instance=NULL, void *deviceContext=NULL)
 Generic static eval stencils function. This function has a same signature as other device kernels have so that it can be called in the same way from OsdMesh template interface. More...
 
 
static void Synchronize (void *)
 synchronize all asynchronous computation invoked on this device. More...
 
\n

Detailed Description

\n
\n-

Definition at line 39 of file cpuEvaluator.h.

\n+

Definition at line 39 of file cpuEvaluator.h.

\n

Member Function Documentation

\n \n

◆ EvalPatches() [1/6]

\n \n
\n
\n \n@@ -707,15 +707,15 @@\n \n \n \n
patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch
instancenot used in the cpu evaluator
deviceContextnot used in the cpu evaluator
\n \n \n \n-

Definition at line 607 of file cpuEvaluator.h.

\n+

Definition at line 607 of file cpuEvaluator.h.

\n \n
\n
\n \n

◆ EvalPatches() [5/6]

\n \n
\n@@ -831,15 +831,15 @@\n patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 517 of file cpuEvaluator.h.

\n+

Definition at line 517 of file cpuEvaluator.h.

\n \n
\n
\n \n

◆ EvalPatches() [6/6]

\n \n
\n@@ -930,15 +930,15 @@\n patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 454 of file cpuEvaluator.h.

\n+

Definition at line 454 of file cpuEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [1/3]

\n \n
\n@@ -1103,15 +1103,15 @@\n fvarChannelface-varying channel\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 1188 of file cpuEvaluator.h.

\n+

Definition at line 1188 of file cpuEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [2/3]

\n \n
\n@@ -1234,15 +1234,15 @@\n fvarChannelface-varying channel\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 1100 of file cpuEvaluator.h.

\n+

Definition at line 1100 of file cpuEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [3/3]

\n \n
\n@@ -1337,15 +1337,15 @@\n fvarChannelface-varying channel\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 1034 of file cpuEvaluator.h.

\n+

Definition at line 1034 of file cpuEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [1/3]

\n \n
\n@@ -1503,15 +1503,15 @@\n patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 971 of file cpuEvaluator.h.

\n+

Definition at line 971 of file cpuEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [2/3]

\n \n
\n@@ -1627,15 +1627,15 @@\n patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 886 of file cpuEvaluator.h.

\n+

Definition at line 886 of file cpuEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [3/3]

\n \n
\n@@ -1723,15 +1723,15 @@\n patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 823 of file cpuEvaluator.h.

\n+

Definition at line 823 of file cpuEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [1/6]

\n \n
\n@@ -2325,15 +2325,15 @@\n stencilTableFar::StencilTable or equivalent\n instancenot used in the cpu kernel (declared as a typed pointer to prevent undesirable template resolution)\n deviceContextnot used in the cpu kernel \n \n \n \n \n-

Definition at line 303 of file cpuEvaluator.h.

\n+

Definition at line 303 of file cpuEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [5/6]

\n \n
\n@@ -2435,15 +2435,15 @@\n stencilTableFar::StencilTable or equivalent\n instancenot used in the cpu kernel (declared as a typed pointer to prevent undesirable template resolution)\n deviceContextnot used in the cpu kernel \n \n \n \n \n-

Definition at line 168 of file cpuEvaluator.h.

\n+

Definition at line 168 of file cpuEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [6/6]

\n \n
\n@@ -2520,15 +2520,15 @@\n stencilTableFar::StencilTable or equivalent\n instancenot used in the cpu kernel (declared as a typed pointer to prevent undesirable template resolution)\n deviceContextnot used in the cpu kernel \n \n \n \n \n-

Definition at line 72 of file cpuEvaluator.h.

\n+

Definition at line 72 of file cpuEvaluator.h.

\n \n
\n \n \n

◆ Synchronize()

\n \n
\n@@ -2553,20 +2553,20 @@\n
\n \n

synchronize all asynchronous computation invoked on this device.

\n
\n

Other methods

\n
\n \n-

Definition at line 1227 of file cpuEvaluator.h.

\n+

Definition at line 1227 of file cpuEvaluator.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • CpuEvaluator
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01165.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01165.html", "unified_diff": "@@ -92,15 +92,15 @@\n
    CpuPatchTable Class Reference
    \n
\n
\n \n

Cpu patch table. \n More...

\n \n-

#include <cpuPatchTable.h>

\n+

#include <cpuPatchTable.h>

\n \n \n \n \n \n \n@@ -161,15 +161,15 @@\n

\n Public Member Functions

 CpuPatchTable (const Far::PatchTable *patchTable)
 
 ~CpuPatchTable ()
 
\n

Detailed Description

\n

Cpu patch table.

\n

XXX: We can use just Far::PatchTable for typical CpuEval use cases.

\n

Currently this class exists because of the template resolution for the CpuEvaluator's generic interface functions (glEvalLimit example uses), and device-specific patch tables such as GLPatchTables internally use as a staging buffer to splice patcharray and interleave sharpnesses.

\n

Ideally Far::PatchTables should have the same data representation and accessors so that we don't have to copy data unnecessarily.

\n \n-

Definition at line 59 of file cpuPatchTable.h.

\n+

Definition at line 59 of file cpuPatchTable.h.

\n

Constructor & Destructor Documentation

\n \n

◆ CpuPatchTable()

\n \n
\n
\n \n@@ -212,15 +212,15 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 68 of file cpuPatchTable.h.

\n+

Definition at line 68 of file cpuPatchTable.h.

\n \n
\n
\n

Member Function Documentation

\n \n

◆ Create()

\n \n@@ -251,15 +251,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 61 of file cpuPatchTable.h.

\n+

Definition at line 61 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchArrayBuffer()

\n \n
\n@@ -279,15 +279,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 109 of file cpuPatchTable.h.

\n+

Definition at line 109 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchIndexBuffer()

\n \n
\n@@ -307,15 +307,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 112 of file cpuPatchTable.h.

\n+

Definition at line 112 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchIndexSize()

\n \n
\n@@ -335,15 +335,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 115 of file cpuPatchTable.h.

\n+

Definition at line 115 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchParamBuffer()

\n \n
\n@@ -363,15 +363,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 118 of file cpuPatchTable.h.

\n+

Definition at line 118 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchParamSize()

\n \n
\n@@ -391,15 +391,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 121 of file cpuPatchTable.h.

\n+

Definition at line 121 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetNumFVarChannels()

\n \n
\n@@ -418,15 +418,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 106 of file cpuPatchTable.h.

\n+

Definition at line 106 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetNumPatchArrays()

\n \n
\n@@ -445,15 +445,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 80 of file cpuPatchTable.h.

\n+

Definition at line 80 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetPatchArrayBuffer()

\n \n
\n@@ -472,15 +472,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 70 of file cpuPatchTable.h.

\n+

Definition at line 70 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetPatchIndexBuffer()

\n \n
\n@@ -499,15 +499,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 73 of file cpuPatchTable.h.

\n+

Definition at line 73 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetPatchIndexSize()

\n \n
\n@@ -526,15 +526,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 83 of file cpuPatchTable.h.

\n+

Definition at line 83 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetPatchParamBuffer()

\n \n
\n@@ -553,15 +553,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 76 of file cpuPatchTable.h.

\n+

Definition at line 76 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetPatchParamSize()

\n \n
\n@@ -580,15 +580,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 86 of file cpuPatchTable.h.

\n+

Definition at line 86 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchArrayBuffer()

\n \n
\n@@ -607,15 +607,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 90 of file cpuPatchTable.h.

\n+

Definition at line 90 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchIndexBuffer()

\n \n
\n@@ -634,15 +634,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 96 of file cpuPatchTable.h.

\n+

Definition at line 96 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchIndexSize()

\n \n
\n@@ -661,15 +661,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 102 of file cpuPatchTable.h.

\n+

Definition at line 102 of file cpuPatchTable.h.

\n \n
\n \n

Member Data Documentation

\n \n

◆ _fvarIndexBuffers

\n \n@@ -686,15 +686,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 134 of file cpuPatchTable.h.

\n+

Definition at line 134 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ _fvarParamBuffers

\n \n
\n@@ -710,15 +710,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 135 of file cpuPatchTable.h.

\n+

Definition at line 135 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ _fvarPatchArrays

\n \n
\n@@ -734,15 +734,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 133 of file cpuPatchTable.h.

\n+

Definition at line 133 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ _indexBuffer

\n \n
\n@@ -758,15 +758,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 127 of file cpuPatchTable.h.

\n+

Definition at line 127 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ _patchArrays

\n \n
\n@@ -782,15 +782,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 126 of file cpuPatchTable.h.

\n+

Definition at line 126 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ _patchParamBuffer

\n \n
\n@@ -806,15 +806,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 128 of file cpuPatchTable.h.

\n+

Definition at line 128 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ _varyingIndexBuffer

\n \n
\n@@ -830,15 +830,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 131 of file cpuPatchTable.h.

\n+

Definition at line 131 of file cpuPatchTable.h.

\n \n
\n \n \n

◆ _varyingPatchArrays

\n \n
\n@@ -854,20 +854,20 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 130 of file cpuPatchTable.h.

\n+

Definition at line 130 of file cpuPatchTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • CpuPatchTable
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01169.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01169.html", "unified_diff": "@@ -92,15 +92,15 @@\n
    CpuVertexBuffer Class Reference
    \n
\n
\n \n

Concrete vertex buffer class for CPU subdivision. \n More...

\n \n-

#include <cpuVertexBuffer.h>

\n+

#include <cpuVertexBuffer.h>

\n \n \n \n \n \n \n@@ -127,15 +127,15 @@\n \n \n

\n Public Member Functions

 ~CpuVertexBuffer ()
 Destructor. More...
 
void UpdateData (const float *src, int startVertex, int numVertices, void *deviceContext=NULL)
 Constructor. More...
 
\n

Detailed Description

\n

Concrete vertex buffer class for CPU subdivision.

\n

CpuVertexBuffer implements the VertexBufferInterface. An instance of this buffer class can be passed to CpuEvaluator

\n \n-

Definition at line 42 of file cpuVertexBuffer.h.

\n+

Definition at line 42 of file cpuVertexBuffer.h.

\n

Constructor & Destructor Documentation

\n \n

◆ ~CpuVertexBuffer()

\n \n
\n
\n \n@@ -330,15 +330,15 @@\n
\n
\n

This method is meant to be used in client code in order to provide coarse vertices data to Osd.

\n \n
\n
\n
The documentation for this class was generated from the following file:\n
\n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • CpuVertexBuffer
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01173.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01173.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    CudaStencilTable Class Reference
    \n
\n
\n \n

CUDA stencil table. \n More...

\n \n-

#include <cudaEvaluator.h>

\n+

#include <cudaEvaluator.h>

\n \n \n \n \n \n \n@@ -134,15 +134,15 @@\n \n

\n Public Member Functions

 CudaStencilTable (Far::StencilTable const *stencilTable)
 
 CudaStencilTable (Far::LimitStencilTable const *limitStencilTable)
 
 
\n

Detailed Description

\n

CUDA stencil table.

\n

This class is a cuda buffer representation of Far::StencilTable.

\n

CudaEvaluator consumes this table to apply stencils

\n \n-

Definition at line 52 of file cudaEvaluator.h.

\n+

Definition at line 52 of file cudaEvaluator.h.

\n

Constructor & Destructor Documentation

\n \n

◆ CudaStencilTable() [1/2]

\n \n
\n
\n \n@@ -240,15 +240,15 @@\n \n \n \n
\n inlinestatic
\n
\n \n-

Definition at line 59 of file cudaEvaluator.h.

\n+

Definition at line 59 of file cudaEvaluator.h.

\n \n
\n
\n \n

◆ Create() [2/2]

\n \n
\n@@ -278,15 +278,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 54 of file cudaEvaluator.h.

\n+

Definition at line 54 of file cudaEvaluator.h.

\n \n
\n
\n \n

◆ GetDuuWeightsBuffer()

\n \n
\n@@ -305,15 +305,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 76 of file cudaEvaluator.h.

\n+

Definition at line 76 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ GetDuvWeightsBuffer()

\n \n
\n@@ -332,15 +332,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 77 of file cudaEvaluator.h.

\n+

Definition at line 77 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ GetDuWeightsBuffer()

\n \n
\n@@ -359,15 +359,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 74 of file cudaEvaluator.h.

\n+

Definition at line 74 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ GetDvvWeightsBuffer()

\n \n
\n@@ -386,15 +386,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 78 of file cudaEvaluator.h.

\n+

Definition at line 78 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ GetDvWeightsBuffer()

\n \n
\n@@ -413,15 +413,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 75 of file cudaEvaluator.h.

\n+

Definition at line 75 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ GetIndicesBuffer()

\n \n
\n@@ -440,15 +440,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 72 of file cudaEvaluator.h.

\n+

Definition at line 72 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ GetNumStencils()

\n \n
\n@@ -467,15 +467,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 79 of file cudaEvaluator.h.

\n+

Definition at line 79 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ GetOffsetsBuffer()

\n \n
\n@@ -494,15 +494,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 71 of file cudaEvaluator.h.

\n+

Definition at line 71 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ GetSizesBuffer()

\n \n
\n@@ -521,15 +521,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 70 of file cudaEvaluator.h.

\n+

Definition at line 70 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ GetWeightsBuffer()

\n \n
\n@@ -548,20 +548,20 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 73 of file cudaEvaluator.h.

\n+

Definition at line 73 of file cudaEvaluator.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
\n \n-

#include <cudaEvaluator.h>

\n+

#include <cudaEvaluator.h>

\n \n \n \n \n \n \n@@ -162,15 +162,15 @@\n \n \n \n \n

\n Static Public Member Functions

template<typename SRC_BUFFER , typename DST_BUFFER , typename STENCIL_TABLE >
static bool EvalStencils (SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, const void *instance=NULL, void *deviceContext=NULL)
 Generic static compute function. This function has a same signature as other device kernels have so that it can be called transparently from OsdMesh template interface. More...
 
 Generic limit eval function. This function has a same signature as other device kernels have so that it can be called in the same way. More...
 
static void Synchronize (void *deviceContext=NULL)
 
\n

Detailed Description

\n
\n-

Definition at line 94 of file cudaEvaluator.h.

\n+

Definition at line 94 of file cudaEvaluator.h.

\n

Member Function Documentation

\n \n

◆ EvalPatches() [1/6]

\n \n
\n
\n \n@@ -706,15 +706,15 @@\n \n \n \n
patchTableCudaPatchTable or equivalent
instancenot used in the cuda evaluator
deviceContextnot used in the cuda evaluator
\n \n \n \n-

Definition at line 648 of file cudaEvaluator.h.

\n+

Definition at line 648 of file cudaEvaluator.h.

\n \n
\n
\n \n

◆ EvalPatches() [5/6]

\n \n
\n@@ -830,15 +830,15 @@\n patchTableCudaPatchTable or equivalent\n instancenot used in the cuda evaluator\n deviceContextnot used in the cuda evaluator \n \n \n \n \n-

Definition at line 565 of file cudaEvaluator.h.

\n+

Definition at line 565 of file cudaEvaluator.h.

\n \n
\n
\n \n

◆ EvalPatches() [6/6]

\n \n
\n@@ -929,15 +929,15 @@\n patchTableCudaPatchTable or equivalent\n instancenot used in the cuda evaluator\n deviceContextnot used in the cuda evaluator \n \n \n \n \n-

Definition at line 504 of file cudaEvaluator.h.

\n+

Definition at line 504 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [1/3]

\n \n
\n@@ -1102,15 +1102,15 @@\n fvarChannelface-varying channel\n instancenot used in the cuda evaluator\n deviceContextnot used in the cuda evaluator \n \n \n \n \n-

Definition at line 1224 of file cudaEvaluator.h.

\n+

Definition at line 1224 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [2/3]

\n \n
\n@@ -1233,15 +1233,15 @@\n fvarChannelface-varying channel\n instancenot used in the cuda evaluator\n deviceContextnot used in the cuda evaluator \n \n \n \n \n-

Definition at line 1136 of file cudaEvaluator.h.

\n+

Definition at line 1136 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [3/3]

\n \n
\n@@ -1336,15 +1336,15 @@\n fvarChannelface-varying channel\n instancenot used in the cuda evaluator\n deviceContextnot used in the cuda evaluator \n \n \n \n \n-

Definition at line 1070 of file cudaEvaluator.h.

\n+

Definition at line 1070 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [1/3]

\n \n
\n@@ -1502,15 +1502,15 @@\n patchTableCudaPatchTable or equivalent\n instancenot used in the cuda evaluator\n deviceContextnot used in the cuda evaluator \n \n \n \n \n-

Definition at line 1007 of file cudaEvaluator.h.

\n+

Definition at line 1007 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [2/3]

\n \n
\n@@ -1626,15 +1626,15 @@\n patchTableCudaPatchTable or equivalent\n instancenot used in the cuda evaluator\n deviceContextnot used in the cuda evaluator \n \n \n \n \n-

Definition at line 922 of file cudaEvaluator.h.

\n+

Definition at line 922 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [3/3]

\n \n
\n@@ -1722,15 +1722,15 @@\n patchTableCudaPatchTable or equivalent\n instancenot used in the cuda evaluator\n deviceContextnot used in the cuda evaluator \n \n \n \n \n-

Definition at line 859 of file cudaEvaluator.h.

\n+

Definition at line 859 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [1/6]

\n \n
\n@@ -2324,15 +2324,15 @@\n stencilTablestencil table to be applied.\n instancenot used in the cuda kernel (declared as a typed pointer to prevent undesirable template resolution)\n deviceContextnot used in the cuda kernel \n \n \n \n \n-

Definition at line 353 of file cudaEvaluator.h.

\n+

Definition at line 353 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [5/6]

\n \n
\n@@ -2434,15 +2434,15 @@\n stencilTablestencil table to be applied.\n instancenot used in the cuda kernel (declared as a typed pointer to prevent undesirable template resolution)\n deviceContextnot used in the cuda kernel \n \n \n \n \n-

Definition at line 218 of file cudaEvaluator.h.

\n+

Definition at line 218 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [6/6]

\n \n
\n@@ -2519,15 +2519,15 @@\n stencilTablestencil table to be applied. The table must have Cuda memory interfaces.\n instancenot used in the CudaEvaluator\n deviceContextnot used in the CudaEvaluator \n \n \n \n \n-

Definition at line 126 of file cudaEvaluator.h.

\n+

Definition at line 126 of file cudaEvaluator.h.

\n \n
\n \n \n

◆ Synchronize()

\n \n
\n@@ -2553,15 +2553,15 @@\n
\n

Other methods

\n
\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • CudaEvaluator
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01181.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01181.html", "unified_diff": "@@ -93,15 +93,15 @@\n
    CudaPatchTable Class Reference
    \n
\n
\n \n

CUDA patch table. \n More...

\n \n-

#include <cudaPatchTable.h>

\n+

#include <cudaPatchTable.h>

\n
\n Inheritance diagram for CudaPatchTable:
\n
\n
\n \"\"/\n
\n \n@@ -171,15 +171,15 @@\n \n
 
\n

Detailed Description

\n

CUDA patch table.

\n

This class is a cuda buffer representation of Far::PatchTable.

\n

CudaEvaluator consumes this table to evaluate on the patches.

\n \n-

Definition at line 51 of file cudaPatchTable.h.

\n+

Definition at line 51 of file cudaPatchTable.h.

\n

Constructor & Destructor Documentation

\n \n

◆ ~CudaPatchTable()

\n \n
\n
\n \n@@ -308,15 +308,15 @@\n inline\n \n
\n
\n \n

Returns the cuda memory of the array of Osd::PatchArray buffer.

\n \n-

Definition at line 81 of file cudaPatchTable.h.

\n+

Definition at line 81 of file cudaPatchTable.h.

\n \n
\n
\n \n

◆ GetFVarPatchIndexBuffer()

\n \n
\n@@ -338,15 +338,15 @@\n inline \n \n \n
\n \n

Returns the cuda memory of the array of face-varying control vertices.

\n \n-

Definition at line 86 of file cudaPatchTable.h.

\n+

Definition at line 86 of file cudaPatchTable.h.

\n \n
\n
\n \n

◆ GetFVarPatchParamBuffer()

\n \n
\n@@ -368,15 +368,15 @@\n inline \n \n \n
\n \n

Returns the cuda memory of the array of face-varying param.

\n \n-

Definition at line 91 of file cudaPatchTable.h.

\n+

Definition at line 91 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ GetNumFVarChannels()

\n \n
\n@@ -397,15 +397,15 @@\n inline \n \n \n
\n \n

Returns the number of face-varying channels buffers.

\n \n-

Definition at line 78 of file cudaPatchTable.h.

\n+

Definition at line 78 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ GetPatchArrayBuffer()

\n \n
\n@@ -426,15 +426,15 @@\n inline \n \n \n
\n \n

Returns the cuda memory of the array of Osd::PatchArray buffer.

\n \n-

Definition at line 60 of file cudaPatchTable.h.

\n+

Definition at line 60 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ GetPatchIndexBuffer()

\n \n
\n@@ -455,15 +455,15 @@\n inline \n \n \n
\n \n

Returns the cuda memory of the patch control vertices.

\n \n-

Definition at line 63 of file cudaPatchTable.h.

\n+

Definition at line 63 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ GetPatchParamBuffer()

\n \n
\n@@ -484,15 +484,15 @@\n inline \n \n \n
\n \n

Returns the cuda memory of the array of Osd::PatchParam buffer.

\n \n-

Definition at line 66 of file cudaPatchTable.h.

\n+

Definition at line 66 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchArrayBuffer()

\n \n
\n@@ -513,15 +513,15 @@\n inline \n \n \n
\n \n

Returns the cuda memory of the array of Osd::PatchArray buffer.

\n \n-

Definition at line 69 of file cudaPatchTable.h.

\n+

Definition at line 69 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchIndexBuffer()

\n \n
\n@@ -542,15 +542,15 @@\n inline \n \n \n
\n \n

Returns the cuda memory of the array of varying control vertices.

\n \n-

Definition at line 73 of file cudaPatchTable.h.

\n+

Definition at line 73 of file cudaPatchTable.h.

\n \n
\n \n

Member Data Documentation

\n \n

◆ _fvarIndexBuffers

\n \n@@ -567,15 +567,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 108 of file cudaPatchTable.h.

\n+

Definition at line 108 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ _fvarParamBuffers

\n \n
\n@@ -591,15 +591,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 109 of file cudaPatchTable.h.

\n+

Definition at line 109 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ _fvarPatchArrays

\n \n
\n@@ -615,15 +615,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 107 of file cudaPatchTable.h.

\n+

Definition at line 107 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ _indexBuffer

\n \n
\n@@ -639,15 +639,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 101 of file cudaPatchTable.h.

\n+

Definition at line 101 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ _patchArrays

\n \n
\n@@ -663,15 +663,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 100 of file cudaPatchTable.h.

\n+

Definition at line 100 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ _patchParamBuffer

\n \n
\n@@ -687,15 +687,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 102 of file cudaPatchTable.h.

\n+

Definition at line 102 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ _varyingIndexBuffer

\n \n
\n@@ -711,15 +711,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 105 of file cudaPatchTable.h.

\n+

Definition at line 105 of file cudaPatchTable.h.

\n \n
\n \n \n

◆ _varyingPatchArrays

\n \n
\n@@ -735,20 +735,20 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 104 of file cudaPatchTable.h.

\n+

Definition at line 104 of file cudaPatchTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • CudaPatchTable
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01189.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01189.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    D3D11StencilTable Class Reference
    \n
\n
\n \n

D3D11 stencil table. \n More...

\n \n-

#include <d3d11ComputeEvaluator.h>

\n+

#include <d3d11ComputeEvaluator.h>

\n \n \n \n \n \n \n@@ -123,15 +123,15 @@\n \n

\n Public Member Functions

 D3D11StencilTable (Far::StencilTable const *stencilTable, ID3D11DeviceContext *deviceContext)
 
 ~D3D11StencilTable ()
 
 
\n

Detailed Description

\n

D3D11 stencil table.

\n

This class is a D3D11 Shader Resource View representation of Far::StencilTable.

\n

D3D11ComputeEvaluator consumes this table to apply stencils

\n \n-

Definition at line 56 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 56 of file d3d11ComputeEvaluator.h.

\n

Constructor & Destructor Documentation

\n \n

◆ D3D11StencilTable()

\n \n
\n
\n \n@@ -205,15 +205,15 @@\n \n \n \n
\n inlinestatic
\n
\n \n-

Definition at line 59 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 59 of file d3d11ComputeEvaluator.h.

\n \n
\n
\n \n

◆ Create() [2/2]

\n \n
\n@@ -243,15 +243,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 64 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 64 of file d3d11ComputeEvaluator.h.

\n \n
\n
\n \n

◆ GetIndicesSRV()

\n \n
\n@@ -270,15 +270,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 77 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 77 of file d3d11ComputeEvaluator.h.

\n \n
\n \n \n

◆ GetNumStencils()

\n \n
\n@@ -297,15 +297,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 79 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 79 of file d3d11ComputeEvaluator.h.

\n \n
\n \n \n

◆ GetOffsetsSRV()

\n \n
\n@@ -324,15 +324,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 76 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 76 of file d3d11ComputeEvaluator.h.

\n \n
\n \n \n

◆ GetSizesSRV()

\n \n
\n@@ -351,15 +351,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 75 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 75 of file d3d11ComputeEvaluator.h.

\n \n
\n \n \n

◆ GetWeightsSRV()

\n \n
\n@@ -378,20 +378,20 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 78 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 78 of file d3d11ComputeEvaluator.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
D3D11ComputeEvaluator Class Reference
\n \n
\n \n-

#include <d3d11ComputeEvaluator.h>

\n+

#include <d3d11ComputeEvaluator.h>

\n \n \n \n \n

\n Public Types

typedef bool Instantiatable
 
\n \n \n \n \n

\n@@ -129,29 +129,29 @@\n

 
static void Synchronize (ID3D11DeviceContext *deviceContext)
 Wait the dispatched kernel finishes. More...
 
\n

Detailed Description

\n
\n-

Definition at line 96 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 96 of file d3d11ComputeEvaluator.h.

\n

Member Typedef Documentation

\n \n

◆ Instantiatable

\n \n
\n
\n \n \n \n \n
typedef bool Instantiatable
\n
\n \n-

Definition at line 98 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 98 of file d3d11ComputeEvaluator.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ D3D11ComputeEvaluator()

\n \n@@ -510,15 +510,15 @@\n stencilTablestencil table to be applied. The table must have SSBO interfaces.\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextID3D11DeviceContext. \n \n \n \n \n-

Definition at line 148 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 148 of file d3d11ComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [3/3]

\n \n
\n@@ -573,15 +573,15 @@\n \n inline \n \n \n
\n

Dispatch the DX compute kernel on GPU asynchronously. returns false if the kernel hasn't been compiled yet.

\n \n-

Definition at line 181 of file d3d11ComputeEvaluator.h.

\n+

Definition at line 181 of file d3d11ComputeEvaluator.h.

\n \n
\n \n \n

◆ Synchronize()

\n \n
\n@@ -606,15 +606,15 @@\n
\n \n

Wait the dispatched kernel finishes.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
D3D11LegacyGregoryPatchTable Class Reference
\n \n
\n \n-

#include <d3d11LegacyGregoryPatchTable.h>

\n+

#include <d3d11LegacyGregoryPatchTable.h>

\n
\n Inheritance diagram for D3D11LegacyGregoryPatchTable:
\n
\n
\n \"\"/\n
\n \n@@ -127,15 +127,15 @@\n \n \n \n

\n Protected Member Functions

 D3D11LegacyGregoryPatchTable ()
 
\n

Detailed Description

\n
\n-

Definition at line 43 of file d3d11LegacyGregoryPatchTable.h.

\n+

Definition at line 43 of file d3d11LegacyGregoryPatchTable.h.

\n

Constructor & Destructor Documentation

\n \n

◆ ~D3D11LegacyGregoryPatchTable()

\n \n
\n
\n \n@@ -206,15 +206,15 @@\n \n \n \n
\n inlinestatic
\n
\n \n-

Definition at line 49 of file d3d11LegacyGregoryPatchTable.h.

\n+

Definition at line 49 of file d3d11LegacyGregoryPatchTable.h.

\n \n
\n
\n \n

◆ Create() [2/2]

\n \n
\n@@ -270,15 +270,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 74 of file d3d11LegacyGregoryPatchTable.h.

\n+

Definition at line 74 of file d3d11LegacyGregoryPatchTable.h.

\n \n
\n
\n \n

◆ GetQuadOffsetsSRV()

\n \n
\n@@ -297,15 +297,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 70 of file d3d11LegacyGregoryPatchTable.h.

\n+

Definition at line 70 of file d3d11LegacyGregoryPatchTable.h.

\n \n
\n \n \n

◆ GetVertexSRV()

\n \n
\n@@ -324,15 +324,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 62 of file d3d11LegacyGregoryPatchTable.h.

\n+

Definition at line 62 of file d3d11LegacyGregoryPatchTable.h.

\n \n
\n \n \n

◆ GetVertexValenceSRV()

\n \n
\n@@ -351,15 +351,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 66 of file d3d11LegacyGregoryPatchTable.h.

\n+

Definition at line 66 of file d3d11LegacyGregoryPatchTable.h.

\n \n
\n \n \n

◆ UpdateVertexBuffer()

\n \n
\n@@ -396,15 +396,15 @@\n \n \n
\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
D3D11PatchTable Class Reference
\n \n
\n \n-

#include <d3d11PatchTable.h>

\n+

#include <d3d11PatchTable.h>

\n
\n Inheritance diagram for D3D11PatchTable:
\n
\n
\n \"\"/\n
\n \n@@ -145,29 +145,29 @@\n \n \n \n \n
ID3D11Buffer * _patchParamBuffer
 
ID3D11ShaderResourceView * _patchParamBufferSRV
 
\n

Detailed Description

\n
\n-

Definition at line 49 of file d3d11PatchTable.h.

\n+

Definition at line 49 of file d3d11PatchTable.h.

\n

Member Typedef Documentation

\n \n

◆ VertexBufferBinding

\n \n
\n
\n \n \n \n \n
typedef ID3D11Buffer* VertexBufferBinding
\n
\n \n-

Definition at line 51 of file d3d11PatchTable.h.

\n+

Definition at line 51 of file d3d11PatchTable.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ D3D11PatchTable()

\n \n@@ -269,15 +269,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 57 of file d3d11PatchTable.h.

\n+

Definition at line 57 of file d3d11PatchTable.h.

\n \n
\n \n \n

◆ Create() [2/2]

\n \n
\n@@ -332,15 +332,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 65 of file d3d11PatchTable.h.

\n+

Definition at line 65 of file d3d11PatchTable.h.

\n \n
\n \n \n

◆ GetPatchIndexBuffer()

\n \n
\n@@ -361,15 +361,15 @@\n inline \n \n \n
\n \n

Returns the index buffer containing the patch control vertices.

\n \n-

Definition at line 70 of file d3d11PatchTable.h.

\n+

Definition at line 70 of file d3d11PatchTable.h.

\n \n
\n \n \n

◆ GetPatchParamSRV()

\n \n
\n@@ -390,15 +390,15 @@\n inline \n \n \n
\n \n

Returns the SRV containing the patch parameter.

\n \n-

Definition at line 75 of file d3d11PatchTable.h.

\n+

Definition at line 75 of file d3d11PatchTable.h.

\n \n
\n \n

Member Data Documentation

\n \n

◆ _indexBuffer

\n \n@@ -415,15 +415,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 86 of file d3d11PatchTable.h.

\n+

Definition at line 86 of file d3d11PatchTable.h.

\n \n
\n \n \n

◆ _patchArrays

\n \n
\n@@ -439,15 +439,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 84 of file d3d11PatchTable.h.

\n+

Definition at line 84 of file d3d11PatchTable.h.

\n \n
\n \n \n

◆ _patchParamBuffer

\n \n
\n@@ -463,15 +463,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 87 of file d3d11PatchTable.h.

\n+

Definition at line 87 of file d3d11PatchTable.h.

\n \n
\n \n \n

◆ _patchParamBufferSRV

\n \n
\n@@ -487,20 +487,20 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 88 of file d3d11PatchTable.h.

\n+

Definition at line 88 of file d3d11PatchTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • D3D11PatchTable
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01205.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01205.html", "unified_diff": "@@ -92,15 +92,15 @@\n
    D3D11VertexBuffer Class Reference
    \n
\n
\n \n

Concrete vertex buffer class for DirectX subdivision and DirectX drawing. \n More...

\n \n-

#include <d3d11VertexBuffer.h>

\n+

#include <d3d11VertexBuffer.h>

\n \n \n \n \n \n \n@@ -135,15 +135,15 @@\n \n \n

\n Public Member Functions

virtual ~D3D11VertexBuffer ()
 Destructor. More...
 
void UpdateData (const float *src, int startVertex, int numVertices, ID3D11DeviceContext *deviceContext)
bool allocate (ID3D11Device *device)
 
\n

Detailed Description

\n

Concrete vertex buffer class for DirectX subdivision and DirectX drawing.

\n

D3D11VertexBuffer implements D3D11VertexBufferInterface. An instance of this buffer class can be passed to D3D11ComputeEvaluator.

\n \n-

Definition at line 46 of file d3d11VertexBuffer.h.

\n+

Definition at line 46 of file d3d11VertexBuffer.h.

\n

Constructor & Destructor Documentation

\n \n

◆ ~D3D11VertexBuffer()

\n \n
\n
\n \n@@ -295,15 +295,15 @@\n inline\n \n
\n
\n \n

Returns the D3D11 buffer object (for Osd::Mesh interface)

\n \n-

Definition at line 70 of file d3d11VertexBuffer.h.

\n+

Definition at line 70 of file d3d11VertexBuffer.h.

\n \n
\n
\n \n

◆ Create()

\n \n
\n@@ -423,15 +423,15 @@\n \n
\n

This method is meant to be used in client code in order to provide coarse vertices data to Osd.

\n \n
\n
\n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • D3D11VertexBuffer
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01209.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01209.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    GLStencilTableSSBO Class Reference
    \n
\n
\n \n

GL stencil table (Shader Storage buffer) \n More...

\n \n-

#include <glComputeEvaluator.h>

\n+

#include <glComputeEvaluator.h>

\n \n \n \n \n \n \n@@ -134,15 +134,15 @@\n \n

\n Public Member Functions

 GLStencilTableSSBO (Far::StencilTable const *stencilTable)
 
 GLStencilTableSSBO (Far::LimitStencilTable const *limitStencilTable)
 
 
\n

Detailed Description

\n

GL stencil table (Shader Storage buffer)

\n

This class is a GLSL SSBO representation of Far::StencilTable.

\n

GLSLComputeKernel consumes this table to apply stencils

\n \n-

Definition at line 51 of file glComputeEvaluator.h.

\n+

Definition at line 51 of file glComputeEvaluator.h.

\n

Constructor & Destructor Documentation

\n \n

◆ GLStencilTableSSBO() [1/2]

\n \n
\n
\n \n@@ -240,15 +240,15 @@\n \n \n \n
\n inlinestatic
\n
\n \n-

Definition at line 58 of file glComputeEvaluator.h.

\n+

Definition at line 58 of file glComputeEvaluator.h.

\n \n
\n
\n \n

◆ Create() [2/2]

\n \n
\n@@ -278,15 +278,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 53 of file glComputeEvaluator.h.

\n+

Definition at line 53 of file glComputeEvaluator.h.

\n \n
\n
\n \n

◆ GetDuuWeightsBuffer()

\n \n
\n@@ -305,15 +305,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 76 of file glComputeEvaluator.h.

\n+

Definition at line 76 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ GetDuvWeightsBuffer()

\n \n
\n@@ -332,15 +332,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 77 of file glComputeEvaluator.h.

\n+

Definition at line 77 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ GetDuWeightsBuffer()

\n \n
\n@@ -359,15 +359,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 74 of file glComputeEvaluator.h.

\n+

Definition at line 74 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ GetDvvWeightsBuffer()

\n \n
\n@@ -386,15 +386,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 78 of file glComputeEvaluator.h.

\n+

Definition at line 78 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ GetDvWeightsBuffer()

\n \n
\n@@ -413,15 +413,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 75 of file glComputeEvaluator.h.

\n+

Definition at line 75 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ GetIndicesBuffer()

\n \n
\n@@ -440,15 +440,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 72 of file glComputeEvaluator.h.

\n+

Definition at line 72 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ GetNumStencils()

\n \n
\n@@ -467,15 +467,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 79 of file glComputeEvaluator.h.

\n+

Definition at line 79 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ GetOffsetsBuffer()

\n \n
\n@@ -494,15 +494,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 71 of file glComputeEvaluator.h.

\n+

Definition at line 71 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ GetSizesBuffer()

\n \n
\n@@ -521,15 +521,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 70 of file glComputeEvaluator.h.

\n+

Definition at line 70 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ GetWeightsBuffer()

\n \n
\n@@ -548,20 +548,20 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 73 of file glComputeEvaluator.h.

\n+

Definition at line 73 of file glComputeEvaluator.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
GLComputeEvaluator Class Reference
\n \n
\n \n-

#include <glComputeEvaluator.h>

\n+

#include <glComputeEvaluator.h>

\n \n \n \n \n

\n Public Types

typedef bool Instantiatable
 
\n \n \n \n \n

\n@@ -226,29 +226,29 @@\n

 
static void Synchronize (void *deviceContext)
 Wait the dispatched kernel finishes. More...
 
\n

Detailed Description

\n
\n-

Definition at line 96 of file glComputeEvaluator.h.

\n+

Definition at line 96 of file glComputeEvaluator.h.

\n

Member Typedef Documentation

\n \n

◆ Instantiatable

\n \n
\n
\n \n \n \n \n
typedef bool Instantiatable
\n
\n \n-

Definition at line 98 of file glComputeEvaluator.h.

\n+

Definition at line 98 of file glComputeEvaluator.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ GLComputeEvaluator()

\n \n@@ -416,15 +416,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 111 of file glComputeEvaluator.h.

\n+

Definition at line 111 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ Create() [2/2]

\n \n
\n@@ -472,15 +472,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 99 of file glComputeEvaluator.h.

\n+

Definition at line 99 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [1/8]

\n \n
\n@@ -848,15 +848,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 1071 of file glComputeEvaluator.h.

\n+

Definition at line 1071 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [4/8]

\n \n
\n@@ -1014,15 +1014,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 870 of file glComputeEvaluator.h.

\n+

Definition at line 870 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [5/8]

\n \n
\n@@ -1124,15 +1124,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 997 of file glComputeEvaluator.h.

\n+

Definition at line 997 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [6/8]

\n \n
\n@@ -1248,15 +1248,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 769 of file glComputeEvaluator.h.

\n+

Definition at line 769 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [7/8]

\n \n
\n@@ -1330,15 +1330,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 943 of file glComputeEvaluator.h.

\n+

Definition at line 943 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [8/8]

\n \n
\n@@ -1429,15 +1429,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 691 of file glComputeEvaluator.h.

\n+

Definition at line 691 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [1/6]

\n \n
\n@@ -1602,15 +1602,15 @@\n fvarChannelface-varying channel\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1903 of file glComputeEvaluator.h.

\n+

Definition at line 1903 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [2/6]

\n \n
\n@@ -1761,15 +1761,15 @@\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent\n fvarChannelface-varying channel \n \n \n \n \n-

Definition at line 2011 of file glComputeEvaluator.h.

\n+

Definition at line 2011 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [3/6]

\n \n
\n@@ -1892,15 +1892,15 @@\n fvarChannelface-varying channel\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1736 of file glComputeEvaluator.h.

\n+

Definition at line 1736 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [4/6]

\n \n
\n@@ -2009,15 +2009,15 @@\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent\n fvarChannelface-varying channel \n \n \n \n \n-

Definition at line 1816 of file glComputeEvaluator.h.

\n+

Definition at line 1816 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [5/6]

\n \n
\n@@ -2112,15 +2112,15 @@\n fvarChannelface-varying channel\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1606 of file glComputeEvaluator.h.

\n+

Definition at line 1606 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [6/6]

\n \n
\n@@ -2201,15 +2201,15 @@\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent\n fvarChannelface-varying channel \n \n \n \n \n-

Definition at line 1669 of file glComputeEvaluator.h.

\n+

Definition at line 1669 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [1/6]

\n \n
\n@@ -2353,15 +2353,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 1544 of file glComputeEvaluator.h.

\n+

Definition at line 1544 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [2/6]

\n \n
\n@@ -2519,15 +2519,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1439 of file glComputeEvaluator.h.

\n+

Definition at line 1439 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [3/6]

\n \n
\n@@ -2629,15 +2629,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 1355 of file glComputeEvaluator.h.

\n+

Definition at line 1355 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [4/6]

\n \n
\n@@ -2753,15 +2753,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1278 of file glComputeEvaluator.h.

\n+

Definition at line 1278 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [5/6]

\n \n
\n@@ -2835,15 +2835,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 1214 of file glComputeEvaluator.h.

\n+

Definition at line 1214 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [6/6]

\n \n
\n@@ -2931,15 +2931,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1154 of file glComputeEvaluator.h.

\n+

Definition at line 1154 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [1/8]

\n \n
\n@@ -3402,15 +3402,15 @@\n dvvBufferOutput buffer 2nd derivative wrt v must have BindVBO() method returning a GL buffer object of destination data\n dvvDescvertex buffer descriptor for the dvvBuffer\n stencilTablestencil table to be applied. The table must have SSBO interfaces. \n \n \n \n \n-

Definition at line 502 of file glComputeEvaluator.h.

\n+

Definition at line 502 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [4/8]

\n \n
\n@@ -3554,15 +3554,15 @@\n stencilTablestencil table to be applied. The table must have SSBO interfaces.\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLSL kernel \n \n \n \n \n-

Definition at line 326 of file glComputeEvaluator.h.

\n+

Definition at line 326 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [5/8]

\n \n
\n@@ -3650,15 +3650,15 @@\n dvBufferOutput buffer derivative wrt v must have BindVBO() method returning a GL buffer object of destination data\n dvDescvertex buffer descriptor for the dvBuffer\n stencilTablestencil table to be applied. The table must have SSBO interfaces. \n \n \n \n \n-

Definition at line 434 of file glComputeEvaluator.h.

\n+

Definition at line 434 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [6/8]

\n \n
\n@@ -3760,15 +3760,15 @@\n stencilTablestencil table to be applied. The table must have SSBO interfaces.\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLSL kernel \n \n \n \n \n-

Definition at line 236 of file glComputeEvaluator.h.

\n+

Definition at line 236 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [7/8]

\n \n
\n@@ -3828,15 +3828,15 @@\n dstBufferOutput primvar buffer must have BindVBO() method returning a GL buffer object of destination data\n dstDescvertex buffer descriptor for the output buffer\n stencilTablestencil table to be applied. The table must have SSBO interfaces. \n \n \n \n \n-

Definition at line 386 of file glComputeEvaluator.h.

\n+

Definition at line 386 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [8/8]

\n \n
\n@@ -3913,15 +3913,15 @@\n stencilTablestencil table to be applied. The table must have SSBO interfaces.\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLSL kernel \n \n \n \n \n-

Definition at line 168 of file glComputeEvaluator.h.

\n+

Definition at line 168 of file glComputeEvaluator.h.

\n \n
\n \n \n

◆ Synchronize()

\n \n
\n@@ -3946,15 +3946,15 @@\n
\n \n

Wait the dispatched kernel finishes.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
GLLegacyGregoryPatchTable Class Reference
\n \n
\n \n-

#include <glLegacyGregoryPatchTable.h>

\n+

#include <glLegacyGregoryPatchTable.h>

\n
\n Inheritance diagram for GLLegacyGregoryPatchTable:
\n
\n
\n \"\"/\n
\n \n@@ -124,15 +124,15 @@\n \n \n \n

\n Protected Member Functions

 GLLegacyGregoryPatchTable ()
 
\n

Detailed Description

\n
\n-

Definition at line 39 of file glLegacyGregoryPatchTable.h.

\n+

Definition at line 39 of file glLegacyGregoryPatchTable.h.

\n

Constructor & Destructor Documentation

\n \n

◆ ~GLLegacyGregoryPatchTable()

\n \n
\n
\n \n@@ -219,15 +219,15 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 60 of file glLegacyGregoryPatchTable.h.

\n+

Definition at line 60 of file glLegacyGregoryPatchTable.h.

\n \n
\n
\n \n

◆ GetQuadOffsetsTextureBuffer()

\n \n
\n@@ -246,15 +246,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 56 of file glLegacyGregoryPatchTable.h.

\n+

Definition at line 56 of file glLegacyGregoryPatchTable.h.

\n \n
\n
\n \n

◆ GetVertexTextureBuffer()

\n \n
\n@@ -273,15 +273,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 48 of file glLegacyGregoryPatchTable.h.

\n+

Definition at line 48 of file glLegacyGregoryPatchTable.h.

\n \n
\n \n \n

◆ GetVertexValenceTextureBuffer()

\n \n
\n@@ -300,15 +300,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 52 of file glLegacyGregoryPatchTable.h.

\n+

Definition at line 52 of file glLegacyGregoryPatchTable.h.

\n \n
\n \n \n

◆ UpdateVertexBuffer()

\n \n
\n@@ -323,15 +323,15 @@\n \n \n
\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
GLPatchTable Class Reference
\n \n
\n \n-

#include <glPatchTable.h>

\n+

#include <glPatchTable.h>

\n
\n Inheritance diagram for GLPatchTable:
\n
\n
\n \"\"/\n
\n \n@@ -194,29 +194,29 @@\n \n \n \n \n
std::vector< GLuint > _fvarParamBuffers
 
std::vector< GLuint > _fvarParamTextures
 
\n

Detailed Description

\n
\n-

Definition at line 45 of file glPatchTable.h.

\n+

Definition at line 45 of file glPatchTable.h.

\n

Member Typedef Documentation

\n \n

◆ VertexBufferBinding

\n \n
\n
\n \n \n \n \n
typedef GLuint VertexBufferBinding
\n
\n \n-

Definition at line 47 of file glPatchTable.h.

\n+

Definition at line 47 of file glPatchTable.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ ~GLPatchTable()

\n \n@@ -344,15 +344,15 @@\n inline \n \n \n
\n \n

Returns the patch arrays for face-varying index buffer data.

\n \n-

Definition at line 98 of file glPatchTable.h.

\n+

Definition at line 98 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchIndexBuffer()

\n \n
\n@@ -374,15 +374,15 @@\n inline \n \n \n
\n \n

Returns the GL index buffer containing face-varying control vertices.

\n \n-

Definition at line 103 of file glPatchTable.h.

\n+

Definition at line 103 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchIndexTextureBuffer()

\n \n
\n@@ -404,15 +404,15 @@\n inline \n \n \n
\n \n

Returns the GL texture buffer containing face-varying control vertices.

\n \n-

Definition at line 108 of file glPatchTable.h.

\n+

Definition at line 108 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchParamBuffer()

\n \n
\n@@ -434,15 +434,15 @@\n inline \n \n \n
\n \n

Returns the GL index buffer containing face-varying patch params.

\n \n-

Definition at line 113 of file glPatchTable.h.

\n+

Definition at line 113 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchParamTextureBuffer()

\n \n
\n@@ -464,15 +464,15 @@\n inline \n \n \n
\n \n

Returns the GL texture buffer containing face-varying patch params.

\n \n-

Definition at line 118 of file glPatchTable.h.

\n+

Definition at line 118 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetNumFVarChannels()

\n \n
\n@@ -493,15 +493,15 @@\n inline \n \n \n
\n \n

Returns the number of face-varying channel buffers.

\n \n-

Definition at line 95 of file glPatchTable.h.

\n+

Definition at line 95 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetPatchArrays()

\n \n
\n@@ -522,15 +522,15 @@\n inline \n \n \n
\n \n

Returns the patch arrays for vertex index buffer data.

\n \n-

Definition at line 55 of file glPatchTable.h.

\n+

Definition at line 55 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetPatchIndexBuffer()

\n \n
\n@@ -551,15 +551,15 @@\n inline \n \n \n
\n \n

Returns the GL index buffer containing the patch control vertices.

\n \n-

Definition at line 60 of file glPatchTable.h.

\n+

Definition at line 60 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetPatchIndexTextureBuffer()

\n \n
\n@@ -580,15 +580,15 @@\n inline \n \n \n
\n \n

Returns the GL texture buffer containing the patch control vertices.

\n \n-

Definition at line 70 of file glPatchTable.h.

\n+

Definition at line 70 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetPatchParamBuffer()

\n \n
\n@@ -609,15 +609,15 @@\n inline \n \n \n
\n \n

Returns the GL index buffer containing the patch parameter.

\n \n-

Definition at line 65 of file glPatchTable.h.

\n+

Definition at line 65 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetPatchParamTextureBuffer()

\n \n
\n@@ -638,15 +638,15 @@\n inline \n \n \n
\n \n

Returns the GL texture buffer containing the patch parameter.

\n \n-

Definition at line 75 of file glPatchTable.h.

\n+

Definition at line 75 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchArrays()

\n \n
\n@@ -667,15 +667,15 @@\n inline \n \n \n
\n \n

Returns the patch arrays for varying index buffer data.

\n \n-

Definition at line 80 of file glPatchTable.h.

\n+

Definition at line 80 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchIndexBuffer()

\n \n
\n@@ -696,15 +696,15 @@\n inline \n \n \n
\n \n

Returns the GL index buffer containing the varying control vertices.

\n \n-

Definition at line 85 of file glPatchTable.h.

\n+

Definition at line 85 of file glPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchIndexTextureBuffer()

\n \n
\n@@ -725,15 +725,15 @@\n inline \n \n \n
\n \n

Returns the GL texture buffer containing the varying control vertices.

\n \n-

Definition at line 90 of file glPatchTable.h.

\n+

Definition at line 90 of file glPatchTable.h.

\n \n
\n \n

Member Data Documentation

\n \n

◆ _fvarIndexBuffers

\n \n@@ -750,15 +750,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 141 of file glPatchTable.h.

\n+

Definition at line 141 of file glPatchTable.h.

\n \n
\n \n \n

◆ _fvarIndexTextures

\n \n
\n@@ -774,15 +774,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 142 of file glPatchTable.h.

\n+

Definition at line 142 of file glPatchTable.h.

\n \n
\n \n \n

◆ _fvarParamBuffers

\n \n
\n@@ -798,15 +798,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 144 of file glPatchTable.h.

\n+

Definition at line 144 of file glPatchTable.h.

\n \n
\n \n \n

◆ _fvarParamTextures

\n \n
\n@@ -822,15 +822,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 145 of file glPatchTable.h.

\n+

Definition at line 145 of file glPatchTable.h.

\n \n
\n \n \n

◆ _fvarPatchArrays

\n \n
\n@@ -846,15 +846,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 140 of file glPatchTable.h.

\n+

Definition at line 140 of file glPatchTable.h.

\n \n
\n \n \n

◆ _patchArrays

\n \n
\n@@ -870,15 +870,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 128 of file glPatchTable.h.

\n+

Definition at line 128 of file glPatchTable.h.

\n \n
\n \n \n

◆ _patchIndexBuffer

\n \n
\n@@ -894,15 +894,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 130 of file glPatchTable.h.

\n+

Definition at line 130 of file glPatchTable.h.

\n \n
\n \n \n

◆ _patchIndexTexture

\n \n
\n@@ -918,15 +918,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 133 of file glPatchTable.h.

\n+

Definition at line 133 of file glPatchTable.h.

\n \n
\n \n \n

◆ _patchParamBuffer

\n \n
\n@@ -942,15 +942,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 131 of file glPatchTable.h.

\n+

Definition at line 131 of file glPatchTable.h.

\n \n
\n \n \n

◆ _patchParamTexture

\n \n
\n@@ -966,15 +966,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 134 of file glPatchTable.h.

\n+

Definition at line 134 of file glPatchTable.h.

\n \n
\n \n \n

◆ _varyingIndexBuffer

\n \n
\n@@ -990,15 +990,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 137 of file glPatchTable.h.

\n+

Definition at line 137 of file glPatchTable.h.

\n \n
\n \n \n

◆ _varyingIndexTexture

\n \n
\n@@ -1014,15 +1014,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 138 of file glPatchTable.h.

\n+

Definition at line 138 of file glPatchTable.h.

\n \n
\n \n \n

◆ _varyingPatchArrays

\n \n
\n@@ -1038,20 +1038,20 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 136 of file glPatchTable.h.

\n+

Definition at line 136 of file glPatchTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
\n \n-

#include <glslPatchShaderSource.h>

\n+

#include <glslPatchShaderSource.h>

\n \n \n \n \n \n \n@@ -104,15 +104,15 @@\n \n \n \n \n

\n Static Public Member Functions

static std::string GetCommonShaderSource ()
 
static std::string GetPatchBasisShaderSource ()
 
static std::string GetTessControlShaderSource (Far::PatchDescriptor::Type type)
 
static std::string GetTessEvalShaderSource (Far::PatchDescriptor::Type type)
 
\n

Detailed Description

\n
\n-

Definition at line 37 of file glslPatchShaderSource.h.

\n+

Definition at line 37 of file glslPatchShaderSource.h.

\n

Member Function Documentation

\n \n

◆ GetCommonShaderSource()

\n \n
\n
\n \n@@ -235,15 +235,15 @@\n \n
\n
\n \n
\n
\n
The documentation for this class was generated from the following file:\n
\n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • GLSLPatchShaderSource
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01237.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01237.html", "unified_diff": "@@ -92,15 +92,15 @@\n
    GLVertexBuffer Class Reference
    \n
\n
\n \n

Concrete vertex buffer class for GLSL subdivision and OpenGL drawing. \n More...

\n \n-

#include <glVertexBuffer.h>

\n+

#include <glVertexBuffer.h>

\n \n \n \n \n \n \n@@ -129,15 +129,15 @@\n \n \n

\n Public Member Functions

 ~GLVertexBuffer ()
 Destructor. More...
 
void UpdateData (const float *src, int startVertex, int numVertices, void *deviceContext=NULL)
bool allocate ()
 
\n

Detailed Description

\n

Concrete vertex buffer class for GLSL subdivision and OpenGL drawing.

\n

GLVertexBuffer implements GLVertexBufferInterface. An instance of this buffer class can be passed to OsdGLComputeEvaluator.

\n \n-

Definition at line 44 of file glVertexBuffer.h.

\n+

Definition at line 44 of file glVertexBuffer.h.

\n

Constructor & Destructor Documentation

\n \n

◆ ~GLVertexBuffer()

\n \n
\n
\n \n@@ -359,15 +359,15 @@\n
\n
\n

This method is meant to be used in client code in order to provide coarse vertices data to Osd.

\n \n
\n
\n
The documentation for this class was generated from the following file:\n
\n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • GLVertexBuffer
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01241.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01241.html", "unified_diff": "@@ -91,15 +91,15 @@\n
    GLStencilTableTBO Class Reference
    \n
\n
\n \n

GL TextureBuffer stencil table. \n More...

\n \n-

#include <glXFBEvaluator.h>

\n+

#include <glXFBEvaluator.h>

\n \n \n \n \n \n \n@@ -134,15 +134,15 @@\n \n

\n Public Member Functions

 GLStencilTableTBO (Far::StencilTable const *stencilTable)
 
 GLStencilTableTBO (Far::LimitStencilTable const *limitStencilTable)
 
 
\n

Detailed Description

\n

GL TextureBuffer stencil table.

\n

This class is a GL Texture Buffer representation of Far::StencilTable.

\n

GLSLTransformFeedback consumes this table to apply stencils

\n \n-

Definition at line 52 of file glXFBEvaluator.h.

\n+

Definition at line 52 of file glXFBEvaluator.h.

\n

Constructor & Destructor Documentation

\n \n

◆ GLStencilTableTBO() [1/2]

\n \n
\n
\n \n@@ -240,15 +240,15 @@\n \n \n \n
\n inlinestatic
\n
\n \n-

Definition at line 60 of file glXFBEvaluator.h.

\n+

Definition at line 60 of file glXFBEvaluator.h.

\n \n
\n
\n \n

◆ Create() [2/2]

\n \n
\n@@ -278,15 +278,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 54 of file glXFBEvaluator.h.

\n+

Definition at line 54 of file glXFBEvaluator.h.

\n \n
\n
\n \n

◆ GetDuuWeightsTexture()

\n \n
\n@@ -305,15 +305,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 78 of file glXFBEvaluator.h.

\n+

Definition at line 78 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ GetDuvWeightsTexture()

\n \n
\n@@ -332,15 +332,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 79 of file glXFBEvaluator.h.

\n+

Definition at line 79 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ GetDuWeightsTexture()

\n \n
\n@@ -359,15 +359,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 76 of file glXFBEvaluator.h.

\n+

Definition at line 76 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ GetDvvWeightsTexture()

\n \n
\n@@ -386,15 +386,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 80 of file glXFBEvaluator.h.

\n+

Definition at line 80 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ GetDvWeightsTexture()

\n \n
\n@@ -413,15 +413,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 77 of file glXFBEvaluator.h.

\n+

Definition at line 77 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ GetIndicesTexture()

\n \n
\n@@ -440,15 +440,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 74 of file glXFBEvaluator.h.

\n+

Definition at line 74 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ GetNumStencils()

\n \n
\n@@ -467,15 +467,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 81 of file glXFBEvaluator.h.

\n+

Definition at line 81 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ GetOffsetsTexture()

\n \n
\n@@ -494,15 +494,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 73 of file glXFBEvaluator.h.

\n+

Definition at line 73 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ GetSizesTexture()

\n \n
\n@@ -521,15 +521,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 72 of file glXFBEvaluator.h.

\n+

Definition at line 72 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ GetWeightsTexture()

\n \n
\n@@ -548,20 +548,20 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 75 of file glXFBEvaluator.h.

\n+

Definition at line 75 of file glXFBEvaluator.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
GLXFBEvaluator Class Reference
\n \n
\n \n-

#include <glXFBEvaluator.h>

\n+

#include <glXFBEvaluator.h>

\n \n \n \n \n

\n Public Types

typedef bool Instantiatable
 
\n \n \n \n \n

\n@@ -240,29 +240,29 @@\n

 
static void Synchronize (void *kernel)
 Wait the dispatched kernel finishes. More...
 
\n

Detailed Description

\n
\n-

Definition at line 98 of file glXFBEvaluator.h.

\n+

Definition at line 98 of file glXFBEvaluator.h.

\n

Member Typedef Documentation

\n \n

◆ Instantiatable

\n \n
\n
\n \n \n \n \n
typedef bool Instantiatable
\n
\n \n-

Definition at line 100 of file glXFBEvaluator.h.

\n+

Definition at line 100 of file glXFBEvaluator.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ GLXFBEvaluator()

\n \n@@ -414,15 +414,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 126 of file glXFBEvaluator.h.

\n+

Definition at line 126 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ Create() [2/6]

\n \n
\n@@ -488,15 +488,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 170 of file glXFBEvaluator.h.

\n+

Definition at line 170 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ Create() [3/6]

\n \n
\n@@ -564,15 +564,15 @@\n inlinestatic \n \n \n
\n \n

Generic creator template.

\n \n-

Definition at line 140 of file glXFBEvaluator.h.

\n+

Definition at line 140 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ Create() [4/6]

\n \n
\n@@ -640,15 +640,15 @@\n inlinestatic \n \n \n
\n \n

Specialization to allow creation without a device context.

\n \n-

Definition at line 157 of file glXFBEvaluator.h.

\n+

Definition at line 157 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ Create() [5/6]

\n \n
\n@@ -698,15 +698,15 @@\n inlinestatic \n \n \n
\n \n

Generic creator template.

\n \n-

Definition at line 104 of file glXFBEvaluator.h.

\n+

Definition at line 104 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ Create() [6/6]

\n \n
\n@@ -756,15 +756,15 @@\n inlinestatic \n \n \n
\n \n

Specialization to allow creation without a device context.

\n \n-

Definition at line 117 of file glXFBEvaluator.h.

\n+

Definition at line 117 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [1/8]

\n \n
\n@@ -1132,15 +1132,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 1144 of file glXFBEvaluator.h.

\n+

Definition at line 1144 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [4/8]

\n \n
\n@@ -1298,15 +1298,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 943 of file glXFBEvaluator.h.

\n+

Definition at line 943 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [5/8]

\n \n
\n@@ -1408,15 +1408,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 1070 of file glXFBEvaluator.h.

\n+

Definition at line 1070 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [6/8]

\n \n
\n@@ -1532,15 +1532,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 843 of file glXFBEvaluator.h.

\n+

Definition at line 843 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [7/8]

\n \n
\n@@ -1614,15 +1614,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 1016 of file glXFBEvaluator.h.

\n+

Definition at line 1016 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [8/8]

\n \n
\n@@ -1713,15 +1713,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 765 of file glXFBEvaluator.h.

\n+

Definition at line 765 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [1/6]

\n \n
\n@@ -1886,15 +1886,15 @@\n fvarChannelface-varying channel\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1976 of file glXFBEvaluator.h.

\n+

Definition at line 1976 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [2/6]

\n \n
\n@@ -2045,15 +2045,15 @@\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent\n fvarChannelface-varying channel \n \n \n \n \n-

Definition at line 2084 of file glXFBEvaluator.h.

\n+

Definition at line 2084 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [3/6]

\n \n
\n@@ -2176,15 +2176,15 @@\n fvarChannelface-varying channel\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1809 of file glXFBEvaluator.h.

\n+

Definition at line 1809 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [4/6]

\n \n
\n@@ -2293,15 +2293,15 @@\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent\n fvarChannelface-varying channel \n \n \n \n \n-

Definition at line 1889 of file glXFBEvaluator.h.

\n+

Definition at line 1889 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [5/6]

\n \n
\n@@ -2396,15 +2396,15 @@\n fvarChannelface-varying channel\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1679 of file glXFBEvaluator.h.

\n+

Definition at line 1679 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [6/6]

\n \n
\n@@ -2485,15 +2485,15 @@\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent\n fvarChannelface-varying channel \n \n \n \n \n-

Definition at line 1742 of file glXFBEvaluator.h.

\n+

Definition at line 1742 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [1/6]

\n \n
\n@@ -2637,15 +2637,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 1617 of file glXFBEvaluator.h.

\n+

Definition at line 1617 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [2/6]

\n \n
\n@@ -2803,15 +2803,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1512 of file glXFBEvaluator.h.

\n+

Definition at line 1512 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [3/6]

\n \n
\n@@ -2913,15 +2913,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 1428 of file glXFBEvaluator.h.

\n+

Definition at line 1428 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [4/6]

\n \n
\n@@ -3037,15 +3037,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1351 of file glXFBEvaluator.h.

\n+

Definition at line 1351 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [5/6]

\n \n
\n@@ -3119,15 +3119,15 @@\n numPatchCoordsnumber of patchCoords.\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableGLPatchTable or equivalent \n \n \n \n \n-

Definition at line 1287 of file glXFBEvaluator.h.

\n+

Definition at line 1287 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [6/6]

\n \n
\n@@ -3215,15 +3215,15 @@\n patchTableGLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLXFB evaluator \n \n \n \n \n-

Definition at line 1227 of file glXFBEvaluator.h.

\n+

Definition at line 1227 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [1/8]

\n \n
\n@@ -3686,15 +3686,15 @@\n dvvBufferOutput buffer 2nd derivative wrt v must have BindVBO() method returning a GL buffer object of destination data\n dvvDescvertex buffer descriptor for the dvvBuffer\n stencilTablestencil table to be applied. The table must have Texture Buffer Object interfaces. \n \n \n \n \n-

Definition at line 575 of file glXFBEvaluator.h.

\n+

Definition at line 575 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [4/8]

\n \n
\n@@ -3838,15 +3838,15 @@\n stencilTablestencil table to be applied. The table must have Texture Buffer Object interfaces.\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLSLTransformFeedback kernel \n \n \n \n \n-

Definition at line 396 of file glXFBEvaluator.h.

\n+

Definition at line 396 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [5/8]

\n \n
\n@@ -3934,15 +3934,15 @@\n dvBufferOutput buffer derivative wrt v must have BindVBO() method returning a GL buffer object of destination data\n dvDescvertex buffer descriptor for the dvBuffer\n stencilTablestencil table to be applied. The table must have Texture Buffer Object interfaces. \n \n \n \n \n-

Definition at line 506 of file glXFBEvaluator.h.

\n+

Definition at line 506 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [6/8]

\n \n
\n@@ -4044,15 +4044,15 @@\n stencilTablestencil table to be applied. The table must have Texture Buffer Object interfaces.\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLSLTransformFeedback kernel \n \n \n \n \n-

Definition at line 306 of file glXFBEvaluator.h.

\n+

Definition at line 306 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [7/8]

\n \n
\n@@ -4112,15 +4112,15 @@\n dstBufferOutput primvar buffer must have BindVBO() method returning a GL buffer object of destination data\n dstDescvertex buffer descriptor for the output buffer\n stencilTablestencil table to be applied. The table must have Texture Buffer Object interfaces. \n \n \n \n \n-

Definition at line 457 of file glXFBEvaluator.h.

\n+

Definition at line 457 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [8/8]

\n \n
\n@@ -4197,15 +4197,15 @@\n stencilTablestencil table to be applied. The table must have Texture Buffer Object interfaces.\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextnot used in the GLSLTransformFeedback kernel \n \n \n \n \n-

Definition at line 238 of file glXFBEvaluator.h.

\n+

Definition at line 238 of file glXFBEvaluator.h.

\n \n
\n \n \n

◆ Synchronize()

\n \n
\n@@ -4230,15 +4230,15 @@\n
\n \n

Wait the dispatched kernel finishes.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
\n \n-

#include <hlslPatchShaderSource.h>

\n+

#include <hlslPatchShaderSource.h>

\n \n \n \n \n \n \n@@ -104,15 +104,15 @@\n \n \n \n \n

\n Static Public Member Functions

static std::string GetCommonShaderSource ()
 
static std::string GetPatchBasisShaderSource ()
 
static std::string GetHullShaderSource (Far::PatchDescriptor::Type type)
 
static std::string GetDomainShaderSource (Far::PatchDescriptor::Type type)
 
\n

Detailed Description

\n
\n-

Definition at line 37 of file hlslPatchShaderSource.h.

\n+

Definition at line 37 of file hlslPatchShaderSource.h.

\n

Member Function Documentation

\n \n

◆ GetCommonShaderSource()

\n \n
\n
\n \n@@ -235,15 +235,15 @@\n \n
\n
\n \n
\n
\n
The documentation for this class was generated from the following file:\n
\n \n \n
\n
\n
MTLStencilTable Class Reference
\n \n
\n \n-

#include <mtlComputeEvaluator.h>

\n+

#include <mtlComputeEvaluator.h>

\n \n \n \n \n \n \n@@ -127,15 +127,15 @@\n Static Public Member Functions\n \n \n \n

\n Public Member Functions

 MTLStencilTable (Far::StencilTable const *stencilTable, MTLContext *context)
 
 MTLStencilTable (Far::LimitStencilTable const *stencilTable, MTLContext *context)
 
template<typename STENCIL_TABLE , typename DEVICE_CONTEXT >
static MTLStencilTable * Create (STENCIL_TABLE *stencilTable, DEVICE_CONTEXT context)
 
\n

Detailed Description

\n
\n-

Definition at line 50 of file mtlComputeEvaluator.h.

\n+

Definition at line 50 of file mtlComputeEvaluator.h.

\n

Constructor & Destructor Documentation

\n \n

◆ MTLStencilTable() [1/2]

\n \n
\n
\n \n@@ -237,15 +237,15 @@\n \n \n \n
\n inlinestatic
\n
\n \n-

Definition at line 54 of file mtlComputeEvaluator.h.

\n+

Definition at line 54 of file mtlComputeEvaluator.h.

\n \n
\n
\n \n

◆ GetDuuWeightsBuffer()

\n \n
\n@@ -264,15 +264,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 71 of file mtlComputeEvaluator.h.

\n+

Definition at line 71 of file mtlComputeEvaluator.h.

\n \n
\n
\n \n

◆ GetDuvWeightsBuffer()

\n \n
\n@@ -291,15 +291,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 72 of file mtlComputeEvaluator.h.

\n+

Definition at line 72 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ GetDuWeightsBuffer()

\n \n
\n@@ -318,15 +318,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 69 of file mtlComputeEvaluator.h.

\n+

Definition at line 69 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ GetDvvWeightsBuffer()

\n \n
\n@@ -345,15 +345,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 73 of file mtlComputeEvaluator.h.

\n+

Definition at line 73 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ GetDvWeightsBuffer()

\n \n
\n@@ -372,15 +372,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 70 of file mtlComputeEvaluator.h.

\n+

Definition at line 70 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ GetIndicesBuffer()

\n \n
\n@@ -399,15 +399,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 67 of file mtlComputeEvaluator.h.

\n+

Definition at line 67 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ GetNumStencils()

\n \n
\n@@ -426,15 +426,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 75 of file mtlComputeEvaluator.h.

\n+

Definition at line 75 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ GetOffsetsBuffer()

\n \n
\n@@ -453,15 +453,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 66 of file mtlComputeEvaluator.h.

\n+

Definition at line 66 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ GetSizesBuffer()

\n \n
\n@@ -480,15 +480,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 65 of file mtlComputeEvaluator.h.

\n+

Definition at line 65 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ GetWeightsBuffer()

\n \n
\n@@ -507,20 +507,20 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 68 of file mtlComputeEvaluator.h.

\n+

Definition at line 68 of file mtlComputeEvaluator.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
MTLComputeEvaluator Class Reference
\n \n
\n \n-

#include <mtlComputeEvaluator.h>

\n+

#include <mtlComputeEvaluator.h>

\n \n \n \n \n

\n Public Types

typedef bool Instantiatable
 
\n \n \n \n \n

\n@@ -224,29 +224,29 @@\n

 
static void Synchronize (MTLContext *context)
 Wait for the dispatched kernel to finish. More...
 
\n

Detailed Description

\n
\n-

Definition at line 91 of file mtlComputeEvaluator.h.

\n+

Definition at line 91 of file mtlComputeEvaluator.h.

\n

Member Typedef Documentation

\n \n

◆ Instantiatable

\n \n
\n
\n \n \n \n \n
typedef bool Instantiatable
\n
\n \n-

Definition at line 94 of file mtlComputeEvaluator.h.

\n+

Definition at line 94 of file mtlComputeEvaluator.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ MTLComputeEvaluator()

\n \n@@ -868,15 +868,15 @@\n patchTableMTLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 889 of file mtlComputeEvaluator.h.

\n+

Definition at line 889 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [4/8]

\n \n
\n@@ -1027,15 +1027,15 @@\n patchCoordsarray of locations to be evaluated.\n patchTableMTLPatchTable or equivalent\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1104 of file mtlComputeEvaluator.h.

\n+

Definition at line 1104 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [5/8]

\n \n
\n@@ -1151,15 +1151,15 @@\n patchTableMTLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 786 of file mtlComputeEvaluator.h.

\n+

Definition at line 786 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [6/8]

\n \n
\n@@ -1268,15 +1268,15 @@\n patchCoordsarray of locations to be evaluated.\n patchTableMTLPatchTable or equivalent\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1025 of file mtlComputeEvaluator.h.

\n+

Definition at line 1025 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [7/8]

\n \n
\n@@ -1367,15 +1367,15 @@\n patchTableMTLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 705 of file mtlComputeEvaluator.h.

\n+

Definition at line 705 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatches() [8/8]

\n \n
\n@@ -1456,15 +1456,15 @@\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableMTLPatchTable or equivalent\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 966 of file mtlComputeEvaluator.h.

\n+

Definition at line 966 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [1/6]

\n \n
\n@@ -1629,15 +1629,15 @@\n fvarChannelface-varying channel\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1991 of file mtlComputeEvaluator.h.

\n+

Definition at line 1991 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [2/6]

\n \n
\n@@ -1795,15 +1795,15 @@\n patchTableMTLPatchTable or equivalent\n fvarChannelface-varying channel\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 2106 of file mtlComputeEvaluator.h.

\n+

Definition at line 2106 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [3/6]

\n \n
\n@@ -1926,15 +1926,15 @@\n fvarChannelface-varying channel\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1813 of file mtlComputeEvaluator.h.

\n+

Definition at line 1813 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [4/6]

\n \n
\n@@ -2050,15 +2050,15 @@\n patchTableMTLPatchTable or equivalent\n fvarChannelface-varying channel\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1900 of file mtlComputeEvaluator.h.

\n+

Definition at line 1900 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [5/6]

\n \n
\n@@ -2153,15 +2153,15 @@\n fvarChannelface-varying channel\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1672 of file mtlComputeEvaluator.h.

\n+

Definition at line 1672 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [6/6]

\n \n
\n@@ -2249,15 +2249,15 @@\n patchTableMTLPatchTable or equivalent\n fvarChannelface-varying channel\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1742 of file mtlComputeEvaluator.h.

\n+

Definition at line 1742 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [1/6]

\n \n
\n@@ -2415,15 +2415,15 @@\n patchTableMTLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1496 of file mtlComputeEvaluator.h.

\n+

Definition at line 1496 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [2/6]

\n \n
\n@@ -2574,15 +2574,15 @@\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableMTLPatchTable or equivalent\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1606 of file mtlComputeEvaluator.h.

\n+

Definition at line 1606 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [3/6]

\n \n
\n@@ -2698,15 +2698,15 @@\n patchTableMTLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1325 of file mtlComputeEvaluator.h.

\n+

Definition at line 1325 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [4/6]

\n \n
\n@@ -2815,15 +2815,15 @@\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableMTLPatchTable or equivalent\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1407 of file mtlComputeEvaluator.h.

\n+

Definition at line 1407 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [5/6]

\n \n
\n@@ -2911,15 +2911,15 @@\n patchTableMTLPatchTable or equivalent\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1192 of file mtlComputeEvaluator.h.

\n+

Definition at line 1192 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [6/6]

\n \n
\n@@ -3000,15 +3000,15 @@\n patchCoordsarray of locations to be evaluated. must have BindVBO() method returning an array of PatchCoord struct in VBO.\n patchTableMTLPatchTable or equivalent\n deviceContextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 1257 of file mtlComputeEvaluator.h.

\n+

Definition at line 1257 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [1/8]

\n \n
\n@@ -3499,15 +3499,15 @@\n stencilTablestencil table to be applied. The table must have MTLBuffer interfaces.\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 312 of file mtlComputeEvaluator.h.

\n+

Definition at line 312 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [4/8]

\n \n
\n@@ -3644,15 +3644,15 @@\n dvvDescvertex buffer descriptor for the dvvBuffer\n stencilTablestencil table to be applied. The table must have MTLBuffer interfaces.\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 504 of file mtlComputeEvaluator.h.

\n+

Definition at line 504 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [5/8]

\n \n
\n@@ -3754,15 +3754,15 @@\n stencilTablestencil table to be applied. The table must have MTLBuffer interfaces.\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 220 of file mtlComputeEvaluator.h.

\n+

Definition at line 220 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [6/8]

\n \n
\n@@ -3857,15 +3857,15 @@\n dvDescvertex buffer descriptor for the dvBuffer\n stencilTablestencil table to be applied. The table must have MTLBuffer interfaces.\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 430 of file mtlComputeEvaluator.h.

\n+

Definition at line 430 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [7/8]

\n \n
\n@@ -3942,15 +3942,15 @@\n stencilTablestencil table to be applied. The table must have MTLBuffer interfaces.\n instancecached compiled instance. Clients are supposed to pre-compile an instance of this class and provide to this function. If it's null the kernel still compute by instantiating on-demand kernel although it may cause a performance problem.\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 149 of file mtlComputeEvaluator.h.

\n+

Definition at line 149 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [8/8]

\n \n
\n@@ -4017,15 +4017,15 @@\n dstDescvertex buffer descriptor for the output buffer\n stencilTablestencil table to be applied. The table must have MTLBuffer interfaces.\n contextused to obtain the MTLDevice object and command queue to obtain command buffers from. \n \n \n \n \n-

Definition at line 376 of file mtlComputeEvaluator.h.

\n+

Definition at line 376 of file mtlComputeEvaluator.h.

\n \n
\n \n \n

◆ Synchronize()

\n \n
\n@@ -4050,15 +4050,15 @@\n
\n \n

Wait for the dispatched kernel to finish.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
MTLLegacyGregoryPatchTable Class Reference
\n \n
\n \n-

#include <mtlLegacyGregoryPatchTable.h>

\n+

#include <mtlLegacyGregoryPatchTable.h>

\n
\n Inheritance diagram for MTLLegacyGregoryPatchTable:
\n
\n
\n \"\"/\n
\n \n@@ -121,15 +121,15 @@\n \n \n \n \n
static MTLLegacyGregoryPatchTable * Create (Far::PatchTable const *farPatchTable, DEVICE_CONTEXT context)
 
static MTLLegacyGregoryPatchTable * Create (Far::PatchTable const *farPatchTable, MTLContext *context)
 
\n

Detailed Description

\n
\n-

Definition at line 41 of file mtlLegacyGregoryPatchTable.h.

\n+

Definition at line 41 of file mtlLegacyGregoryPatchTable.h.

\n

Constructor & Destructor Documentation

\n \n

◆ ~MTLLegacyGregoryPatchTable()

\n \n
\n
\n \n@@ -175,15 +175,15 @@\n \n \n \n
\n inlinestatic
\n
\n \n-

Definition at line 48 of file mtlLegacyGregoryPatchTable.h.

\n+

Definition at line 48 of file mtlLegacyGregoryPatchTable.h.

\n \n
\n
\n \n

◆ Create() [2/2]

\n \n
\n@@ -239,15 +239,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 71 of file mtlLegacyGregoryPatchTable.h.

\n+

Definition at line 71 of file mtlLegacyGregoryPatchTable.h.

\n \n
\n
\n \n

◆ GetQuadOffsetsBuffer()

\n \n
\n@@ -266,15 +266,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 66 of file mtlLegacyGregoryPatchTable.h.

\n+

Definition at line 66 of file mtlLegacyGregoryPatchTable.h.

\n \n
\n \n \n

◆ GetVertexBuffer()

\n \n
\n@@ -293,15 +293,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 56 of file mtlLegacyGregoryPatchTable.h.

\n+

Definition at line 56 of file mtlLegacyGregoryPatchTable.h.

\n \n
\n \n \n

◆ GetVertexValenceBuffer()

\n \n
\n@@ -320,15 +320,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 61 of file mtlLegacyGregoryPatchTable.h.

\n+

Definition at line 61 of file mtlLegacyGregoryPatchTable.h.

\n \n
\n \n \n

◆ UpdateVertexBuffer()

\n \n
\n@@ -365,15 +365,15 @@\n \n \n
\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
MTLPatchTable Class Reference
\n \n
\n \n-

#include <mtlPatchTable.h>

\n+

#include <mtlPatchTable.h>

\n
\n Inheritance diagram for MTLPatchTable:
\n
\n
\n \"\"/\n
\n \n@@ -163,29 +163,29 @@\n \n \n \n \n
std::vector< id< MTLBuffer > > _fvarIndexBuffers
 
std::vector< id< MTLBuffer > > _fvarParamBuffers
 
\n

Detailed Description

\n
\n-

Definition at line 46 of file mtlPatchTable.h.

\n+

Definition at line 46 of file mtlPatchTable.h.

\n

Member Typedef Documentation

\n \n

◆ VertexBufferBinding

\n \n
\n
\n \n \n \n \n
typedef id<MTLBuffer> VertexBufferBinding
\n
\n \n-

Definition at line 48 of file mtlPatchTable.h.

\n+

Definition at line 48 of file mtlPatchTable.h.

\n \n
\n
\n

Constructor & Destructor Documentation

\n \n

◆ MTLPatchTable()

\n \n@@ -287,15 +287,15 @@\n \n \n inlinestatic \n \n \n
\n \n-

Definition at line 54 of file mtlPatchTable.h.

\n+

Definition at line 54 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ Create() [2/2]

\n \n
\n@@ -351,15 +351,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 69 of file mtlPatchTable.h.

\n+

Definition at line 69 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchIndexBuffer()

\n \n
\n@@ -379,15 +379,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 70 of file mtlPatchTable.h.

\n+

Definition at line 70 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ GetFVarPatchParamBuffer()

\n \n
\n@@ -407,15 +407,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 71 of file mtlPatchTable.h.

\n+

Definition at line 71 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ GetNumFVarChannels()

\n \n
\n@@ -434,15 +434,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 68 of file mtlPatchTable.h.

\n+

Definition at line 68 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ GetPatchArrays()

\n \n
\n@@ -461,15 +461,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 61 of file mtlPatchTable.h.

\n+

Definition at line 61 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ GetPatchIndexBuffer()

\n \n
\n@@ -488,15 +488,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 62 of file mtlPatchTable.h.

\n+

Definition at line 62 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ GetPatchParamBuffer()

\n \n
\n@@ -515,15 +515,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 63 of file mtlPatchTable.h.

\n+

Definition at line 63 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchArrays()

\n \n
\n@@ -542,15 +542,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 65 of file mtlPatchTable.h.

\n+

Definition at line 65 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ GetVaryingPatchIndexBuffer()

\n \n
\n@@ -569,15 +569,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 66 of file mtlPatchTable.h.

\n+

Definition at line 66 of file mtlPatchTable.h.

\n \n
\n \n

Member Data Documentation

\n \n

◆ _fvarIndexBuffers

\n \n@@ -594,15 +594,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 86 of file mtlPatchTable.h.

\n+

Definition at line 86 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ _fvarParamBuffers

\n \n
\n@@ -618,15 +618,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 87 of file mtlPatchTable.h.

\n+

Definition at line 87 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ _fvarPatchArrays

\n \n
\n@@ -642,15 +642,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 85 of file mtlPatchTable.h.

\n+

Definition at line 85 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ _indexBuffer

\n \n
\n@@ -666,15 +666,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 78 of file mtlPatchTable.h.

\n+

Definition at line 78 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ _patchArrays

\n \n
\n@@ -690,15 +690,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 76 of file mtlPatchTable.h.

\n+

Definition at line 76 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ _patchParamBuffer

\n \n
\n@@ -714,15 +714,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 79 of file mtlPatchTable.h.

\n+

Definition at line 79 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ _varyingPatchArrays

\n \n
\n@@ -738,15 +738,15 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 81 of file mtlPatchTable.h.

\n+

Definition at line 81 of file mtlPatchTable.h.

\n \n
\n \n \n

◆ _varyingPatchIndexBuffer

\n \n
\n@@ -762,20 +762,20 @@\n \n \n protected \n \n \n
\n \n-

Definition at line 83 of file mtlPatchTable.h.

\n+

Definition at line 83 of file mtlPatchTable.h.

\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
CPUMTLVertexBuffer Class Reference
\n \n
\n \n-

#include <mtlVertexBuffer.h>

\n+

#include <mtlVertexBuffer.h>

\n \n \n \n \n \n \n@@ -120,15 +120,15 @@\n \n \n \n \n

\n Public Member Functions

void UpdateData (const float *src, int startVertex, int numVertices, MTLContext *context)
 
int GetNumElements () const
 
 CPUMTLVertexBuffer (int numElements, int numVertices)
 
bool allocate (MTLContext *context)
 
\n

Detailed Description

\n
\n-

Definition at line 39 of file mtlVertexBuffer.h.

\n+

Definition at line 39 of file mtlVertexBuffer.h.

\n

Constructor & Destructor Documentation

\n \n

◆ CPUMTLVertexBuffer()

\n \n
\n
\n \n@@ -244,15 +244,15 @@\n \n \n \n
\n inline
\n
\n \n-

Definition at line 58 of file mtlVertexBuffer.h.

\n+

Definition at line 58 of file mtlVertexBuffer.h.

\n \n
\n
\n \n

◆ Create()

\n \n
\n@@ -313,15 +313,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 45 of file mtlVertexBuffer.h.

\n+

Definition at line 45 of file mtlVertexBuffer.h.

\n \n
\n
\n \n

◆ GetNumVertices()

\n \n
\n@@ -340,15 +340,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 50 of file mtlVertexBuffer.h.

\n+

Definition at line 50 of file mtlVertexBuffer.h.

\n \n
\n \n \n

◆ UpdateData()

\n \n
\n@@ -385,15 +385,15 @@\n \n \n
\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
\n
\n \n-

#include <tbbEvaluator.h>

\n+

#include <tbbEvaluator.h>

\n \n \n \n \n \n \n@@ -166,15 +166,15 @@\n \n \n \n \n

\n Static Public Member Functions

template<typename SRC_BUFFER , typename DST_BUFFER , typename STENCIL_TABLE >
static bool EvalStencils (SRC_BUFFER *srcBuffer, BufferDescriptor const &srcDesc, DST_BUFFER *dstBuffer, BufferDescriptor const &dstDesc, STENCIL_TABLE const *stencilTable, TbbEvaluator const *instance=NULL, void *deviceContext=NULL)
 Generic static eval stencils function. This function has a same signature as other device kernels have so that it can be called in the same way from OsdMesh template interface. More...
 
 
static void SetNumThreads (int numThreads)
 initialize tbb task schedular (optional: client may use tbb::task_scheduler_init) More...
 
\n

Detailed Description

\n
\n-

Definition at line 39 of file tbbEvaluator.h.

\n+

Definition at line 39 of file tbbEvaluator.h.

\n

Member Function Documentation

\n \n

◆ EvalPatches() [1/6]

\n \n
\n
\n \n@@ -710,15 +710,15 @@\n \n \n \n
patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch
instancenot used in the cpu evaluator
deviceContextnot used in the cpu evaluator
\n \n \n \n-

Definition at line 607 of file tbbEvaluator.h.

\n+

Definition at line 607 of file tbbEvaluator.h.

\n \n
\n
\n \n

◆ EvalPatches() [5/6]

\n \n
\n@@ -834,15 +834,15 @@\n patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 517 of file tbbEvaluator.h.

\n+

Definition at line 517 of file tbbEvaluator.h.

\n \n
\n
\n \n

◆ EvalPatches() [6/6]

\n \n
\n@@ -933,15 +933,15 @@\n patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 454 of file tbbEvaluator.h.

\n+

Definition at line 454 of file tbbEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [1/3]

\n \n
\n@@ -1106,15 +1106,15 @@\n fvarChannelface-varying channel\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 1188 of file tbbEvaluator.h.

\n+

Definition at line 1188 of file tbbEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [2/3]

\n \n
\n@@ -1237,15 +1237,15 @@\n fvarChannelface-varying channel\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 1100 of file tbbEvaluator.h.

\n+

Definition at line 1100 of file tbbEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesFaceVarying() [3/3]

\n \n
\n@@ -1340,15 +1340,15 @@\n fvarChannelface-varying channel\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 1034 of file tbbEvaluator.h.

\n+

Definition at line 1034 of file tbbEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [1/3]

\n \n
\n@@ -1506,15 +1506,15 @@\n patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 971 of file tbbEvaluator.h.

\n+

Definition at line 971 of file tbbEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [2/3]

\n \n
\n@@ -1630,15 +1630,15 @@\n patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 886 of file tbbEvaluator.h.

\n+

Definition at line 886 of file tbbEvaluator.h.

\n \n
\n \n \n

◆ EvalPatchesVarying() [3/3]

\n \n
\n@@ -1726,15 +1726,15 @@\n patchTableCpuPatchTable or equivalent XXX: currently Far::PatchTable can't be used due to interface mismatch\n instancenot used in the cpu evaluator\n deviceContextnot used in the cpu evaluator \n \n \n \n \n-

Definition at line 823 of file tbbEvaluator.h.

\n+

Definition at line 823 of file tbbEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [1/6]

\n \n
\n@@ -2328,15 +2328,15 @@\n stencilTableFar::StencilTable or equivalent\n instancenot used in the tbb kernel (declared as a typed pointer to prevent undesirable template resolution)\n deviceContextnot used in the tbb kernel \n \n \n \n \n-

Definition at line 303 of file tbbEvaluator.h.

\n+

Definition at line 303 of file tbbEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [5/6]

\n \n
\n@@ -2438,15 +2438,15 @@\n stencilTableFar::StencilTable or equivalent\n instancenot used in the tbb kernel (declared as a typed pointer to prevent undesirable template resolution)\n deviceContextnot used in the tbb kernel \n \n \n \n \n-

Definition at line 168 of file tbbEvaluator.h.

\n+

Definition at line 168 of file tbbEvaluator.h.

\n \n
\n \n \n

◆ EvalStencils() [6/6]

\n \n
\n@@ -2523,15 +2523,15 @@\n stencilTableFar::StencilTable or equivalent\n instancenot used in the tbb kernel (declared as a typed pointer to prevent undesirable template resolution)\n deviceContextnot used in the tbb kernel \n \n \n \n \n-

Definition at line 72 of file tbbEvaluator.h.

\n+

Definition at line 72 of file tbbEvaluator.h.

\n \n
\n \n \n

◆ SetNumThreads()

\n \n
\n@@ -2593,15 +2593,15 @@\n
\n

Other methods

\n
\n \n
\n \n
The documentation for this class was generated from the following file:\n \n \n \n
\n
    \n
  • OpenSubdiv
  • OPENSUBDIV_VERSION
  • Osd
  • TbbEvaluator
    • \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01333.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01333.html", "unified_diff": "@@ -358,15 +358,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 84 of file bilinearScheme.h.

\n+

Definition at line 84 of file bilinearScheme.h.

\n \n
\n \n \n

◆ assignCornerLimitMask() [2/4]

\n \n
\n@@ -396,15 +396,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 208 of file catmarkScheme.h.

\n+

Definition at line 208 of file catmarkScheme.h.

\n \n
\n \n \n

◆ assignCornerLimitMask() [3/4]

\n \n
\n@@ -514,15 +514,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 118 of file bilinearScheme.h.

\n+

Definition at line 118 of file bilinearScheme.h.

\n \n
\n \n \n

◆ assignCornerLimitTangentMasks() [2/4]

\n \n
\n@@ -600,15 +600,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 301 of file catmarkScheme.h.

\n+

Definition at line 301 of file catmarkScheme.h.

\n \n
\n \n \n

◆ assignCornerLimitTangentMasks() [4/4]

\n \n
\n@@ -806,15 +806,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 221 of file catmarkScheme.h.

\n+

Definition at line 221 of file catmarkScheme.h.

\n \n
\n \n \n

◆ assignCreaseLimitMask() [3/4]

\n \n
\n@@ -894,15 +894,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 97 of file bilinearScheme.h.

\n+

Definition at line 97 of file bilinearScheme.h.

\n \n
\n \n \n

◆ assignCreaseLimitTangentMasks() [1/4]

\n \n
\n@@ -992,15 +992,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 334 of file catmarkScheme.h.

\n+

Definition at line 334 of file catmarkScheme.h.

\n \n
\n \n \n

◆ assignCreaseLimitTangentMasks() [3/4]

\n \n
\n@@ -1092,15 +1092,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 143 of file bilinearScheme.h.

\n+

Definition at line 143 of file bilinearScheme.h.

\n \n
\n \n \n

◆ assignCreaseMaskForEdge() [1/2]

\n \n
\n@@ -1254,15 +1254,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 148 of file catmarkScheme.h.

\n+

Definition at line 148 of file catmarkScheme.h.

\n \n
\n \n \n

◆ assignCreaseMaskForVertex() [3/3]

\n \n
\n@@ -1372,15 +1372,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 106 of file bilinearScheme.h.

\n+

Definition at line 106 of file bilinearScheme.h.

\n \n
\n \n \n

◆ assignSmoothLimitMask() [3/4]

\n \n
\n@@ -1410,15 +1410,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 247 of file catmarkScheme.h.

\n+

Definition at line 247 of file catmarkScheme.h.

\n \n
\n \n \n

◆ assignSmoothLimitMask() [4/4]

\n \n
\n@@ -1534,15 +1534,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 152 of file bilinearScheme.h.

\n+

Definition at line 152 of file bilinearScheme.h.

\n \n
\n \n \n

◆ assignSmoothLimitTangentMasks() [3/4]

\n \n
\n@@ -1578,15 +1578,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 453 of file catmarkScheme.h.

\n+

Definition at line 453 of file catmarkScheme.h.

\n \n
\n \n \n

◆ assignSmoothLimitTangentMasks() [4/4]

\n \n
\n@@ -1660,15 +1660,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 71 of file catmarkScheme.h.

\n+

Definition at line 71 of file catmarkScheme.h.

\n \n
\n \n \n

◆ assignSmoothMaskForEdge() [2/3]

\n \n
\n@@ -1808,15 +1808,15 @@\n \n \n inlineprotected \n \n \n
\n \n-

Definition at line 173 of file catmarkScheme.h.

\n+

Definition at line 173 of file catmarkScheme.h.

\n \n
\n \n \n

◆ assignSmoothMaskForVertex() [3/3]

\n \n
\n@@ -1888,15 +1888,15 @@\n \n )\n const\n \n \n
\n \n-

Definition at line 62 of file bilinearScheme.h.

\n+

Definition at line 62 of file bilinearScheme.h.

\n \n
\n \n \n

◆ ComputeEdgeVertexMask() [2/2]

\n \n
\n@@ -2095,15 +2095,15 @@\n \n )\n const\n \n \n
\n \n-

Definition at line 71 of file bilinearScheme.h.

\n+

Definition at line 71 of file bilinearScheme.h.

\n \n
\n \n \n

◆ ComputeVertexVertexMask() [2/2]

\n \n
\n@@ -2166,15 +2166,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 53 of file bilinearScheme.h.

\n+

Definition at line 53 of file bilinearScheme.h.

\n \n
\n \n \n

◆ GetLocalNeighborhoodSize() [2/4]

\n \n
\n@@ -2193,15 +2193,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 56 of file catmarkScheme.h.

\n+

Definition at line 56 of file catmarkScheme.h.

\n \n
\n \n \n

◆ GetLocalNeighborhoodSize() [3/4]

\n \n
\n@@ -2299,15 +2299,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 47 of file bilinearScheme.h.

\n+

Definition at line 47 of file bilinearScheme.h.

\n \n
\n \n \n

◆ GetRegularFaceSize() [2/4]

\n \n
\n@@ -2326,15 +2326,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 50 of file catmarkScheme.h.

\n+

Definition at line 50 of file catmarkScheme.h.

\n \n
\n \n \n

◆ GetRegularFaceSize() [3/4]

\n \n
\n@@ -2405,15 +2405,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 50 of file bilinearScheme.h.

\n+

Definition at line 50 of file bilinearScheme.h.

\n \n
\n \n \n

◆ GetRegularVertexValence() [2/4]

\n \n
\n@@ -2432,15 +2432,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 53 of file catmarkScheme.h.

\n+

Definition at line 53 of file catmarkScheme.h.

\n \n
\n \n \n

◆ GetRegularVertexValence() [3/4]

\n \n
\n@@ -2511,15 +2511,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 44 of file bilinearScheme.h.

\n+

Definition at line 44 of file bilinearScheme.h.

\n \n
\n \n \n

◆ GetTopologicalSplitType() [2/4]

\n \n
\n@@ -2538,15 +2538,15 @@\n \n \n inline \n \n \n
\n \n-

Definition at line 47 of file catmarkScheme.h.

\n+

Definition at line 47 of file catmarkScheme.h.

\n \n
\n \n \n

◆ GetTopologicalSplitType() [3/4]

\n \n
\n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/a01481_source.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/a01481_source.html", "unified_diff": "@@ -217,17 +217,17 @@\n
140} // end namespace OpenSubdiv
\n
141
\n
142
\n
143#endif // OPENSUBDIV3_OSD_TYPES_H
\n \n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchParamVector
std::vector< PatchParam > PatchParamVector
Definition: types.h:133
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArrayVector
std::vector< PatchArray > PatchArrayVector
Definition: types.h:132
\n-\n-
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetType
Type GetType() const
Returns the type of the patch.
Definition: patchDescriptor.h:87
\n-\n+\n+
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::GetType
Type GetType() const
Returns the type of the patch.
Definition: patchDescriptor.h:87
\n+\n
OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::PatchHandle
Handle that can be used as unique patch identifier within PatchTable.
Definition: patchTable.h:60
\n \n \n \n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord
Coordinates set on a patch table.
Definition: types.h:42
\n
OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord::handle
Far::PatchTable::PatchHandle handle
patch handle
Definition: types.h:62
\n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/dir_2483050b711c1829a6434006d0beb10c.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/dir_2483050b711c1829a6434006d0beb10c.html", "unified_diff": "@@ -86,91 +86,91 @@\n
\n
osd Directory Reference
\n
\n
\n \n \n-\n+\n \n-\n+\n \n-\n+\n \n \n \n-\n+\n \n-\n+\n \n \n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n \n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n \n \n \n \n-\n+\n \n-\n+\n \n-\n+\n \n \n \n-\n+\n \n-\n+\n \n \n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n \n \n

\n Files

file  bufferDescriptor.h [code]
file  bufferDescriptor.h [code]
 
file  clEvaluator.h [code]
file  clEvaluator.h [code]
 
file  clPatchTable.h [code]
file  clPatchTable.h [code]
 
file  clVertexBuffer.h [code]
 
file  cpuD3D11VertexBuffer.h [code]
file  cpuD3D11VertexBuffer.h [code]
 
file  cpuEvaluator.h [code]
file  cpuEvaluator.h [code]
 
file  cpuGLVertexBuffer.h [code]
 
file  cpuPatchTable.h [code]
file  cpuPatchTable.h [code]
 
file  cpuVertexBuffer.h [code]
file  cpuVertexBuffer.h [code]
 
file  cudaEvaluator.h [code]
file  cudaEvaluator.h [code]
 
file  cudaPatchTable.h [code]
file  cudaPatchTable.h [code]
 
file  cudaVertexBuffer.h [code]
 
file  d3d11ComputeEvaluator.h [code]
file  d3d11ComputeEvaluator.h [code]
 
file  d3d11LegacyGregoryPatchTable.h [code]
file  d3d11LegacyGregoryPatchTable.h [code]
 
file  d3d11Mesh.h [code]
file  d3d11Mesh.h [code]
 
file  d3d11PatchTable.h [code]
file  d3d11PatchTable.h [code]
 
file  d3d11VertexBuffer.h [code]
file  d3d11VertexBuffer.h [code]
 
file  glComputeEvaluator.h [code]
file  glComputeEvaluator.h [code]
 
file  glLegacyGregoryPatchTable.h [code]
file  glLegacyGregoryPatchTable.h [code]
 
file  glMesh.h [code]
file  glMesh.h [code]
 
file  glPatchTable.h [code]
file  glPatchTable.h [code]
 
file  glslPatchShaderSource.h [code]
file  glslPatchShaderSource.h [code]
 
file  glVertexBuffer.h [code]
file  glVertexBuffer.h [code]
 
file  glXFBEvaluator.h [code]
file  glXFBEvaluator.h [code]
 
file  hlslPatchShaderSource.h [code]
file  hlslPatchShaderSource.h [code]
 
file  mesh.h [code]
 
file  mtlCommon.h [code]
 
file  mtlComputeEvaluator.h [code]
file  mtlComputeEvaluator.h [code]
 
file  mtlLegacyGregoryPatchTable.h [code]
file  mtlLegacyGregoryPatchTable.h [code]
 
file  mtlMesh.h [code]
file  mtlMesh.h [code]
 
file  mtlPatchShaderSource.h [code]
 
file  mtlPatchTable.h [code]
file  mtlPatchTable.h [code]
 
file  mtlVertexBuffer.h [code]
file  mtlVertexBuffer.h [code]
 
file  ompEvaluator.h [code]
 
file  ompKernel.h [code]
file  ompKernel.h [code]
 
file  opencl.h [code]
file  opencl.h [code]
 
file  opengl.h [code]
file  opengl.h [code]
 
file  tbbEvaluator.h [code]
file  tbbEvaluator.h [code]
 
file  tbbKernel.h [code]
file  tbbKernel.h [code]
 
file  types.h [code]
 
\n
\n
\n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/dir_2483050b711c1829a6434006d0beb10c.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/dir_2483050b711c1829a6434006d0beb10c.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,42 +1,42 @@\n var dir_2483050b711c1829a6434006d0beb10c = [\n- [\"bufferDescriptor.h\", \"a00896.html\", \"a00896\"],\n- [\"clEvaluator.h\", \"a00800.html\", \"a00800\"],\n- [\"clPatchTable.h\", \"a00803.html\", \"a00803\"],\n+ [\"bufferDescriptor.h\", \"a00899.html\", \"a00899\"],\n+ [\"clEvaluator.h\", \"a00803.html\", \"a00803\"],\n+ [\"clPatchTable.h\", \"a00800.html\", \"a00800\"],\n [\"clVertexBuffer.h\", \"a00797.html\", \"a00797\"],\n- [\"cpuD3D11VertexBuffer.h\", \"a00842.html\", \"a00842\"],\n- [\"cpuEvaluator.h\", \"a00899.html\", \"a00899\"],\n+ [\"cpuD3D11VertexBuffer.h\", \"a00848.html\", \"a00848\"],\n+ [\"cpuEvaluator.h\", \"a00896.html\", \"a00896\"],\n [\"cpuGLVertexBuffer.h\", \"a00872.html\", \"a00872\"],\n- [\"cpuPatchTable.h\", \"a00890.html\", \"a00890\"],\n- [\"cpuVertexBuffer.h\", \"a00893.html\", \"a00893\"],\n- [\"cudaEvaluator.h\", \"a00791.html\", \"a00791\"],\n- [\"cudaPatchTable.h\", \"a00788.html\", \"a00788\"],\n+ [\"cpuPatchTable.h\", \"a00893.html\", \"a00893\"],\n+ [\"cpuVertexBuffer.h\", \"a00890.html\", \"a00890\"],\n+ [\"cudaEvaluator.h\", \"a00794.html\", \"a00794\"],\n+ [\"cudaPatchTable.h\", \"a00791.html\", \"a00791\"],\n [\"cudaVertexBuffer.h\", \"a00902.html\", \"a00902\"],\n- [\"d3d11ComputeEvaluator.h\", \"a00839.html\", \"a00839\"],\n- [\"d3d11LegacyGregoryPatchTable.h\", \"a00836.html\", \"a00836\"],\n- [\"d3d11Mesh.h\", \"a00827.html\", \"a00827\"],\n- [\"d3d11PatchTable.h\", \"a00845.html\", \"a00845\"],\n- [\"d3d11VertexBuffer.h\", \"a00830.html\", \"a00830\"],\n- [\"glComputeEvaluator.h\", \"a00848.html\", \"a00848\"],\n- [\"glLegacyGregoryPatchTable.h\", \"a00866.html\", \"a00866\"],\n- [\"glMesh.h\", \"a00854.html\", \"a00854\"],\n- [\"glPatchTable.h\", \"a00863.html\", \"a00863\"],\n- [\"glslPatchShaderSource.h\", \"a00860.html\", \"a00860\"],\n- [\"glVertexBuffer.h\", \"a00857.html\", \"a00857\"],\n- [\"glXFBEvaluator.h\", \"a00851.html\", \"a00851\"],\n- [\"hlslPatchShaderSource.h\", \"a00833.html\", \"a00833\"],\n+ [\"d3d11ComputeEvaluator.h\", \"a00842.html\", \"a00842\"],\n+ [\"d3d11LegacyGregoryPatchTable.h\", \"a00839.html\", \"a00839\"],\n+ [\"d3d11Mesh.h\", \"a00830.html\", \"a00830\"],\n+ [\"d3d11PatchTable.h\", \"a00836.html\", \"a00836\"],\n+ [\"d3d11VertexBuffer.h\", \"a00833.html\", \"a00833\"],\n+ [\"glComputeEvaluator.h\", \"a00851.html\", \"a00851\"],\n+ [\"glLegacyGregoryPatchTable.h\", \"a00863.html\", \"a00863\"],\n+ [\"glMesh.h\", \"a00857.html\", \"a00857\"],\n+ [\"glPatchTable.h\", \"a00860.html\", \"a00860\"],\n+ [\"glslPatchShaderSource.h\", \"a00854.html\", \"a00854\"],\n+ [\"glVertexBuffer.h\", \"a00866.html\", \"a00866\"],\n+ [\"glXFBEvaluator.h\", \"a00845.html\", \"a00845\"],\n+ [\"hlslPatchShaderSource.h\", \"a00827.html\", \"a00827\"],\n [\"mesh.h\", \"a00887.html\", \"a00887\"],\n [\"mtlCommon.h\", \"a00806.html\", \"a00806\"],\n- [\"mtlComputeEvaluator.h\", \"a00824.html\", \"a00824\"],\n- [\"mtlLegacyGregoryPatchTable.h\", \"a00818.html\", \"a00818\"],\n- [\"mtlMesh.h\", \"a00812.html\", \"a00812\"],\n+ [\"mtlComputeEvaluator.h\", \"a00821.html\", \"a00821\"],\n+ [\"mtlLegacyGregoryPatchTable.h\", \"a00815.html\", \"a00815\"],\n+ [\"mtlMesh.h\", \"a00818.html\", \"a00818\"],\n [\"mtlPatchShaderSource.h\", \"a00809.html\", \"a00809\"],\n- [\"mtlPatchTable.h\", \"a00815.html\", \"a00815\"],\n- [\"mtlVertexBuffer.h\", \"a00821.html\", \"a00821\"],\n+ [\"mtlPatchTable.h\", \"a00812.html\", \"a00812\"],\n+ [\"mtlVertexBuffer.h\", \"a00824.html\", \"a00824\"],\n [\"ompEvaluator.h\", \"a00878.html\", \"a00878\"],\n- [\"ompKernel.h\", \"a00881.html\", \"a00881\"],\n- [\"opencl.h\", \"a00794.html\", null],\n- [\"opengl.h\", \"a00884.html\", null],\n- [\"tbbEvaluator.h\", \"a00875.html\", \"a00875\"],\n- [\"tbbKernel.h\", \"a00869.html\", \"a00869\"],\n+ [\"ompKernel.h\", \"a00884.html\", \"a00884\"],\n+ [\"opencl.h\", \"a00788.html\", null],\n+ [\"opengl.h\", \"a00881.html\", null],\n+ [\"tbbEvaluator.h\", \"a00869.html\", \"a00869\"],\n+ [\"tbbKernel.h\", \"a00875.html\", \"a00875\"],\n [\"types.h\", \"a01481.html\", \"a01481\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/dir_7f420b72a822828cfa8b0561c695a5cc.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/dir_7f420b72a822828cfa8b0561c695a5cc.html", "unified_diff": "@@ -88,35 +88,35 @@\n \n
\n \n \n \n \n-\n+\n \n \n \n-\n+\n \n \n \n \n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n-\n+\n \n \n \n \n \n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/dir_7f420b72a822828cfa8b0561c695a5cc.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/dir_7f420b72a822828cfa8b0561c695a5cc.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,17 +1,17 @@\n var dir_7f420b72a822828cfa8b0561c695a5cc = [\n [\"error.h\", \"a00743.html\", \"a00743\"],\n- [\"patchDescriptor.h\", \"a00740.html\", \"a00740\"],\n+ [\"patchDescriptor.h\", \"a00737.html\", \"a00737\"],\n [\"patchMap.h\", \"a00734.html\", \"a00734\"],\n- [\"patchParam.h\", \"a00737.html\", \"a00737\"],\n+ [\"patchParam.h\", \"a00740.html\", \"a00740\"],\n [\"patchTable.h\", \"a00731.html\", \"a00731\"],\n [\"patchTableFactory.h\", \"a00728.html\", \"a00728\"],\n- [\"primvarRefiner.h\", \"a00719.html\", \"a00719\"],\n- [\"ptexIndices.h\", \"a00716.html\", \"a00716\"],\n- [\"stencilTable.h\", \"a00725.html\", \"a00725\"],\n- [\"stencilTableFactory.h\", \"a00710.html\", \"a00710\"],\n- [\"topologyDescriptor.h\", \"a00713.html\", \"a00713\"],\n- [\"topologyLevel.h\", \"a00722.html\", \"a00722\"],\n+ [\"primvarRefiner.h\", \"a00725.html\", \"a00725\"],\n+ [\"ptexIndices.h\", \"a00722.html\", \"a00722\"],\n+ [\"stencilTable.h\", \"a00713.html\", \"a00713\"],\n+ [\"stencilTableFactory.h\", \"a00716.html\", \"a00716\"],\n+ [\"topologyDescriptor.h\", \"a00710.html\", \"a00710\"],\n+ [\"topologyLevel.h\", \"a00719.html\", \"a00719\"],\n [\"topologyRefiner.h\", \"a00707.html\", \"a00707\"],\n [\"topologyRefinerFactory.h\", \"a00704.html\", \"a00704\"],\n [\"types.h\", \"a01478.html\", \"a01478\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/dir_bd5473d7d5841d48919aaa9ad0cb1f9b.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/dir_bd5473d7d5841d48919aaa9ad0cb1f9b.html", "unified_diff": "@@ -94,17 +94,17 @@\n \n \n \n \n \n \n \n-\n+\n \n-\n+\n \n \n \n \n \n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/dir_bd5473d7d5841d48919aaa9ad0cb1f9b.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/dir_bd5473d7d5841d48919aaa9ad0cb1f9b.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,11 +1,11 @@\n var dir_bd5473d7d5841d48919aaa9ad0cb1f9b = [\n [\"array.h\", \"a00764.html\", \"a00764\"],\n [\"componentInterfaces.h\", \"a00767.html\", null],\n [\"fvarLevel.h\", \"a00761.html\", null],\n [\"fvarRefinement.h\", \"a00758.html\", null],\n- [\"level.h\", \"a00752.html\", null],\n- [\"refinement.h\", \"a00755.html\", null],\n+ [\"level.h\", \"a00755.html\", null],\n+ [\"refinement.h\", \"a00752.html\", null],\n [\"sparseSelector.h\", \"a00749.html\", null],\n [\"stackBuffer.h\", \"a00746.html\", null],\n [\"types.h\", \"a01487.html\", \"a01487\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/dir_f61368944c63c2c1f8c69bc232e59c39.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/dir_f61368944c63c2c1f8c69bc232e59c39.html", "unified_diff": "@@ -86,17 +86,17 @@\n
\n
sdc Directory Reference
\n
\n
\n

\n Files

file  error.h [code]
 
file  patchDescriptor.h [code]
file  patchDescriptor.h [code]
 
file  patchMap.h [code]
 
file  patchParam.h [code]
file  patchParam.h [code]
 
file  patchTable.h [code]
 
file  patchTableFactory.h [code]
 
file  primvarRefiner.h [code]
file  primvarRefiner.h [code]
 
file  ptexIndices.h [code]
file  ptexIndices.h [code]
 
file  stencilTable.h [code]
file  stencilTable.h [code]
 
file  stencilTableFactory.h [code]
file  stencilTableFactory.h [code]
 
file  topologyDescriptor.h [code]
file  topologyDescriptor.h [code]
 
file  topologyLevel.h [code]
file  topologyLevel.h [code]
 
file  topologyRefiner.h [code]
 
file  topologyRefinerFactory.h [code]
 
file  types.h [code]
 
 
file  componentInterfaces.h [code]
 
file  fvarLevel.h [code]
 
file  fvarRefinement.h [code]
 
file  level.h [code]
file  level.h [code]
 
file  refinement.h [code]
file  refinement.h [code]
 
file  sparseSelector.h [code]
 
file  stackBuffer.h [code]
 
file  types.h [code]
 
\n \n-\n+\n \n-\n+\n \n \n \n \n \n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/dir_f61368944c63c2c1f8c69bc232e59c39.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/dir_f61368944c63c2c1f8c69bc232e59c39.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,9 +1,9 @@\n var dir_f61368944c63c2c1f8c69bc232e59c39 = [\n- [\"bilinearScheme.h\", \"a00782.html\", null],\n- [\"catmarkScheme.h\", \"a00785.html\", null],\n+ [\"bilinearScheme.h\", \"a00785.html\", null],\n+ [\"catmarkScheme.h\", \"a00782.html\", null],\n [\"crease.h\", \"a00779.html\", \"a00779\"],\n [\"loopScheme.h\", \"a00776.html\", null],\n [\"options.h\", \"a00773.html\", \"a00773\"],\n [\"scheme.h\", \"a00770.html\", \"a00770\"],\n [\"types.h\", \"a01484.html\", \"a01484\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/dir_fa5edef2a780a73fbfb6a11f1af88ceb.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/dir_fa5edef2a780a73fbfb6a11f1af88ceb.html", "unified_diff": "@@ -90,17 +90,17 @@\n

\n Files

file  bilinearScheme.h [code]
file  bilinearScheme.h [code]
 
file  catmarkScheme.h [code]
file  catmarkScheme.h [code]
 
file  crease.h [code]
 
file  loopScheme.h [code]
 
file  options.h [code]
 
\n \n \n \n \n \n-\n+\n \n-\n+\n \n \n \n \n \n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/dir_fa5edef2a780a73fbfb6a11f1af88ceb.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/dir_fa5edef2a780a73fbfb6a11f1af88ceb.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,11 +1,11 @@\n var dir_fa5edef2a780a73fbfb6a11f1af88ceb = [\n [\"limits.h\", \"a00695.html\", \"a00695\"],\n [\"parameterization.h\", \"a00698.html\", \"a00698\"],\n- [\"refinerSurfaceFactory.h\", \"a00689.html\", \"a00689\"],\n- [\"surface.h\", \"a00692.html\", \"a00692\"],\n+ [\"refinerSurfaceFactory.h\", \"a00692.html\", \"a00692\"],\n+ [\"surface.h\", \"a00689.html\", \"a00689\"],\n [\"surfaceFactory.h\", \"a00686.html\", \"a00686\"],\n [\"surfaceFactoryCache.h\", \"a00680.html\", \"a00680\"],\n [\"surfaceFactoryMeshAdapter.h\", \"a00683.html\", \"a00683\"],\n [\"tessellation.h\", \"a00677.html\", \"a00677\"],\n [\"vertexDescriptor.h\", \"a00674.html\", \"a00674\"]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/files.html", "source2": "./usr/share/doc/opensubdiv/doxy_html/files.html", "unified_diff": "@@ -89,93 +89,93 @@\n
\n
Here is a list of all files with brief descriptions:
\n
[detail level 123]

\n Files

file  limits.h [code]
 
file  parameterization.h [code]
 
file  refinerSurfaceFactory.h [code]
file  refinerSurfaceFactory.h [code]
 
file  surface.h [code]
file  surface.h [code]
 
file  surfaceFactory.h [code]
 
file  surfaceFactoryCache.h [code]
 
file  surfaceFactoryMeshAdapter.h [code]
 
\n \n \n \n \n-\n-\n+\n+\n \n \n \n \n \n \n \n-\n+\n \n-\n+\n \n \n-\n-\n-\n-\n-\n-\n+\n+\n+\n+\n+\n+\n \n \n \n \n-\n-\n-\n+\n+\n+\n \n-\n-\n+\n+\n \n-\n-\n-\n-\n+\n+\n+\n+\n \n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n-\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n+\n \n \n-\n-\n-\n+\n+\n+\n \n-\n-\n+\n+\n \n-\n-\n-\n-\n-\n+\n+\n+\n+\n+\n \n \n-\n-\n+\n+\n \n \n \n \n \n \n \n \n \n \n-\n-\n+\n+\n \n \n \n
 ▼ opensubdiv
 ▼ bfr
 limits.h
 parameterization.h
 refinerSurfaceFactory.h
 surface.h
 refinerSurfaceFactory.h
 surface.h
 surfaceFactory.h
 surfaceFactoryCache.h
 surfaceFactoryMeshAdapter.h
 tessellation.h
 vertexDescriptor.h
 ▼ far
 error.h
 patchDescriptor.h
 patchDescriptor.h
 patchMap.h
 patchParam.h
 patchParam.h
 patchTable.h
 patchTableFactory.h
 primvarRefiner.h
 ptexIndices.h
 stencilTable.h
 stencilTableFactory.h
 topologyDescriptor.h
 topologyLevel.h
 primvarRefiner.h
 ptexIndices.h
 stencilTable.h
 stencilTableFactory.h
 topologyDescriptor.h
 topologyLevel.h
 topologyRefiner.h
 topologyRefinerFactory.h
 types.h
 ▼ osd
 bufferDescriptor.h
 clEvaluator.h
 clPatchTable.h
 bufferDescriptor.h
 clEvaluator.h
 clPatchTable.h
 clVertexBuffer.h
 cpuD3D11VertexBuffer.h
 cpuEvaluator.h
 cpuD3D11VertexBuffer.h
 cpuEvaluator.h
 cpuGLVertexBuffer.h
 cpuPatchTable.h
 cpuVertexBuffer.h
 cudaEvaluator.h
 cudaPatchTable.h
 cpuPatchTable.h
 cpuVertexBuffer.h
 cudaEvaluator.h
 cudaPatchTable.h
 cudaVertexBuffer.h
 d3d11ComputeEvaluator.h
 d3d11LegacyGregoryPatchTable.h
 d3d11Mesh.h
 d3d11PatchTable.h
 d3d11VertexBuffer.h
 glComputeEvaluator.h
 glLegacyGregoryPatchTable.h
 glMesh.h
 glPatchTable.h
 glslPatchShaderSource.h
 glVertexBuffer.h
 glXFBEvaluator.h
 hlslPatchShaderSource.h
 d3d11ComputeEvaluator.h
 d3d11LegacyGregoryPatchTable.h
 d3d11Mesh.h
 d3d11PatchTable.h
 d3d11VertexBuffer.h
 glComputeEvaluator.h
 glLegacyGregoryPatchTable.h
 glMesh.h
 glPatchTable.h
 glslPatchShaderSource.h
 glVertexBuffer.h
 glXFBEvaluator.h
 hlslPatchShaderSource.h
 mesh.h
 mtlCommon.h
 mtlComputeEvaluator.h
 mtlLegacyGregoryPatchTable.h
 mtlMesh.h
 mtlComputeEvaluator.h
 mtlLegacyGregoryPatchTable.h
 mtlMesh.h
 mtlPatchShaderSource.h
 mtlPatchTable.h
 mtlVertexBuffer.h
 mtlPatchTable.h
 mtlVertexBuffer.h
 ompEvaluator.h
 ompKernel.h
 opencl.h
 opengl.h
 tbbEvaluator.h
 tbbKernel.h
 ompKernel.h
 opencl.h
 opengl.h
 tbbEvaluator.h
 tbbKernel.h
 types.h
 ▼ sdc
 bilinearScheme.h
 catmarkScheme.h
 bilinearScheme.h
 catmarkScheme.h
 crease.h
 loopScheme.h
 options.h
 scheme.h
 types.h
 ▼ vtr
 array.h
 componentInterfaces.h
 fvarLevel.h
 fvarRefinement.h
 level.h
 refinement.h
 level.h
 refinement.h
 sparseSelector.h
 stackBuffer.h
 types.h
\n
\n \n \n"}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/navtreeindex0.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/navtreeindex0.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -5,52 +5,52 @@\n \"a00677_source.html\": [2, 0, 0, 0, 7],\n \"a00680.html\": [2, 0, 0, 0, 5],\n \"a00680_source.html\": [2, 0, 0, 0, 5],\n \"a00683.html\": [2, 0, 0, 0, 6],\n \"a00683_source.html\": [2, 0, 0, 0, 6],\n \"a00686.html\": [2, 0, 0, 0, 4],\n \"a00686_source.html\": [2, 0, 0, 0, 4],\n- \"a00689.html\": [2, 0, 0, 0, 2],\n- \"a00689_source.html\": [2, 0, 0, 0, 2],\n- \"a00692.html\": [2, 0, 0, 0, 3],\n- \"a00692_source.html\": [2, 0, 0, 0, 3],\n+ \"a00689.html\": [2, 0, 0, 0, 3],\n+ \"a00689_source.html\": [2, 0, 0, 0, 3],\n+ \"a00692.html\": [2, 0, 0, 0, 2],\n+ \"a00692_source.html\": [2, 0, 0, 0, 2],\n \"a00695.html\": [2, 0, 0, 0, 0],\n \"a00695_source.html\": [2, 0, 0, 0, 0],\n \"a00698.html\": [2, 0, 0, 0, 1],\n \"a00698_source.html\": [2, 0, 0, 0, 1],\n \"a00704.html\": [2, 0, 0, 1, 13],\n \"a00704_source.html\": [2, 0, 0, 1, 13],\n \"a00707.html\": [2, 0, 0, 1, 12],\n \"a00707_source.html\": [2, 0, 0, 1, 12],\n- \"a00710.html\": [2, 0, 0, 1, 9],\n- \"a00710_source.html\": [2, 0, 0, 1, 9],\n- \"a00713.html\": [2, 0, 0, 1, 10],\n- \"a00713_source.html\": [2, 0, 0, 1, 10],\n- \"a00716.html\": [2, 0, 0, 1, 7],\n- \"a00716_source.html\": [2, 0, 0, 1, 7],\n- \"a00719.html\": [2, 0, 0, 1, 6],\n- \"a00719_source.html\": [2, 0, 0, 1, 6],\n- \"a00722.html\": [2, 0, 0, 1, 11],\n- \"a00722_source.html\": [2, 0, 0, 1, 11],\n- \"a00725.html\": [2, 0, 0, 1, 8],\n- \"a00725_source.html\": [2, 0, 0, 1, 8],\n+ \"a00710.html\": [2, 0, 0, 1, 10],\n+ \"a00710_source.html\": [2, 0, 0, 1, 10],\n+ \"a00713.html\": [2, 0, 0, 1, 8],\n+ \"a00713_source.html\": [2, 0, 0, 1, 8],\n+ \"a00716.html\": [2, 0, 0, 1, 9],\n+ \"a00716_source.html\": [2, 0, 0, 1, 9],\n+ \"a00719.html\": [2, 0, 0, 1, 11],\n+ \"a00719_source.html\": [2, 0, 0, 1, 11],\n+ \"a00722.html\": [2, 0, 0, 1, 7],\n+ \"a00722_source.html\": [2, 0, 0, 1, 7],\n+ \"a00725.html\": [2, 0, 0, 1, 6],\n+ \"a00725_source.html\": [2, 0, 0, 1, 6],\n \"a00728.html\": [2, 0, 0, 1, 5],\n \"a00728_source.html\": [2, 0, 0, 1, 5],\n \"a00731.html\": [2, 0, 0, 1, 4],\n \"a00731_source.html\": [2, 0, 0, 1, 4],\n \"a00734.html\": [2, 0, 0, 1, 2],\n \"a00734_source.html\": [2, 0, 0, 1, 2],\n- \"a00737.html\": [2, 0, 0, 1, 3],\n- \"a00737.html#a3139c7c52ddf26b2403e8e95ff63eeb3\": [2, 0, 0, 1, 3, 1],\n- \"a00737.html#a46ea68db89cc3d55c4191ef7648281ce\": [2, 0, 0, 1, 3, 3],\n- \"a00737.html#abe07210a029a01571c54276ae28b4d88\": [2, 0, 0, 1, 3, 2],\n- \"a00737_source.html\": [2, 0, 0, 1, 3],\n- \"a00740.html\": [2, 0, 0, 1, 1],\n- \"a00740.html#ad258c2d8f7871bc0d40cf3a9ded6e08a\": [2, 0, 0, 1, 1, 1],\n- \"a00740_source.html\": [2, 0, 0, 1, 1],\n+ \"a00737.html\": [2, 0, 0, 1, 1],\n+ \"a00737.html#ad258c2d8f7871bc0d40cf3a9ded6e08a\": [2, 0, 0, 1, 1, 1],\n+ \"a00737_source.html\": [2, 0, 0, 1, 1],\n+ \"a00740.html\": [2, 0, 0, 1, 3],\n+ \"a00740.html#a3139c7c52ddf26b2403e8e95ff63eeb3\": [2, 0, 0, 1, 3, 1],\n+ \"a00740.html#a46ea68db89cc3d55c4191ef7648281ce\": [2, 0, 0, 1, 3, 3],\n+ \"a00740.html#abe07210a029a01571c54276ae28b4d88\": [2, 0, 0, 1, 3, 2],\n+ \"a00740_source.html\": [2, 0, 0, 1, 3],\n \"a00743.html\": [2, 0, 0, 1, 0],\n \"a00743.html#a25d51c3a61fb3dacded6e48d01fd2fdb\": [2, 0, 0, 1, 0, 3],\n \"a00743.html#a2bd300390ebc686e71a3690f4a0b1726\": [2, 0, 0, 1, 0, 1],\n \"a00743.html#a3ba4fd4ed534a5c9c32ad01fa88bb8d9\": [2, 0, 0, 1, 0, 4],\n \"a00743.html#a73bbe6b41209e4c34234f192f0193a63\": [2, 0, 0, 1, 0, 5],\n \"a00743.html#a97d41b6faaf421937e6b82bd1c3067db\": [2, 0, 0, 1, 0, 6],\n \"a00743.html#aa1630f297a91dfc683dfcfddeeebb2ac\": [2, 0, 0, 1, 0, 0],\n@@ -61,18 +61,18 @@\n \"a00743.html#ab0df38968e4f03a3f1f6d6df0f31f45aabeca977bc439b451aea1325f48a52a0b\": [2, 0, 0, 1, 0, 2, 0],\n \"a00743.html#ab0df38968e4f03a3f1f6d6df0f31f45aac565fa39d681e2766584dd87246c3bdd\": [2, 0, 0, 1, 0, 2, 4],\n \"a00743_source.html\": [2, 0, 0, 1, 0],\n \"a00746.html\": [2, 0, 0, 4, 7],\n \"a00746_source.html\": [2, 0, 0, 4, 7],\n \"a00749.html\": [2, 0, 0, 4, 6],\n \"a00749_source.html\": [2, 0, 0, 4, 6],\n- \"a00752.html\": [2, 0, 0, 4, 4],\n- \"a00752_source.html\": [2, 0, 0, 4, 4],\n- \"a00755.html\": [2, 0, 0, 4, 5],\n- \"a00755_source.html\": [2, 0, 0, 4, 5],\n+ \"a00752.html\": [2, 0, 0, 4, 5],\n+ \"a00752_source.html\": [2, 0, 0, 4, 5],\n+ \"a00755.html\": [2, 0, 0, 4, 4],\n+ \"a00755_source.html\": [2, 0, 0, 4, 4],\n \"a00758.html\": [2, 0, 0, 4, 3],\n \"a00758_source.html\": [2, 0, 0, 4, 3],\n \"a00761.html\": [2, 0, 0, 4, 2],\n \"a00761_source.html\": [2, 0, 0, 4, 2],\n \"a00764.html\": [2, 0, 0, 4, 0],\n \"a00764_source.html\": [2, 0, 0, 4, 0],\n \"a00767.html\": [2, 0, 0, 4, 1],\n@@ -81,95 +81,95 @@\n \"a00770_source.html\": [2, 0, 0, 3, 5],\n \"a00773.html\": [2, 0, 0, 3, 4],\n \"a00773_source.html\": [2, 0, 0, 3, 4],\n \"a00776.html\": [2, 0, 0, 3, 3],\n \"a00776_source.html\": [2, 0, 0, 3, 3],\n \"a00779.html\": [2, 0, 0, 3, 2],\n \"a00779_source.html\": [2, 0, 0, 3, 2],\n- \"a00782.html\": [2, 0, 0, 3, 0],\n- \"a00782_source.html\": [2, 0, 0, 3, 0],\n- \"a00785.html\": [2, 0, 0, 3, 1],\n- \"a00785_source.html\": [2, 0, 0, 3, 1],\n- \"a00788.html\": [2, 0, 0, 2, 10],\n- \"a00788_source.html\": [2, 0, 0, 2, 10],\n- \"a00791.html\": [2, 0, 0, 2, 9],\n- \"a00791_source.html\": [2, 0, 0, 2, 9],\n- \"a00794.html\": [2, 0, 0, 2, 35],\n- \"a00794_source.html\": [2, 0, 0, 2, 35],\n+ \"a00782.html\": [2, 0, 0, 3, 1],\n+ \"a00782_source.html\": [2, 0, 0, 3, 1],\n+ \"a00785.html\": [2, 0, 0, 3, 0],\n+ \"a00785_source.html\": [2, 0, 0, 3, 0],\n+ \"a00788.html\": [2, 0, 0, 2, 35],\n+ \"a00788_source.html\": [2, 0, 0, 2, 35],\n+ \"a00791.html\": [2, 0, 0, 2, 10],\n+ \"a00791_source.html\": [2, 0, 0, 2, 10],\n+ \"a00794.html\": [2, 0, 0, 2, 9],\n+ \"a00794_source.html\": [2, 0, 0, 2, 9],\n \"a00797.html\": [2, 0, 0, 2, 3],\n \"a00797_source.html\": [2, 0, 0, 2, 3],\n- \"a00800.html\": [2, 0, 0, 2, 1],\n- \"a00800_source.html\": [2, 0, 0, 2, 1],\n- \"a00803.html\": [2, 0, 0, 2, 2],\n- \"a00803_source.html\": [2, 0, 0, 2, 2],\n+ \"a00800.html\": [2, 0, 0, 2, 2],\n+ \"a00800_source.html\": [2, 0, 0, 2, 2],\n+ \"a00803.html\": [2, 0, 0, 2, 1],\n+ \"a00803_source.html\": [2, 0, 0, 2, 1],\n \"a00806.html\": [2, 0, 0, 2, 26],\n \"a00806_source.html\": [2, 0, 0, 2, 26],\n \"a00809.html\": [2, 0, 0, 2, 30],\n \"a00809_source.html\": [2, 0, 0, 2, 30],\n- \"a00812.html\": [2, 0, 0, 2, 29],\n- \"a00812.html#aa41e5b43fe46eeb0f97ff6ebdb1935aa\": [2, 0, 0, 2, 29, 0],\n- \"a00812_source.html\": [2, 0, 0, 2, 29],\n- \"a00815.html\": [2, 0, 0, 2, 31],\n- \"a00815_source.html\": [2, 0, 0, 2, 31],\n- \"a00818.html\": [2, 0, 0, 2, 28],\n- \"a00818_source.html\": [2, 0, 0, 2, 28],\n- \"a00821.html\": [2, 0, 0, 2, 32],\n- \"a00821_source.html\": [2, 0, 0, 2, 32],\n- \"a00824.html\": [2, 0, 0, 2, 27],\n- \"a00824_source.html\": [2, 0, 0, 2, 27],\n- \"a00827.html\": [2, 0, 0, 2, 14],\n- \"a00827.html#ae5c8e80d05d6eaee105d8eecfa9ab609\": [2, 0, 0, 2, 14, 0],\n- \"a00827_source.html\": [2, 0, 0, 2, 14],\n- \"a00830.html\": [2, 0, 0, 2, 16],\n- \"a00830_source.html\": [2, 0, 0, 2, 16],\n- \"a00833.html\": [2, 0, 0, 2, 24],\n- \"a00833_source.html\": [2, 0, 0, 2, 24],\n- \"a00836.html\": [2, 0, 0, 2, 13],\n- \"a00836_source.html\": [2, 0, 0, 2, 13],\n- \"a00839.html\": [2, 0, 0, 2, 12],\n- \"a00839_source.html\": [2, 0, 0, 2, 12],\n- \"a00842.html\": [2, 0, 0, 2, 4],\n- \"a00842_source.html\": [2, 0, 0, 2, 4],\n- \"a00845.html\": [2, 0, 0, 2, 15],\n- \"a00845_source.html\": [2, 0, 0, 2, 15],\n- \"a00848.html\": [2, 0, 0, 2, 17],\n- \"a00848_source.html\": [2, 0, 0, 2, 17],\n- \"a00851.html\": [2, 0, 0, 2, 23],\n- \"a00851_source.html\": [2, 0, 0, 2, 23],\n- \"a00854.html\": [2, 0, 0, 2, 19],\n- \"a00854.html#a7f9cf2b729178bdbb847fc4afd91af9e\": [2, 0, 0, 2, 19, 0],\n- \"a00854_source.html\": [2, 0, 0, 2, 19],\n- \"a00857.html\": [2, 0, 0, 2, 22],\n- \"a00857_source.html\": [2, 0, 0, 2, 22],\n- \"a00860.html\": [2, 0, 0, 2, 21],\n- \"a00860_source.html\": [2, 0, 0, 2, 21],\n- \"a00863.html\": [2, 0, 0, 2, 20],\n- \"a00863_source.html\": [2, 0, 0, 2, 20],\n- \"a00866.html\": [2, 0, 0, 2, 18],\n- \"a00866_source.html\": [2, 0, 0, 2, 18],\n- \"a00869.html\": [2, 0, 0, 2, 38],\n- \"a00869.html#a20e08be34d6183b6943ed7a1586eca4a\": [2, 0, 0, 2, 38, 1],\n- \"a00869.html#a4a6a0fcc01638f2a0316b32ad158b485\": [2, 0, 0, 2, 38, 4],\n- \"a00869.html#aa0c3c264a43eeddc5cb3a58b006d3d32\": [2, 0, 0, 2, 38, 0],\n- \"a00869.html#aa65ed860a2fa5935d99129a898a8f0b6\": [2, 0, 0, 2, 38, 2],\n- \"a00869.html#af23c35537952bfdd493441e5c42ff2ed\": [2, 0, 0, 2, 38, 3],\n- \"a00869_source.html\": [2, 0, 0, 2, 38],\n+ \"a00812.html\": [2, 0, 0, 2, 31],\n+ \"a00812_source.html\": [2, 0, 0, 2, 31],\n+ \"a00815.html\": [2, 0, 0, 2, 28],\n+ \"a00815_source.html\": [2, 0, 0, 2, 28],\n+ \"a00818.html\": [2, 0, 0, 2, 29],\n+ \"a00818.html#aa41e5b43fe46eeb0f97ff6ebdb1935aa\": [2, 0, 0, 2, 29, 0],\n+ \"a00818_source.html\": [2, 0, 0, 2, 29],\n+ \"a00821.html\": [2, 0, 0, 2, 27],\n+ \"a00821_source.html\": [2, 0, 0, 2, 27],\n+ \"a00824.html\": [2, 0, 0, 2, 32],\n+ \"a00824_source.html\": [2, 0, 0, 2, 32],\n+ \"a00827.html\": [2, 0, 0, 2, 24],\n+ \"a00827_source.html\": [2, 0, 0, 2, 24],\n+ \"a00830.html\": [2, 0, 0, 2, 14],\n+ \"a00830.html#ae5c8e80d05d6eaee105d8eecfa9ab609\": [2, 0, 0, 2, 14, 0],\n+ \"a00830_source.html\": [2, 0, 0, 2, 14],\n+ \"a00833.html\": [2, 0, 0, 2, 16],\n+ \"a00833_source.html\": [2, 0, 0, 2, 16],\n+ \"a00836.html\": [2, 0, 0, 2, 15],\n+ \"a00836_source.html\": [2, 0, 0, 2, 15],\n+ \"a00839.html\": [2, 0, 0, 2, 13],\n+ \"a00839_source.html\": [2, 0, 0, 2, 13],\n+ \"a00842.html\": [2, 0, 0, 2, 12],\n+ \"a00842_source.html\": [2, 0, 0, 2, 12],\n+ \"a00845.html\": [2, 0, 0, 2, 23],\n+ \"a00845_source.html\": [2, 0, 0, 2, 23],\n+ \"a00848.html\": [2, 0, 0, 2, 4],\n+ \"a00848_source.html\": [2, 0, 0, 2, 4],\n+ \"a00851.html\": [2, 0, 0, 2, 17],\n+ \"a00851_source.html\": [2, 0, 0, 2, 17],\n+ \"a00854.html\": [2, 0, 0, 2, 21],\n+ \"a00854_source.html\": [2, 0, 0, 2, 21],\n+ \"a00857.html\": [2, 0, 0, 2, 19],\n+ \"a00857.html#a7f9cf2b729178bdbb847fc4afd91af9e\": [2, 0, 0, 2, 19, 0],\n+ \"a00857_source.html\": [2, 0, 0, 2, 19],\n+ \"a00860.html\": [2, 0, 0, 2, 20],\n+ \"a00860_source.html\": [2, 0, 0, 2, 20],\n+ \"a00863.html\": [2, 0, 0, 2, 18],\n+ \"a00863_source.html\": [2, 0, 0, 2, 18],\n+ \"a00866.html\": [2, 0, 0, 2, 22],\n+ \"a00866_source.html\": [2, 0, 0, 2, 22],\n+ \"a00869.html\": [2, 0, 0, 2, 37],\n+ \"a00869_source.html\": [2, 0, 0, 2, 37],\n \"a00872.html\": [2, 0, 0, 2, 6],\n \"a00872_source.html\": [2, 0, 0, 2, 6],\n- \"a00875.html\": [2, 0, 0, 2, 37],\n- \"a00875_source.html\": [2, 0, 0, 2, 37],\n+ \"a00875.html\": [2, 0, 0, 2, 38],\n+ \"a00875.html#a20e08be34d6183b6943ed7a1586eca4a\": [2, 0, 0, 2, 38, 1],\n+ \"a00875.html#a4a6a0fcc01638f2a0316b32ad158b485\": [2, 0, 0, 2, 38, 4],\n+ \"a00875.html#aa0c3c264a43eeddc5cb3a58b006d3d32\": [2, 0, 0, 2, 38, 0],\n+ \"a00875.html#aa65ed860a2fa5935d99129a898a8f0b6\": [2, 0, 0, 2, 38, 2],\n+ \"a00875.html#af23c35537952bfdd493441e5c42ff2ed\": [2, 0, 0, 2, 38, 3],\n+ \"a00875_source.html\": [2, 0, 0, 2, 38],\n \"a00878.html\": [2, 0, 0, 2, 33],\n \"a00878_source.html\": [2, 0, 0, 2, 33],\n- \"a00881.html\": [2, 0, 0, 2, 34],\n- \"a00881.html#a44a05502feb5015035bc81b93cffb99c\": [2, 0, 0, 2, 34, 0],\n- \"a00881.html#a870745ff3e883b32547d4db42b3be2bd\": [2, 0, 0, 2, 34, 2],\n- \"a00881.html#a907dc53421460557871ddc35f77751f5\": [2, 0, 0, 2, 34, 1],\n- \"a00881_source.html\": [2, 0, 0, 2, 34],\n- \"a00884.html\": [2, 0, 0, 2, 36],\n- \"a00884_source.html\": [2, 0, 0, 2, 36],\n+ \"a00881.html\": [2, 0, 0, 2, 36],\n+ \"a00881_source.html\": [2, 0, 0, 2, 36],\n+ \"a00884.html\": [2, 0, 0, 2, 34],\n+ \"a00884.html#a44a05502feb5015035bc81b93cffb99c\": [2, 0, 0, 2, 34, 0],\n+ \"a00884.html#a870745ff3e883b32547d4db42b3be2bd\": [2, 0, 0, 2, 34, 2],\n+ \"a00884.html#a907dc53421460557871ddc35f77751f5\": [2, 0, 0, 2, 34, 1],\n+ \"a00884_source.html\": [2, 0, 0, 2, 34],\n \"a00887.html\": [2, 0, 0, 2, 25],\n \"a00887.html#a06ceddf21c024b341a5b946049cd701a\": [2, 0, 0, 2, 25, 6],\n \"a00887.html#a1af6b50e9f8ae13d3d2f9d5b03b35492\": [2, 0, 0, 2, 25, 5],\n \"a00887.html#a1af6b50e9f8ae13d3d2f9d5b03b35492a4124233e4f787f999c7975afceab34f3\": [2, 0, 0, 2, 25, 5, 7],\n \"a00887.html#a1af6b50e9f8ae13d3d2f9d5b03b35492a5676712c77d29cfdd330e2802db1c464\": [2, 0, 0, 2, 25, 5, 11],\n \"a00887.html#a1af6b50e9f8ae13d3d2f9d5b03b35492a8fab1b3018f94b5b2432c7435e795925\": [2, 0, 0, 2, 25, 5, 6],\n \"a00887.html#a1af6b50e9f8ae13d3d2f9d5b03b35492a9058d69e1089fc77f1974a76c34e2071\": [2, 0, 0, 2, 25, 5, 5],\n@@ -182,22 +182,22 @@\n \"a00887.html#a1af6b50e9f8ae13d3d2f9d5b03b35492ac10e41fef659f3b3a1b730319bdedb45\": [2, 0, 0, 2, 25, 5, 2],\n \"a00887.html#a1af6b50e9f8ae13d3d2f9d5b03b35492adef344bcbdd50a1a93136eb53183c718\": [2, 0, 0, 2, 25, 5, 0],\n \"a00887.html#a33f9830228f847fae1c30c9dffe02b5a\": [2, 0, 0, 2, 25, 9],\n \"a00887.html#a4c2520cc1a8a5a03e3f53584629d7318\": [2, 0, 0, 2, 25, 7],\n \"a00887.html#a85ba9df501aea9af1ec9da1558c255bc\": [2, 0, 0, 2, 25, 4],\n \"a00887.html#ab5753a2b0a0995c8e2259d2c4765f7df\": [2, 0, 0, 2, 25, 8],\n \"a00887_source.html\": [2, 0, 0, 2, 25],\n- \"a00890.html\": [2, 0, 0, 2, 7],\n- \"a00890_source.html\": [2, 0, 0, 2, 7],\n- \"a00893.html\": [2, 0, 0, 2, 8],\n- \"a00893_source.html\": [2, 0, 0, 2, 8],\n- \"a00896.html\": [2, 0, 0, 2, 0],\n- \"a00896_source.html\": [2, 0, 0, 2, 0],\n- \"a00899.html\": [2, 0, 0, 2, 5],\n- \"a00899_source.html\": [2, 0, 0, 2, 5],\n+ \"a00890.html\": [2, 0, 0, 2, 8],\n+ \"a00890_source.html\": [2, 0, 0, 2, 8],\n+ \"a00893.html\": [2, 0, 0, 2, 7],\n+ \"a00893_source.html\": [2, 0, 0, 2, 7],\n+ \"a00896.html\": [2, 0, 0, 2, 5],\n+ \"a00896_source.html\": [2, 0, 0, 2, 5],\n+ \"a00899.html\": [2, 0, 0, 2, 0],\n+ \"a00899_source.html\": [2, 0, 0, 2, 0],\n \"a00902.html\": [2, 0, 0, 2, 11],\n \"a00902_source.html\": [2, 0, 0, 2, 11],\n \"a00905.html\": [0, 0, 0],\n \"a00906.html\": [0, 0, 0, 0],\n \"a00907.html\": [0, 0, 0, 0, 0],\n \"a00908.html\": [0, 0, 0, 0, 1],\n \"a00908.html#a200172f1c19545bc01f292afa2167e2f\": [0, 0, 0, 0, 1, 32],\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_10.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_10.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -2,23 +2,23 @@\n ['reference_0', ['reference', ['../a01345.html#a9025ff89fc9afe7d9054abb58123bbda', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstArray::reference()'],\n ['../a01349.html#a9025ff89fc9afe7d9054abb58123bbda', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Vtr::Array::reference()']\n ]],\n ['refine_1', ['Refine', ['../a01261.html#a9a90398ba2073fe3da454f6b00deb33c', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface::Refine()'],\n ['../a01273.html#a32c8d641fe9ea1565be646a10c5d6c85', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::Mesh::Refine()']\n ]],\n ['refineadaptive_2', ['RefineAdaptive', ['../a01113.html#aff92309935db19988908471bee4fef4d', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner']]],\n- ['refinement_2eh_3', ['refinement.h', ['../a00755.html', 1, '']]],\n+ ['refinement_2eh_3', ['refinement.h', ['../a00752.html', 1, '']]],\n ['refinementlevel_4', ['refinementLevel', ['../a01117.html#a240d1e5f5b743486e8122db6223d4562', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::UniformOptions']]],\n ['refinemesh_5', ['refineMesh', ['../a01261.html#a51f237c3f08f6ddc2ea7ba5cb0d00b36', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface::refineMesh(Far::TopologyRefiner &refiner, int level, bool adaptive, bool singleCreasePatch)'],\n ['../a01261.html#a173851f742adc30ac139af96fe63732d', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::MeshInterface::refineMesh(Far::TopologyRefiner &refiner, int level, MeshBitset bits)']\n ]],\n ['refinersurfacefactory_6', ['RefinerSurfaceFactory', ['../a00929.html', 1, 'RefinerSurfaceFactory< CACHE_TYPE >'],\n ['../a00929.html#af74dafe122d798d4069b675c52f6665a', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactory::RefinerSurfaceFactory()']\n ]],\n- ['refinersurfacefactory_2eh_7', ['refinerSurfaceFactory.h', ['../a00689.html', 1, '']]],\n+ ['refinersurfacefactory_2eh_7', ['refinerSurfaceFactory.h', ['../a00692.html', 1, '']]],\n ['refinersurfacefactorybase_8', ['RefinerSurfaceFactoryBase', ['../a00925.html', 1, 'RefinerSurfaceFactoryBase'],\n ['../a00925.html#af457c931d6d611e35737454f2c483f84', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::RefinerSurfaceFactoryBase::RefinerSurfaceFactoryBase()']\n ]],\n ['refineuniform_9', ['RefineUniform', ['../a01113.html#ad22b7ffd9f8f3b488e5f220cf00c8c82', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner']]],\n ['regdesc_10', ['regDesc', ['../a01317.html#a00efd3a5502664756617ee930cb4324c', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArray']]],\n ['regular_11', ['REGULAR', ['../a00973.html#a1d1cfd8ffb84e947f82999c682b666a7afc488491175a6dac9e5881e363ad4c41', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor']]],\n ['reportinvalidtopology_12', ['reportInvalidTopology', ['../a01109.html#a0c4c1d44df4c480bd0adfa9e86891b39', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory']]],\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_11.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_11.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -100,17 +100,17 @@\n ['stencilreal_3c_20float_20_3e_75', ['StencilReal< float >', ['../a01045.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far']]],\n ['stenciltable_76', ['StencilTable', ['../a01057.html', 1, 'StencilTable'],\n ['../a01057.html#a3d12830eb0d421206de500d7abc9decf', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::StencilTable(int numControlVerts, std::vector< int > const &offsets, std::vector< int > const &sizes, std::vector< int > const &sources, std::vector< float > const &weights, bool includeCoarseVerts, size_t firstOffset)'],\n ['../a01057.html#a54db76b34112b5cb3c623be8789dd8f4', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::StencilTable(int numControlVerts)'],\n ['../a01057.html#acc553d2d5624ab2a790ae9613812393e', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::StencilTable()'],\n ['../a01273.html#a6b90e8290c540c78bedd053e2cc42f31', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::Mesh::StencilTable()']\n ]],\n- ['stenciltable_2eh_77', ['stencilTable.h', ['../a00725.html', 1, '']]],\n+ ['stenciltable_2eh_77', ['stencilTable.h', ['../a00713.html', 1, '']]],\n ['stenciltablefactory_78', ['StencilTableFactory', ['../a01089.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far']]],\n- ['stenciltablefactory_2eh_79', ['stencilTableFactory.h', ['../a00710.html', 1, '']]],\n+ ['stenciltablefactory_2eh_79', ['stencilTableFactory.h', ['../a00716.html', 1, '']]],\n ['stenciltablefactoryreal_80', ['StencilTableFactoryReal', ['../a01037.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far']]],\n ['stenciltablefactoryreal_3c_20float_20_3e_81', ['StencilTableFactoryReal< float >', ['../a01037.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far']]],\n ['stenciltablefactoryreal_3c_20real_20_3e_82', ['StencilTableFactoryReal< REAL >', ['../a01045.html#af58f6a3f9b167c3330097c1a69ca36a8', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilReal::StencilTableFactoryReal< REAL >()'],\n ['../a01053.html#af58f6a3f9b167c3330097c1a69ca36a8', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::StencilTableFactoryReal< REAL >()'],\n ['../a01061.html#af58f6a3f9b167c3330097c1a69ca36a8', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilReal::StencilTableFactoryReal< REAL >()']\n ]],\n ['stenciltablereal_83', ['StencilTableReal', ['../a01053.html', 1, 'StencilTableReal< REAL >'],\n@@ -127,15 +127,15 @@\n ['subdivideedgesharpnessesaroundvertex_87', ['SubdivideEdgeSharpnessesAroundVertex', ['../a01325.html#a7c27382a0be83617c49b8587ba75b438', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease']]],\n ['subdivideuniformsharpness_88', ['SubdivideUniformSharpness', ['../a01325.html#a7e5126bd5cff203bcea84ef35845fb40', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease']]],\n ['subdividevertexsharpness_89', ['SubdivideVertexSharpness', ['../a01325.html#a03f4bc921d6ee6fad4440ac134e92e36', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease']]],\n ['surface_90', ['Surface', ['../a00933.html', 1, 'Surface< REAL >'],\n ['../a00933.html#a6e1a6ad293d9318f65c2e28c468c66ba', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Surface()'],\n ['../a00933.html#aa9500016cdd5cf88992541cc263a75c9', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Surface(Surface const &src)=default']\n ]],\n- ['surface_2eh_91', ['surface.h', ['../a00692.html', 1, '']]],\n+ ['surface_2eh_91', ['surface.h', ['../a00689.html', 1, '']]],\n ['surfacefactory_92', ['SurfaceFactory', ['../a00941.html', 1, 'SurfaceFactory'],\n ['../a00941.html#ae54ee5a9e2bcb2bfdf395084395a0719', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::SurfaceFactory(Sdc::SchemeType schemeType, Sdc::Options const &schemeOptions, Options const &limitOptions)'],\n ['../a00941.html#a5ac6bb6255796f79debe11e25b816093', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::SurfaceFactory(SurfaceFactory const &)=delete'],\n ['../a00933.html#a3bdc23048cdd1242d0a429413fa89256', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::SurfaceFactory()']\n ]],\n ['surfacefactory_2eh_93', ['surfaceFactory.h', ['../a00686.html', 1, '']]],\n ['surfacefactorycache_94', ['SurfaceFactoryCache', ['../a00949.html', 1, 'SurfaceFactoryCache'],\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_12.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_12.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -6,33 +6,33 @@\n ['../a00912.html#a20e08be34d6183b6943ed7a1586eca4a', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalPatches(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int numPatchCoords, const PatchCoord *patchCoords, const PatchArray *patchArrayBuffer, const int *patchIndexBuffer, const PatchParam *patchParamBuffer)']\n ]],\n ['tbbevalstencils_2', ['TbbEvalStencils', ['../a00912.html#aa65ed860a2fa5935d99129a898a8f0b6', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalStencils(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv, BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const &dstDvvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, float const *duuWeights, float const *duvWeights, float const *dvvWeights, int start, int end)'],\n ['../a00912.html#af23c35537952bfdd493441e5c42ff2ed', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalStencils(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, int start, int end)'],\n ['../a00912.html#a4a6a0fcc01638f2a0316b32ad158b485', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::TbbEvalStencils(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, int start, int end)']\n ]],\n ['tbbevaluator_3', ['TbbEvaluator', ['../a01309.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n- ['tbbevaluator_2eh_4', ['tbbEvaluator.h', ['../a00875.html', 1, '']]],\n- ['tbbkernel_2eh_5', ['tbbKernel.h', ['../a00869.html', 1, '']]],\n+ ['tbbevaluator_2eh_4', ['tbbEvaluator.h', ['../a00869.html', 1, '']]],\n+ ['tbbkernel_2eh_5', ['tbbKernel.h', ['../a00875.html', 1, '']]],\n ['tessellation_6', ['Tessellation', ['../a00961.html', 1, 'Tessellation'],\n ['../a00961.html#ab03191ef31cf0d8b0bcc44f226804e87', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Tessellation::Tessellation(Tessellation const &)=delete'],\n ['../a00961.html#a7a3e89992f3cc492804ec9bc184a1674', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Tessellation::Tessellation()=delete'],\n ['../a00961.html#a65ee652b3607c97bb2b04a64c55d7f99', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Tessellation::Tessellation(Parameterization const &p, int numRates, int const rates[], Options const &options=Options())'],\n ['../a00961.html#a7f0af117d53701f7ca6e3eb2368b89bb', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Tessellation::Tessellation(Parameterization const &p, int uniformRate, Options const &options=Options())']\n ]],\n ['tessellation_2eh_7', ['tessellation.h', ['../a00677.html', 1, '']]],\n ['topologycallback_8', ['TopologyCallback', ['../a01125.html#a896a964625491e084b3d776acd10f053', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactoryBase']]],\n ['topologydescriptor_9', ['TopologyDescriptor', ['../a01097.html', 1, 'TopologyDescriptor'],\n ['../a01097.html#a8209f1b69b0162ee2e7065d91088113d', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor::TopologyDescriptor()']\n ]],\n- ['topologydescriptor_2eh_10', ['topologyDescriptor.h', ['../a00713.html', 1, '']]],\n+ ['topologydescriptor_2eh_10', ['topologyDescriptor.h', ['../a00710.html', 1, '']]],\n ['topologyerror_11', ['TopologyError', ['../a01109.html#a5458be6ce3f2e63c8d8555f4ab3b2b62', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefinerFactory']]],\n ['topologylevel_12', ['TopologyLevel', ['../a01105.html', 1, 'TopologyLevel'],\n ['../a01105.html#a03a8183081b34983ade155e3636ba835', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::TopologyLevel()']\n ]],\n- ['topologylevel_2eh_13', ['topologyLevel.h', ['../a00722.html', 1, '']]],\n+ ['topologylevel_2eh_13', ['topologyLevel.h', ['../a00719.html', 1, '']]],\n ['topologyrefiner_14', ['TopologyRefiner', ['../a01113.html', 1, 'TopologyRefiner'],\n ['../a01105.html#a094d2e51437ecd5ecc6a0cf2c4ce1e78', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel::TopologyRefiner()'],\n ['../a01113.html#a91353962838bff64448014136e0e11ff', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::TopologyRefiner(Sdc::SchemeType type, Sdc::Options options=Sdc::Options())'],\n ['../a01113.html#a13c709a24feeed80409924510e3efc5e', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::TopologyRefiner(TopologyRefiner const &source)']\n ]],\n ['topologyrefiner_2eh_15', ['topologyRefiner.h', ['../a00707.html', 1, '']]],\n ['topologyrefinerfactory_16', ['TopologyRefinerFactory', ['../a01109.html', 1, 'TopologyRefinerFactory< MESH >'],\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_2.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_2.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -4,15 +4,15 @@\n ]],\n ['basetable_1', ['BaseTable', ['../a01057.html#a54f0bb223290a17a59f51f4cbcfbc2e2', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTable::BaseTable()'],\n ['../a01073.html#a13d2ee8ca9cdc9073905193ea1cefd95', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTable::BaseTable()']\n ]],\n ['begin_2', ['begin', ['../a01345.html#a29305669b60ca1680752e2fc3592ba99', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Vtr::ConstArray::begin()'],\n ['../a01349.html#ad69bd11391be1a1dba5c8202259664f8', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Vtr::Array::begin()']\n ]],\n- ['bilinearscheme_2eh_3', ['bilinearScheme.h', ['../a00782.html', 1, '']]],\n+ ['bilinearscheme_2eh_3', ['bilinearScheme.h', ['../a00785.html', 1, '']]],\n ['bindclbuffer_4', ['BindCLBuffer', ['../a01149.html#ad9c25d5b96f9d5e971f677d32b406ec6', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLVertexBuffer']]],\n ['bindcpubuffer_5', ['BindCpuBuffer', ['../a01153.html#adf7a9df5d3fc4d803bf5f1f374d90274', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::BindCpuBuffer()'],\n ['../a01161.html#adf7a9df5d3fc4d803bf5f1f374d90274', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuGLVertexBuffer::BindCpuBuffer()'],\n ['../a01169.html#adf7a9df5d3fc4d803bf5f1f374d90274', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::BindCpuBuffer()'],\n ['../a01301.html#adf7a9df5d3fc4d803bf5f1f374d90274', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::BindCpuBuffer()']\n ]],\n ['bindcudabuffer_6', ['BindCudaBuffer', ['../a01185.html#a59eb144c23606529ba730ae15598af09', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaVertexBuffer']]],\n@@ -35,9 +35,9 @@\n ]],\n ['boundcontrolpoints_13', ['BoundControlPoints', ['../a00933.html#a22604a4489ed5fdedb0ef272de0fa637', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface']]],\n ['boundcontrolpointsfrommesh_14', ['BoundControlPointsFromMesh', ['../a00933.html#a8df5cf06cea51aeb6419fd698651ae3b', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface']]],\n ['bufferdescriptor_15', ['BufferDescriptor', ['../a01133.html#a446218762f5de8c77d06ee61e14b6b61', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::BufferDescriptor()'],\n ['../a01133.html#a958068066d1d10ae2a452ac57d9c0401', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor::BufferDescriptor(int o, int l, int s)'],\n ['../a01133.html', 1, 'BufferDescriptor']\n ]],\n- ['bufferdescriptor_2eh_16', ['bufferDescriptor.h', ['../a00896.html', 1, '']]]\n+ ['bufferdescriptor_2eh_16', ['bufferDescriptor.h', ['../a00899.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_3.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_3.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,27 +1,27 @@\n var searchData = [\n- ['catmarkscheme_2eh_0', ['catmarkScheme.h', ['../a00785.html', 1, '']]],\n+ ['catmarkscheme_2eh_0', ['catmarkScheme.h', ['../a00782.html', 1, '']]],\n ['child_1', ['Child', ['../a00985.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchMap::QuadNode']]],\n ['clear_2', ['clear', ['../a01017.html#ac8bb3912a3ce86b15842e79d0b421204', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTableFactory::PatchFaceTag']]],\n ['clear_3', ['Clear', ['../a01069.html#aa71d36872f416feaa853788a7a7a7ef8', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableReal::Clear()'],\n ['../a01053.html#aa71d36872f416feaa853788a7a7a7ef8', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableReal::Clear()'],\n ['../a00989.html#aa71d36872f416feaa853788a7a7a7ef8', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchParam::Clear()'],\n ['../a00933.html#aa71d36872f416feaa853788a7a7a7ef8', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::Clear()']\n ]],\n ['clearedgesharpness_4', ['ClearEdgeSharpness', ['../a00969.html#a831025d6d817c4afb1df7d5a705af378', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::VertexDescriptor']]],\n ['clearincidentfacesizes_5', ['ClearIncidentFaceSizes', ['../a00969.html#a5c719e43580059139938a6ed0f851853', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::VertexDescriptor']]],\n ['clearvertexsharpness_6', ['ClearVertexSharpness', ['../a00969.html#a68afc60df5a617f5d7e1580548b72f9b', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::VertexDescriptor']]],\n ['clevaluator_7', ['CLEvaluator', ['../a01141.html#adbb0aafb7db3284f164499e8007c4cac', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLEvaluator::CLEvaluator()'],\n ['../a01141.html', 1, 'CLEvaluator']\n ]],\n- ['clevaluator_2eh_8', ['clEvaluator.h', ['../a00800.html', 1, '']]],\n+ ['clevaluator_2eh_8', ['clEvaluator.h', ['../a00803.html', 1, '']]],\n ['clpatchtable_9', ['CLPatchTable', ['../a01145.html#ae69549a9493f9a51edfa7d11cec1d72f', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLPatchTable::CLPatchTable()'],\n ['../a01145.html', 1, 'CLPatchTable']\n ]],\n- ['clpatchtable_2eh_10', ['clPatchTable.h', ['../a00803.html', 1, '']]],\n+ ['clpatchtable_2eh_10', ['clPatchTable.h', ['../a00800.html', 1, '']]],\n ['clstenciltable_11', ['CLStencilTable', ['../a01137.html#af2994e79cf93c872e39659321750351a', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::CLStencilTable(Far::LimitStencilTable const *limitStencilTable, cl_context clContext)'],\n ['../a01137.html#a33ca6bb1dba9f8e390a58395d164431e', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLStencilTable::CLStencilTable(Far::StencilTable const *stencilTable, cl_context clContext)'],\n ['../a01137.html', 1, 'CLStencilTable']\n ]],\n ['clvertexbuffer_12', ['CLVertexBuffer', ['../a01149.html#a48e2b665b557d5f3f838bc552152160b', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CLVertexBuffer::CLVertexBuffer()'],\n ['../a01149.html', 1, 'CLVertexBuffer']\n ]],\n@@ -77,32 +77,32 @@\n ['converttocompatiblestenciltable_3c_20far_3a_3astenciltable_2c_20far_3a_3astenciltable_2c_20id3d11devicecontext_20_3e_42', ['convertToCompatibleStencilTable< Far::StencilTable, Far::StencilTable, ID3D11DeviceContext >', ['../a00912.html#ab5753a2b0a0995c8e2259d2c4765f7df', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n ['converttocompatiblestenciltable_3c_20far_3a_3astenciltable_2c_20far_3a_3astenciltable_2c_20void_20_3e_43', ['convertToCompatibleStencilTable< Far::StencilTable, Far::StencilTable, void >', ['../a00912.html#a33f9830228f847fae1c30c9dffe02b5a', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n ['cornervertexindices_44', ['cornerVertexIndices', ['../a01097.html#a3bf794743edda29fbee0653b65b75ba4', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor']]],\n ['cornerweights_45', ['cornerWeights', ['../a01097.html#a6d938f6b0a9ff0fa377ef675aa2c7fda', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor']]],\n ['cpud3d11vertexbuffer_46', ['CpuD3D11VertexBuffer', ['../a01153.html#a632143bc1ec4fbbca2148b0e9e87bc08', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuD3D11VertexBuffer::CpuD3D11VertexBuffer()'],\n ['../a01153.html', 1, 'CpuD3D11VertexBuffer']\n ]],\n- ['cpud3d11vertexbuffer_2eh_47', ['cpuD3D11VertexBuffer.h', ['../a00842.html', 1, '']]],\n+ ['cpud3d11vertexbuffer_2eh_47', ['cpuD3D11VertexBuffer.h', ['../a00848.html', 1, '']]],\n ['cpuevaluator_48', ['CpuEvaluator', ['../a01157.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n- ['cpuevaluator_2eh_49', ['cpuEvaluator.h', ['../a00899.html', 1, '']]],\n+ ['cpuevaluator_2eh_49', ['cpuEvaluator.h', ['../a00896.html', 1, '']]],\n ['cpuglvertexbuffer_50', ['CpuGLVertexBuffer', ['../a01161.html#a32dd2317d622c1b079e3219f97c86955', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuGLVertexBuffer::CpuGLVertexBuffer()'],\n ['../a01161.html', 1, 'CpuGLVertexBuffer']\n ]],\n ['cpuglvertexbuffer_2eh_51', ['cpuGLVertexBuffer.h', ['../a00872.html', 1, '']]],\n ['cpumtlvertexbuffer_52', ['CPUMTLVertexBuffer', ['../a01301.html#a1457f342bc3007386104aa3d29277a67', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CPUMTLVertexBuffer::CPUMTLVertexBuffer()'],\n ['../a01301.html', 1, 'CPUMTLVertexBuffer']\n ]],\n ['cpupatchtable_53', ['CpuPatchTable', ['../a01165.html#a3c19d1748852f36ba69a6557e5069e08', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuPatchTable::CpuPatchTable()'],\n ['../a01165.html', 1, 'CpuPatchTable']\n ]],\n- ['cpupatchtable_2eh_54', ['cpuPatchTable.h', ['../a00890.html', 1, '']]],\n+ ['cpupatchtable_2eh_54', ['cpuPatchTable.h', ['../a00893.html', 1, '']]],\n ['cpuvertexbuffer_55', ['CpuVertexBuffer', ['../a01169.html#afa9168f4f1f046aea4ff94933ac48972', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CpuVertexBuffer::CpuVertexBuffer()'],\n ['../a01169.html', 1, 'CpuVertexBuffer']\n ]],\n- ['cpuvertexbuffer_2eh_56', ['cpuVertexBuffer.h', ['../a00893.html', 1, '']]],\n+ ['cpuvertexbuffer_2eh_56', ['cpuVertexBuffer.h', ['../a00890.html', 1, '']]],\n ['crease_57', ['Crease', ['../a01325.html#a9659c3f2c73b80d9dc22763a6054079b', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::Crease()'],\n ['../a01325.html#a75d5d7e54170ccf6481932916e0fa9c1', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::Crease(Options const &options)'],\n ['../a01325.html', 1, 'Crease']\n ]],\n ['crease_2eh_58', ['crease.h', ['../a00779.html', 1, '']]],\n ['crease_5fchaikin_59', ['CREASE_CHAIKIN', ['../a01329.html#a16ec522dc0398bf1435a912cd669ffb0a1e31a583663dd65a7b2feffeb24268b5', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Options']]],\n ['crease_5funiform_60', ['CREASE_UNIFORM', ['../a01329.html#a16ec522dc0398bf1435a912cd669ffb0a10f2956db48d2975424be2e52ebd544d', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Options']]],\n@@ -174,19 +174,19 @@\n ]],\n ['createfacevaryingsurface_66', ['CreateFaceVaryingSurface', ['../a00941.html#a67a5dcab87b052d231634dd9fcf5dee2', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::CreateFaceVaryingSurface(Index faceIndex) const'],\n ['../a00941.html#a59227e292ea63aa9641a090521cfd457', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::CreateFaceVaryingSurface(Index faceIndex, FVarID id) const']\n ]],\n ['createvaryingsurface_67', ['CreateVaryingSurface', ['../a00941.html#a81638f2ff6818fd96aa7cd1ad3ed0703', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory']]],\n ['createvertexsurface_68', ['CreateVertexSurface', ['../a00941.html#a676cf019574c99fa77341ac09889192d', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory']]],\n ['cudaevaluator_69', ['CudaEvaluator', ['../a01177.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n- ['cudaevaluator_2eh_70', ['cudaEvaluator.h', ['../a00791.html', 1, '']]],\n+ ['cudaevaluator_2eh_70', ['cudaEvaluator.h', ['../a00794.html', 1, '']]],\n ['cudapatchtable_71', ['CudaPatchTable', ['../a01181.html#aefa9c1a9175327a40e7cf46ed32ced2f', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaPatchTable::CudaPatchTable()'],\n ['../a01181.html', 1, 'CudaPatchTable']\n ]],\n- ['cudapatchtable_2eh_72', ['cudaPatchTable.h', ['../a00788.html', 1, '']]],\n+ ['cudapatchtable_2eh_72', ['cudaPatchTable.h', ['../a00791.html', 1, '']]],\n ['cudastenciltable_73', ['CudaStencilTable', ['../a01173.html#a45150c630c07bbeabd4b91145cecf087', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::CudaStencilTable(Far::LimitStencilTable const *limitStencilTable)'],\n ['../a01173.html#af872fc469b3696a745d30d22c2b89c4c', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaStencilTable::CudaStencilTable(Far::StencilTable const *stencilTable)'],\n ['../a01173.html', 1, 'CudaStencilTable']\n ]],\n ['cudavertexbuffer_74', ['CudaVertexBuffer', ['../a01185.html#af6cb32d896f7815cac1969c857845ba8', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::CudaVertexBuffer::CudaVertexBuffer()'],\n ['../a01185.html', 1, 'CudaVertexBuffer']\n ]],\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_4.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_4.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,29 +1,29 @@\n var searchData = [\n ['d3d11computeevaluator_0', ['D3D11ComputeEvaluator', ['../a01193.html#aa81511d3938d70d67a62217951cb8eb7', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11ComputeEvaluator::D3D11ComputeEvaluator()'],\n ['../a01193.html', 1, 'D3D11ComputeEvaluator']\n ]],\n- ['d3d11computeevaluator_2eh_1', ['d3d11ComputeEvaluator.h', ['../a00839.html', 1, '']]],\n+ ['d3d11computeevaluator_2eh_1', ['d3d11ComputeEvaluator.h', ['../a00842.html', 1, '']]],\n ['d3d11legacygregorypatchtable_2', ['D3D11LegacyGregoryPatchTable', ['../a01197.html#af138b4535e4db5f013225c5924e0b243', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11LegacyGregoryPatchTable::D3D11LegacyGregoryPatchTable()'],\n ['../a01197.html', 1, 'D3D11LegacyGregoryPatchTable']\n ]],\n- ['d3d11legacygregorypatchtable_2eh_3', ['d3d11LegacyGregoryPatchTable.h', ['../a00836.html', 1, '']]],\n- ['d3d11mesh_2eh_4', ['d3d11Mesh.h', ['../a00827.html', 1, '']]],\n+ ['d3d11legacygregorypatchtable_2eh_3', ['d3d11LegacyGregoryPatchTable.h', ['../a00839.html', 1, '']]],\n+ ['d3d11mesh_2eh_4', ['d3d11Mesh.h', ['../a00830.html', 1, '']]],\n ['d3d11meshinterface_5', ['D3D11MeshInterface', ['../a00912.html#ae5c8e80d05d6eaee105d8eecfa9ab609', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n ['d3d11patchtable_6', ['D3D11PatchTable', ['../a01201.html#ac81ed5f09e3d3a2aae5a01802b343a6d', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11PatchTable::D3D11PatchTable()'],\n ['../a01201.html', 1, 'D3D11PatchTable']\n ]],\n- ['d3d11patchtable_2eh_7', ['d3d11PatchTable.h', ['../a00845.html', 1, '']]],\n+ ['d3d11patchtable_2eh_7', ['d3d11PatchTable.h', ['../a00836.html', 1, '']]],\n ['d3d11stenciltable_8', ['D3D11StencilTable', ['../a01189.html#a7e3a2f5c281c55308400138222797dbc', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable::D3D11StencilTable()'],\n ['../a01189.html', 1, 'D3D11StencilTable']\n ]],\n ['d3d11vertexbuffer_9', ['D3D11VertexBuffer', ['../a01205.html#aacfe93ef7f7ede21dcf7088b62467742', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11VertexBuffer::D3D11VertexBuffer()'],\n ['../a01205.html', 1, 'D3D11VertexBuffer']\n ]],\n- ['d3d11vertexbuffer_2eh_10', ['d3d11VertexBuffer.h', ['../a00830.html', 1, '']]],\n+ ['d3d11vertexbuffer_2eh_10', ['d3d11VertexBuffer.h', ['../a00833.html', 1, '']]],\n ['decrementsharpness_11', ['decrementSharpness', ['../a01325.html#a453a232c0edcafd9d484197d46ed39ec', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease']]],\n ['desc_12', ['desc', ['../a01317.html#ac84ff682fda703f1f552a05977050f20', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArray']]],\n ['determinevertexvertexrule_13', ['DetermineVertexVertexRule', ['../a01325.html#a41ebb75d28c5973328e0e3883495aeba', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::DetermineVertexVertexRule(float vertexSharpness, int incidentEdgeCount, float const *incidentEdgeSharpness) const'],\n ['../a01325.html#a500fadaa8cec78ab9690ad7331dba3d6', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Sdc::Crease::DetermineVertexVertexRule(float vertexSharpness, int sharpEdgeCount) const']\n ]],\n ['device_14', ['device', ['../a01277.html#a7d70f5acf4efd89092645b4c44e480cf', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLContext']]],\n ['devicecontext_15', ['DeviceContext', ['../a01273.html#a3e2f358c45731eceb819ae70e7c8ca6e', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::Mesh']]],\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_7.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_7.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -519,41 +519,41 @@\n ['../a01281.html#ab22907c9cfe58b82ddfcad8241304500', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::GetWeightsBuffer()']\n ]],\n ['getweightssrv_270', ['GetWeightsSRV', ['../a01189.html#ac3de1233503ee41bac41618702b4302c', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::D3D11StencilTable']]],\n ['getweightstexture_271', ['GetWeightsTexture', ['../a01241.html#ad5f4a852f17762c50c5d3011b11155cb', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO']]],\n ['glcomputeevaluator_272', ['GLComputeEvaluator', ['../a01213.html#a9fe54ab76e32fb2dc6f7cd631ce2e678', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLComputeEvaluator::GLComputeEvaluator()'],\n ['../a01213.html', 1, 'GLComputeEvaluator']\n ]],\n- ['glcomputeevaluator_2eh_273', ['glComputeEvaluator.h', ['../a00848.html', 1, '']]],\n+ ['glcomputeevaluator_2eh_273', ['glComputeEvaluator.h', ['../a00851.html', 1, '']]],\n ['gllegacygregorypatchtable_274', ['GLLegacyGregoryPatchTable', ['../a01225.html#a5b78188b846b316c5c0cdb42f4ee9563', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLLegacyGregoryPatchTable::GLLegacyGregoryPatchTable()'],\n ['../a01225.html', 1, 'GLLegacyGregoryPatchTable']\n ]],\n- ['gllegacygregorypatchtable_2eh_275', ['glLegacyGregoryPatchTable.h', ['../a00866.html', 1, '']]],\n- ['glmesh_2eh_276', ['glMesh.h', ['../a00854.html', 1, '']]],\n+ ['gllegacygregorypatchtable_2eh_275', ['glLegacyGregoryPatchTable.h', ['../a00863.html', 1, '']]],\n+ ['glmesh_2eh_276', ['glMesh.h', ['../a00857.html', 1, '']]],\n ['glmeshinterface_277', ['GLMeshInterface', ['../a00912.html#a7f9cf2b729178bdbb847fc4afd91af9e', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n ['glpatchtable_278', ['GLPatchTable', ['../a01229.html#a88c12681f62189c398c1d2dc32111264', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLPatchTable::GLPatchTable()'],\n ['../a01229.html', 1, 'GLPatchTable']\n ]],\n- ['glpatchtable_2eh_279', ['glPatchTable.h', ['../a00863.html', 1, '']]],\n+ ['glpatchtable_2eh_279', ['glPatchTable.h', ['../a00860.html', 1, '']]],\n ['glslpatchshadersource_280', ['GLSLPatchShaderSource', ['../a01233.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n- ['glslpatchshadersource_2eh_281', ['glslPatchShaderSource.h', ['../a00860.html', 1, '']]],\n+ ['glslpatchshadersource_2eh_281', ['glslPatchShaderSource.h', ['../a00854.html', 1, '']]],\n ['glstenciltablessbo_282', ['GLStencilTableSSBO', ['../a01209.html#a27a911fea9736fbe935c0edc4de9b998', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GLStencilTableSSBO(Far::LimitStencilTable const *limitStencilTable)'],\n ['../a01209.html#aaacee6173a99bb3d7bbfb0a352ef9828', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableSSBO::GLStencilTableSSBO(Far::StencilTable const *stencilTable)'],\n ['../a01209.html', 1, 'GLStencilTableSSBO']\n ]],\n ['glstenciltabletbo_283', ['GLStencilTableTBO', ['../a01241.html#abbba46ba67be00a0eac751e3e731db6e', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GLStencilTableTBO(Far::StencilTable const *stencilTable)'],\n ['../a01241.html#ac74a769daf1fe45575a16b738bb761cd', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLStencilTableTBO::GLStencilTableTBO(Far::LimitStencilTable const *limitStencilTable)'],\n ['../a01241.html', 1, 'GLStencilTableTBO']\n ]],\n ['glvertexbuffer_284', ['GLVertexBuffer', ['../a01237.html#aa6f38a53fba4a64d600b9c4695477049', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLVertexBuffer::GLVertexBuffer()'],\n ['../a01237.html', 1, 'GLVertexBuffer']\n ]],\n- ['glvertexbuffer_2eh_285', ['glVertexBuffer.h', ['../a00857.html', 1, '']]],\n+ ['glvertexbuffer_2eh_285', ['glVertexBuffer.h', ['../a00866.html', 1, '']]],\n ['glxfbevaluator_286', ['GLXFBEvaluator', ['../a01245.html#a1932f725b28e271ab11ca768ba3bb01e', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::GLXFBEvaluator::GLXFBEvaluator()'],\n ['../a01245.html', 1, 'GLXFBEvaluator']\n ]],\n- ['glxfbevaluator_2eh_287', ['glXFBEvaluator.h', ['../a00851.html', 1, '']]],\n+ ['glxfbevaluator_2eh_287', ['glXFBEvaluator.h', ['../a00845.html', 1, '']]],\n ['gregory_288', ['GREGORY', ['../a00973.html#a1d1cfd8ffb84e947f82999c682b666a7ae765e809376e5557d348d3e991a7273e', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor']]],\n ['gregory_5fbasis_289', ['GREGORY_BASIS', ['../a00973.html#a1d1cfd8ffb84e947f82999c682b666a7ad7d1a88a40ae8938e07af7a7745054b0', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor']]],\n ['gregory_5fboundary_290', ['GREGORY_BOUNDARY', ['../a00973.html#a1d1cfd8ffb84e947f82999c682b666a7ac4ae10df9c13e44880ca7be341e44f2a', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor']]],\n ['gregory_5ftriangle_291', ['GREGORY_TRIANGLE', ['../a00973.html#a1d1cfd8ffb84e947f82999c682b666a7a8bf2cfb939c78155cae7549621ea73f6', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_8.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_8.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -3,10 +3,10 @@\n ['handle_1', ['handle', ['../a01313.html#a06ae6afa9ea8e2028021f8ee3f28b824', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord']]],\n ['hasedgesharpness_2', ['HasEdgeSharpness', ['../a00969.html#a2c6d182581c88931784f62fe2b8ae666', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::VertexDescriptor']]],\n ['hasholes_3', ['HasHoles', ['../a01113.html#a1b68f0501aabb94ba410e006181d2cc5', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner']]],\n ['hasincidentfacesizes_4', ['HasIncidentFaceSizes', ['../a00969.html#ad56c50bc5582768806d7890c55b63141', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::VertexDescriptor']]],\n ['hassubfaces_5', ['HasSubFaces', ['../a00921.html#adc4cd156a8fdb843b0e233a922bb8a17', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Parameterization']]],\n ['hasvertexsharpness_6', ['HasVertexSharpness', ['../a00969.html#a4f46be1cfd41b34d8808363c83e9ff56', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::VertexDescriptor']]],\n ['hlslpatchshadersource_7', ['HLSLPatchShaderSource', ['../a01257.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n- ['hlslpatchshadersource_2eh_8', ['hlslPatchShaderSource.h', ['../a00833.html', 1, '']]],\n+ ['hlslpatchshadersource_2eh_8', ['hlslPatchShaderSource.h', ['../a00827.html', 1, '']]],\n ['holeindices_9', ['holeIndices', ['../a01097.html#a4fcd8e9a636d89045d364607f2b035a1', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyDescriptor']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_a.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_a.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,10 +1,10 @@\n var searchData = [\n ['length_0', ['length', ['../a01133.html#a9f59b34b1f25fe00023291b678246bcc', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor']]],\n- ['level_2eh_1', ['level.h', ['../a00752.html', 1, '']]],\n+ ['level_2eh_1', ['level.h', ['../a00755.html', 1, '']]],\n ['limit_2', ['Limit', ['../a01021.html#a2cf4c50a1117c44ba4d11db92b067401', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Limit(T const &src, U &dstPos, U1 &dstTan1, U2 &dstTan2) const'],\n ['../a01021.html#a0357839a7fabaa02f1d6d6a3c75e1ec2', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::Limit(T const &src, U &dstPos) const']\n ]],\n ['limitfacevarying_3', ['LimitFaceVarying', ['../a01021.html#afdee0e9e9c6edd34c96892efff808a87', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal']]],\n ['limits_4', ['Limits', ['../a00917.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr']]],\n ['limits_2eh_5', ['limits.h', ['../a00695.html', 1, '']]],\n ['limitstencil_6', ['LimitStencil', ['../a01065.html#a51e1d601373aa0cd9ce70b630ce22adf', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencil::LimitStencil(int *size, Index *indices, float *weights, float *duWeights=0, float *dvWeights=0, float *duuWeights=0, float *duvWeights=0, float *dvvWeights=0)'],\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_b.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_b.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -26,25 +26,25 @@\n ['mode_20', ['Mode', ['../a01037.html#a46c8a310cf4c094f8c80e1cb8dc1f911', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::StencilTableFactoryReal::Mode()'],\n ['../a01041.html#a46c8a310cf4c094f8c80e1cb8dc1f911', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::Mode()']\n ]],\n ['mtlcommon_2eh_21', ['mtlCommon.h', ['../a00806.html', 1, '']]],\n ['mtlcomputeevaluator_22', ['MTLComputeEvaluator', ['../a01285.html#a10015b2c4f90f3a11e60cc88dcbb9d6b', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLComputeEvaluator::MTLComputeEvaluator()'],\n ['../a01285.html', 1, 'MTLComputeEvaluator']\n ]],\n- ['mtlcomputeevaluator_2eh_23', ['mtlComputeEvaluator.h', ['../a00824.html', 1, '']]],\n+ ['mtlcomputeevaluator_2eh_23', ['mtlComputeEvaluator.h', ['../a00821.html', 1, '']]],\n ['mtlcontext_24', ['MTLContext', ['../a01277.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n ['mtllegacygregorypatchtable_25', ['MTLLegacyGregoryPatchTable', ['../a01289.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n- ['mtllegacygregorypatchtable_2eh_26', ['mtlLegacyGregoryPatchTable.h', ['../a00818.html', 1, '']]],\n- ['mtlmesh_2eh_27', ['mtlMesh.h', ['../a00812.html', 1, '']]],\n+ ['mtllegacygregorypatchtable_2eh_26', ['mtlLegacyGregoryPatchTable.h', ['../a00815.html', 1, '']]],\n+ ['mtlmesh_2eh_27', ['mtlMesh.h', ['../a00818.html', 1, '']]],\n ['mtlmeshinterface_28', ['MTLMeshInterface', ['../a00912.html#aa41e5b43fe46eeb0f97ff6ebdb1935aa', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n ['mtlpatchshadersource_29', ['MTLPatchShaderSource', ['../a01293.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n ['mtlpatchshadersource_2eh_30', ['mtlPatchShaderSource.h', ['../a00809.html', 1, '']]],\n ['mtlpatchtable_31', ['MTLPatchTable', ['../a01297.html#a3f197c2ae3862390d72e7443a2525187', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLPatchTable::MTLPatchTable()'],\n ['../a01297.html', 1, 'MTLPatchTable']\n ]],\n- ['mtlpatchtable_2eh_32', ['mtlPatchTable.h', ['../a00815.html', 1, '']]],\n+ ['mtlpatchtable_2eh_32', ['mtlPatchTable.h', ['../a00812.html', 1, '']]],\n ['mtlstenciltable_33', ['MTLStencilTable', ['../a01281.html#a015dc72c3360b4841ee2b3feb06c106f', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::MTLStencilTable(Far::LimitStencilTable const *stencilTable, MTLContext *context)'],\n ['../a01281.html#a45e329ffe9fe4e32a1b21bfc89a6763f', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::MTLStencilTable::MTLStencilTable(Far::StencilTable const *stencilTable, MTLContext *context)'],\n ['../a01281.html', 1, 'MTLStencilTable']\n ]],\n- ['mtlvertexbuffer_2eh_34', ['mtlVertexBuffer.h', ['../a00821.html', 1, '']]]\n+ ['mtlvertexbuffer_2eh_34', ['mtlVertexBuffer.h', ['../a00824.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_d.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_d.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -4,17 +4,17 @@\n ['offset_2', ['offset', ['../a01133.html#aed7ea92f45bd273dde380a45ddced592', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor']]],\n ['ompevalstencils_3', ['OmpEvalStencils', ['../a00912.html#a907dc53421460557871ddc35f77751f5', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvalStencils(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, int start, int end)'],\n ['../a00912.html#a44a05502feb5015035bc81b93cffb99c', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvalStencils(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, float *dstDu, BufferDescriptor const &dstDuDesc, float *dstDv, BufferDescriptor const &dstDvDesc, float *dstDuu, BufferDescriptor const &dstDuuDesc, float *dstDuv, BufferDescriptor const &dstDuvDesc, float *dstDvv, BufferDescriptor const &dstDvvDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, float const *duWeights, float const *dvWeights, float const *duuWeights, float const *duvWeights, float const *dvvWeights, int start, int end)'],\n ['../a00912.html#a870745ff3e883b32547d4db42b3be2bd', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::OmpEvalStencils(float const *src, BufferDescriptor const &srcDesc, float *dst, BufferDescriptor const &dstDesc, int const *sizes, int const *offsets, int const *indices, float const *weights, int start, int end)']\n ]],\n ['ompevaluator_4', ['OmpEvaluator', ['../a01305.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n ['ompevaluator_2eh_5', ['ompEvaluator.h', ['../a00878.html', 1, '']]],\n- ['ompkernel_2eh_6', ['ompKernel.h', ['../a00881.html', 1, '']]],\n- ['opencl_2eh_7', ['opencl.h', ['../a00794.html', 1, '']]],\n- ['opengl_2eh_8', ['opengl.h', ['../a00884.html', 1, '']]],\n+ ['ompkernel_2eh_6', ['ompKernel.h', ['../a00884.html', 1, '']]],\n+ ['opencl_2eh_7', ['opencl.h', ['../a00788.html', 1, '']]],\n+ ['opengl_2eh_8', ['opengl.h', ['../a00881.html', 1, '']]],\n ['opensubdiv_9', ['OpenSubdiv', ['../a00905.html', 1, '']]],\n ['opensubdiv_5fversion_10', ['OPENSUBDIV_VERSION', ['../a00906.html', 1, 'OpenSubdiv']]],\n ['operator_21_3d_11', ['operator!=', ['../a01133.html#a9f79e5aef7aa7198087e68c98c9804d3', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::BufferDescriptor']]],\n ['operator_3c_12', ['operator<', ['../a00973.html#abd9f5b68153db907eb7ccd0ed177dfdd', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor']]],\n ['operator_3d_13', ['operator=', ['../a00933.html#a85251d7cc5b0aa5838fa1767b6e31e5d', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Surface::operator=()'],\n ['../a00941.html#a5d089ab30f0402395635d7b389710594', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::SurfaceFactory::operator=()'],\n ['../a00921.html#af488514a0e7868dd273ebe767420c416', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Parameterization::operator=()'],\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/all_e.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/all_e.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -16,27 +16,27 @@\n ['../a01313.html#a5e6a814c3f060dc9575af4082f4d905e', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchCoord::PatchCoord()']\n ]],\n ['patchdescriptor_6', ['PatchDescriptor', ['../a00973.html', 1, 'PatchDescriptor'],\n ['../a00973.html#a6c6de8b1c8212039f4702f70bb75cf02', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::PatchDescriptor(PatchDescriptor const &d)'],\n ['../a00973.html#a8e33c50700963e8dc99a5f598a32aaa3', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::PatchDescriptor(int type)'],\n ['../a00973.html#af3f91a4a134cb2e5786990ec81f30d17', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::PatchDescriptor()']\n ]],\n- ['patchdescriptor_2eh_7', ['patchDescriptor.h', ['../a00740.html', 1, '']]],\n+ ['patchdescriptor_2eh_7', ['patchDescriptor.h', ['../a00737.html', 1, '']]],\n ['patchfacetag_8', ['PatchFaceTag', ['../a01017.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTableFactory']]],\n ['patchhandle_9', ['PatchHandle', ['../a00997.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable']]],\n ['patchindex_10', ['patchIndex', ['../a00997.html#a31bdf15525ab73560c2b725ff7f8ae3a', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::PatchHandle']]],\n ['patchmap_11', ['PatchMap', ['../a00977.html', 1, 'PatchMap'],\n ['../a00997.html#aff380d8d7f072959898c0a4e25150d2a', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::PatchHandle::PatchMap()'],\n ['../a00977.html#a016619de52fd79dbcdb879cc12b79200', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchMap::PatchMap()']\n ]],\n ['patchmap_2eh_12', ['patchMap.h', ['../a00734.html', 1, '']]],\n ['patchparam_13', ['PatchParam', ['../a00989.html', 1, 'PatchParam'],\n ['../a01321.html', 1, 'PatchParam']\n ]],\n- ['patchparam_2eh_14', ['patchParam.h', ['../a00737.html', 1, '']]],\n+ ['patchparam_2eh_14', ['patchParam.h', ['../a00740.html', 1, '']]],\n ['patchparamarray_15', ['PatchParamArray', ['../a00908.html#abe07210a029a01571c54276ae28b4d88', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far']]],\n ['patchparamtable_16', ['PatchParamTable', ['../a00908.html#a46ea68db89cc3d55c4191ef7648281ce', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far']]],\n ['patchparamvector_17', ['PatchParamVector', ['../a00912.html#a1e22001c627b0364d6c6d62a3ed05fee', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd']]],\n ['patchprecisiondouble_18', ['patchPrecisionDouble', ['../a01013.html#a91ce5fe0ce9b9ec6c5afcfb6e1ccaa40', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTableFactory::Options']]],\n ['patchtable_19', ['PatchTable', ['../a00993.html', 1, 'PatchTable'],\n ['../a00997.html#ad4867d502f9704b7e45a7974e8918dc3', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::PatchHandle::PatchTable()'],\n ['../a00993.html#a37e24b0244219eef61cc1ab2a5a5450e', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::PatchTable(int maxvalence)'],\n@@ -69,24 +69,24 @@\n ['preservequads_36', ['PreserveQuads', ['../a00965.html#a98820ec764e776bd556a9b80b872b78f', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Tessellation::Options::PreserveQuads(bool on)'],\n ['../a00965.html#ae2ec197b850e8bd12878cc67a503a07a', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Bfr::Tessellation::Options::PreserveQuads() const']\n ]],\n ['primitiveidbase_37', ['primitiveIdBase', ['../a01317.html#a5a054ce6862ecc6ea8c404fadb1c9a15', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Osd::PatchArray']]],\n ['primvarrefiner_38', ['PrimvarRefiner', ['../a01029.html', 1, 'PrimvarRefiner'],\n ['../a01029.html#a643d81a2ae20ba1897433f5af5a3697a', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefiner::PrimvarRefiner()']\n ]],\n- ['primvarrefiner_2eh_39', ['primvarRefiner.h', ['../a00719.html', 1, '']]],\n+ ['primvarrefiner_2eh_39', ['primvarRefiner.h', ['../a00725.html', 1, '']]],\n ['primvarrefinerreal_40', ['PrimvarRefinerReal', ['../a01021.html', 1, 'PrimvarRefinerReal< REAL >'],\n ['../a01113.html#a2cb48b58c1c2ecef25abb39339a95ac1', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::PrimvarRefinerReal()'],\n ['../a01021.html#a1fe1fbda93c01c326f13929d1c3c6247', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PrimvarRefinerReal::PrimvarRefinerReal()']\n ]],\n ['primvarrefinerreal_3c_20float_20_3e_41', ['PrimvarRefinerReal< float >', ['../a01021.html', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far']]],\n ['print_42', ['print', ['../a00973.html#acd0f82d3c49b8a5dd436d4003c44b713', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchDescriptor::print()'],\n ['../a00993.html#acd0f82d3c49b8a5dd436d4003c44b713', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PatchTable::print()']\n ]],\n ['printtopology_43', ['PrintTopology', ['../a01105.html#a3f963f5352658762e7a2e8d82fb888a0', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyLevel']]],\n ['ptexidx_44', ['ptexIdx', ['../a01085.html#afb60a8380cb9df186de7183b32bd84e2', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::LimitStencilTableFactoryReal::LocationArray']]],\n ['ptexindices_45', ['PtexIndices', ['../a01033.html', 1, 'PtexIndices'],\n ['../a01113.html#a3e6c896f40ac50aec4b28f19bc4eae9c', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::TopologyRefiner::PtexIndices()'],\n ['../a01033.html#a7b5dd5f4c76d684d924adf477eeea0a3', 1, 'OpenSubdiv::OPENSUBDIV_VERSION::Far::PtexIndices::PtexIndices()']\n ]],\n- ['ptexindices_2eh_46', ['ptexIndices.h', ['../a00716.html', 1, '']]]\n+ ['ptexindices_2eh_46', ['ptexIndices.h', ['../a00722.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_1.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_1.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,4 @@\n var searchData = [\n- ['bilinearscheme_2eh_0', ['bilinearScheme.h', ['../a00782.html', 1, '']]],\n- ['bufferdescriptor_2eh_1', ['bufferDescriptor.h', ['../a00896.html', 1, '']]]\n+ ['bilinearscheme_2eh_0', ['bilinearScheme.h', ['../a00785.html', 1, '']]],\n+ ['bufferdescriptor_2eh_1', ['bufferDescriptor.h', ['../a00899.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_2.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_2.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,16 +1,16 @@\n var searchData = [\n- ['catmarkscheme_2eh_0', ['catmarkScheme.h', ['../a00785.html', 1, '']]],\n- ['clevaluator_2eh_1', ['clEvaluator.h', ['../a00800.html', 1, '']]],\n- ['clpatchtable_2eh_2', ['clPatchTable.h', ['../a00803.html', 1, '']]],\n+ ['catmarkscheme_2eh_0', ['catmarkScheme.h', ['../a00782.html', 1, '']]],\n+ ['clevaluator_2eh_1', ['clEvaluator.h', ['../a00803.html', 1, '']]],\n+ ['clpatchtable_2eh_2', ['clPatchTable.h', ['../a00800.html', 1, '']]],\n ['clvertexbuffer_2eh_3', ['clVertexBuffer.h', ['../a00797.html', 1, '']]],\n ['componentinterfaces_2eh_4', ['componentInterfaces.h', ['../a00767.html', 1, '']]],\n- ['cpud3d11vertexbuffer_2eh_5', ['cpuD3D11VertexBuffer.h', ['../a00842.html', 1, '']]],\n- ['cpuevaluator_2eh_6', ['cpuEvaluator.h', ['../a00899.html', 1, '']]],\n+ ['cpud3d11vertexbuffer_2eh_5', ['cpuD3D11VertexBuffer.h', ['../a00848.html', 1, '']]],\n+ ['cpuevaluator_2eh_6', ['cpuEvaluator.h', ['../a00896.html', 1, '']]],\n ['cpuglvertexbuffer_2eh_7', ['cpuGLVertexBuffer.h', ['../a00872.html', 1, '']]],\n- ['cpupatchtable_2eh_8', ['cpuPatchTable.h', ['../a00890.html', 1, '']]],\n- ['cpuvertexbuffer_2eh_9', ['cpuVertexBuffer.h', ['../a00893.html', 1, '']]],\n+ ['cpupatchtable_2eh_8', ['cpuPatchTable.h', ['../a00893.html', 1, '']]],\n+ ['cpuvertexbuffer_2eh_9', ['cpuVertexBuffer.h', ['../a00890.html', 1, '']]],\n ['crease_2eh_10', ['crease.h', ['../a00779.html', 1, '']]],\n- ['cudaevaluator_2eh_11', ['cudaEvaluator.h', ['../a00791.html', 1, '']]],\n- ['cudapatchtable_2eh_12', ['cudaPatchTable.h', ['../a00788.html', 1, '']]],\n+ ['cudaevaluator_2eh_11', ['cudaEvaluator.h', ['../a00794.html', 1, '']]],\n+ ['cudapatchtable_2eh_12', ['cudaPatchTable.h', ['../a00791.html', 1, '']]],\n ['cudavertexbuffer_2eh_13', ['cudaVertexBuffer.h', ['../a00902.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_3.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_3.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,7 +1,7 @@\n var searchData = [\n- ['d3d11computeevaluator_2eh_0', ['d3d11ComputeEvaluator.h', ['../a00839.html', 1, '']]],\n- ['d3d11legacygregorypatchtable_2eh_1', ['d3d11LegacyGregoryPatchTable.h', ['../a00836.html', 1, '']]],\n- ['d3d11mesh_2eh_2', ['d3d11Mesh.h', ['../a00827.html', 1, '']]],\n- ['d3d11patchtable_2eh_3', ['d3d11PatchTable.h', ['../a00845.html', 1, '']]],\n- ['d3d11vertexbuffer_2eh_4', ['d3d11VertexBuffer.h', ['../a00830.html', 1, '']]]\n+ ['d3d11computeevaluator_2eh_0', ['d3d11ComputeEvaluator.h', ['../a00842.html', 1, '']]],\n+ ['d3d11legacygregorypatchtable_2eh_1', ['d3d11LegacyGregoryPatchTable.h', ['../a00839.html', 1, '']]],\n+ ['d3d11mesh_2eh_2', ['d3d11Mesh.h', ['../a00830.html', 1, '']]],\n+ ['d3d11patchtable_2eh_3', ['d3d11PatchTable.h', ['../a00836.html', 1, '']]],\n+ ['d3d11vertexbuffer_2eh_4', ['d3d11VertexBuffer.h', ['../a00833.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_6.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_6.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,9 +1,9 @@\n var searchData = [\n- ['glcomputeevaluator_2eh_0', ['glComputeEvaluator.h', ['../a00848.html', 1, '']]],\n- ['gllegacygregorypatchtable_2eh_1', ['glLegacyGregoryPatchTable.h', ['../a00866.html', 1, '']]],\n- ['glmesh_2eh_2', ['glMesh.h', ['../a00854.html', 1, '']]],\n- ['glpatchtable_2eh_3', ['glPatchTable.h', ['../a00863.html', 1, '']]],\n- ['glslpatchshadersource_2eh_4', ['glslPatchShaderSource.h', ['../a00860.html', 1, '']]],\n- ['glvertexbuffer_2eh_5', ['glVertexBuffer.h', ['../a00857.html', 1, '']]],\n- ['glxfbevaluator_2eh_6', ['glXFBEvaluator.h', ['../a00851.html', 1, '']]]\n+ ['glcomputeevaluator_2eh_0', ['glComputeEvaluator.h', ['../a00851.html', 1, '']]],\n+ ['gllegacygregorypatchtable_2eh_1', ['glLegacyGregoryPatchTable.h', ['../a00863.html', 1, '']]],\n+ ['glmesh_2eh_2', ['glMesh.h', ['../a00857.html', 1, '']]],\n+ ['glpatchtable_2eh_3', ['glPatchTable.h', ['../a00860.html', 1, '']]],\n+ ['glslpatchshadersource_2eh_4', ['glslPatchShaderSource.h', ['../a00854.html', 1, '']]],\n+ ['glvertexbuffer_2eh_5', ['glVertexBuffer.h', ['../a00866.html', 1, '']]],\n+ ['glxfbevaluator_2eh_6', ['glXFBEvaluator.h', ['../a00845.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_7.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_7.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,3 +1,3 @@\n var searchData = [\n- ['hlslpatchshadersource_2eh_0', ['hlslPatchShaderSource.h', ['../a00833.html', 1, '']]]\n+ ['hlslpatchshadersource_2eh_0', ['hlslPatchShaderSource.h', ['../a00827.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_8.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_8.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,5 +1,5 @@\n var searchData = [\n- ['level_2eh_0', ['level.h', ['../a00752.html', 1, '']]],\n+ ['level_2eh_0', ['level.h', ['../a00755.html', 1, '']]],\n ['limits_2eh_1', ['limits.h', ['../a00695.html', 1, '']]],\n ['loopscheme_2eh_2', ['loopScheme.h', ['../a00776.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_9.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_9.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,10 +1,10 @@\n var searchData = [\n ['mesh_2eh_0', ['mesh.h', ['../a00887.html', 1, '']]],\n ['mtlcommon_2eh_1', ['mtlCommon.h', ['../a00806.html', 1, '']]],\n- ['mtlcomputeevaluator_2eh_2', ['mtlComputeEvaluator.h', ['../a00824.html', 1, '']]],\n- ['mtllegacygregorypatchtable_2eh_3', ['mtlLegacyGregoryPatchTable.h', ['../a00818.html', 1, '']]],\n- ['mtlmesh_2eh_4', ['mtlMesh.h', ['../a00812.html', 1, '']]],\n+ ['mtlcomputeevaluator_2eh_2', ['mtlComputeEvaluator.h', ['../a00821.html', 1, '']]],\n+ ['mtllegacygregorypatchtable_2eh_3', ['mtlLegacyGregoryPatchTable.h', ['../a00815.html', 1, '']]],\n+ ['mtlmesh_2eh_4', ['mtlMesh.h', ['../a00818.html', 1, '']]],\n ['mtlpatchshadersource_2eh_5', ['mtlPatchShaderSource.h', ['../a00809.html', 1, '']]],\n- ['mtlpatchtable_2eh_6', ['mtlPatchTable.h', ['../a00815.html', 1, '']]],\n- ['mtlvertexbuffer_2eh_7', ['mtlVertexBuffer.h', ['../a00821.html', 1, '']]]\n+ ['mtlpatchtable_2eh_6', ['mtlPatchTable.h', ['../a00812.html', 1, '']]],\n+ ['mtlvertexbuffer_2eh_7', ['mtlVertexBuffer.h', ['../a00824.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_a.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_a.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,7 +1,7 @@\n var searchData = [\n ['ompevaluator_2eh_0', ['ompEvaluator.h', ['../a00878.html', 1, '']]],\n- ['ompkernel_2eh_1', ['ompKernel.h', ['../a00881.html', 1, '']]],\n- ['opencl_2eh_2', ['opencl.h', ['../a00794.html', 1, '']]],\n- ['opengl_2eh_3', ['opengl.h', ['../a00884.html', 1, '']]],\n+ ['ompkernel_2eh_1', ['ompKernel.h', ['../a00884.html', 1, '']]],\n+ ['opencl_2eh_2', ['opencl.h', ['../a00788.html', 1, '']]],\n+ ['opengl_2eh_3', ['opengl.h', ['../a00881.html', 1, '']]],\n ['options_2eh_4', ['options.h', ['../a00773.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_b.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_b.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,10 +1,10 @@\n var searchData = [\n ['parameterization_2eh_0', ['parameterization.h', ['../a00698.html', 1, '']]],\n- ['patchdescriptor_2eh_1', ['patchDescriptor.h', ['../a00740.html', 1, '']]],\n+ ['patchdescriptor_2eh_1', ['patchDescriptor.h', ['../a00737.html', 1, '']]],\n ['patchmap_2eh_2', ['patchMap.h', ['../a00734.html', 1, '']]],\n- ['patchparam_2eh_3', ['patchParam.h', ['../a00737.html', 1, '']]],\n+ ['patchparam_2eh_3', ['patchParam.h', ['../a00740.html', 1, '']]],\n ['patchtable_2eh_4', ['patchTable.h', ['../a00731.html', 1, '']]],\n ['patchtablefactory_2eh_5', ['patchTableFactory.h', ['../a00728.html', 1, '']]],\n- ['primvarrefiner_2eh_6', ['primvarRefiner.h', ['../a00719.html', 1, '']]],\n- ['ptexindices_2eh_7', ['ptexIndices.h', ['../a00716.html', 1, '']]]\n+ ['primvarrefiner_2eh_6', ['primvarRefiner.h', ['../a00725.html', 1, '']]],\n+ ['ptexindices_2eh_7', ['ptexIndices.h', ['../a00722.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_c.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_c.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,4 +1,4 @@\n var searchData = [\n- ['refinement_2eh_0', ['refinement.h', ['../a00755.html', 1, '']]],\n- ['refinersurfacefactory_2eh_1', ['refinerSurfaceFactory.h', ['../a00689.html', 1, '']]]\n+ ['refinement_2eh_0', ['refinement.h', ['../a00752.html', 1, '']]],\n+ ['refinersurfacefactory_2eh_1', ['refinerSurfaceFactory.h', ['../a00692.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_d.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_d.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,11 +1,11 @@\n var searchData = [\n ['scheme_2eh_0', ['scheme.h', ['../a00770.html', 1, '']]],\n ['sparseselector_2eh_1', ['sparseSelector.h', ['../a00749.html', 1, '']]],\n ['stackbuffer_2eh_2', ['stackBuffer.h', ['../a00746.html', 1, '']]],\n- ['stenciltable_2eh_3', ['stencilTable.h', ['../a00725.html', 1, '']]],\n- ['stenciltablefactory_2eh_4', ['stencilTableFactory.h', ['../a00710.html', 1, '']]],\n- ['surface_2eh_5', ['surface.h', ['../a00692.html', 1, '']]],\n+ ['stenciltable_2eh_3', ['stencilTable.h', ['../a00713.html', 1, '']]],\n+ ['stenciltablefactory_2eh_4', ['stencilTableFactory.h', ['../a00716.html', 1, '']]],\n+ ['surface_2eh_5', ['surface.h', ['../a00689.html', 1, '']]],\n ['surfacefactory_2eh_6', ['surfaceFactory.h', ['../a00686.html', 1, '']]],\n ['surfacefactorycache_2eh_7', ['surfaceFactoryCache.h', ['../a00680.html', 1, '']]],\n ['surfacefactorymeshadapter_2eh_8', ['surfaceFactoryMeshAdapter.h', ['../a00683.html', 1, '']]]\n ];\n"}]}, {"source1": "./usr/share/doc/opensubdiv/doxy_html/search/files_e.js", "source2": "./usr/share/doc/opensubdiv/doxy_html/search/files_e.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,13 +1,13 @@\n var searchData = [\n- ['tbbevaluator_2eh_0', ['tbbEvaluator.h', ['../a00875.html', 1, '']]],\n- ['tbbkernel_2eh_1', ['tbbKernel.h', ['../a00869.html', 1, '']]],\n+ ['tbbevaluator_2eh_0', ['tbbEvaluator.h', ['../a00869.html', 1, '']]],\n+ ['tbbkernel_2eh_1', ['tbbKernel.h', ['../a00875.html', 1, '']]],\n ['tessellation_2eh_2', ['tessellation.h', ['../a00677.html', 1, '']]],\n- ['topologydescriptor_2eh_3', ['topologyDescriptor.h', ['../a00713.html', 1, '']]],\n- ['topologylevel_2eh_4', ['topologyLevel.h', ['../a00722.html', 1, '']]],\n+ ['topologydescriptor_2eh_3', ['topologyDescriptor.h', ['../a00710.html', 1, '']]],\n+ ['topologylevel_2eh_4', ['topologyLevel.h', ['../a00719.html', 1, '']]],\n ['topologyrefiner_2eh_5', ['topologyRefiner.h', ['../a00707.html', 1, '']]],\n ['topologyrefinerfactory_2eh_6', ['topologyRefinerFactory.h', ['../a00704.html', 1, '']]],\n ['types_2eh_7', ['types.h', ['../a01478.html', 1, '(Global Namespace)'],\n ['../a01481.html', 1, '(Global Namespace)'],\n ['../a01484.html', 1, '(Global Namespace)'],\n ['../a01487.html', 1, '(Global Namespace)']\n ]]\n"}]}, {"source1": "./usr/share/doc/opensubdiv/dxptexviewer.html", "source2": "./usr/share/doc/opensubdiv/dxptexviewer.html", "unified_diff": "@@ -266,15 +266,15 @@\n

SEE ALSO

\n

Other examples :\n glViewer, glFVarViewer, glEvalLimit, glStencilViewer, glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -124,14 +124,14 @@\n esc : hide GUI\n f : fit frame\n +/- : increase / decrese tessellation rate\n **** SEE ALSO ****\n Other examples : glViewer, glFVarViewer, glEvalLimit, glStencilViewer,\n glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/dxviewer.html", "source2": "./usr/share/doc/opensubdiv/dxviewer.html", "unified_diff": "@@ -220,15 +220,15 @@\n

SEE ALSO

\n

Other examples :\n glViewer, glFVarViewer, glEvalLimit, glStencilViewer, glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -85,14 +85,14 @@\n **** OPTIONS ****\n See the description of the common_comand_line_options for the subset of common\n options supported here.\n **** SEE ALSO ****\n Other examples : glViewer, glFVarViewer, glEvalLimit, glStencilViewer,\n glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_overview.html", "source2": "./usr/share/doc/opensubdiv/far_overview.html", "unified_diff": "@@ -780,15 +780,15 @@\n controlStencils.UpdateDerivs<StencilType>( &controlPoints[0], &utan[0], &vtan[0] );\n \n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -475,14 +475,14 @@\n // Update points by applying stencils\n controlStencils.UpdateValues( &controlPoints[0], &points[0] );\n \n // Update tangents by applying derivative stencils\n controlStencils.UpdateDerivs( &controlPoints[0], &utan[0], &vtan\n [0] );\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_1_1.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_1_1.html", "unified_diff": "@@ -356,15 +356,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -243,14 +243,14 @@\n delete refiner;\n return EXIT_SUCCESS;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_1_2.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_1_2.html", "unified_diff": "@@ -461,15 +461,15 @@\n delete refiner;\n return EXIT_SUCCESS;\n }\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -345,14 +345,14 @@\n printf(\"\\n\");\n }\n \n delete refiner;\n return EXIT_SUCCESS;\n }\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_2_1.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_2_1.html", "unified_diff": "@@ -411,15 +411,15 @@\n return refiner;\n }\n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -302,14 +302,14 @@\n Far::TopologyRefinerFactory::Options(type, options));\n \n return refiner;\n }\n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_2_2.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_2_2.html", "unified_diff": "@@ -528,15 +528,15 @@\n return EXIT_SUCCESS;\n }\n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -425,14 +425,14 @@\n \n delete refiner;\n return EXIT_SUCCESS;\n }\n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_2_3.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_2_3.html", "unified_diff": "@@ -696,15 +696,15 @@\n return EXIT_SUCCESS;\n }\n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -613,14 +613,14 @@\n \n delete refiner;\n return EXIT_SUCCESS;\n }\n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_3_1.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_3_1.html", "unified_diff": "@@ -662,15 +662,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -562,14 +562,14 @@\n delete refiner;\n return EXIT_SUCCESS;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_4_1.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_4_1.html", "unified_diff": "@@ -351,15 +351,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -237,14 +237,14 @@\n options));\n \n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_4_2.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_4_2.html", "unified_diff": "@@ -405,15 +405,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -295,14 +295,14 @@\n \n return refiner;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_4_3.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_4_3.html", "unified_diff": "@@ -405,15 +405,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -293,14 +293,14 @@\n Far::TopologyRefinerFactory::Options(type, options));\n \n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_5_1.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_5_1.html", "unified_diff": "@@ -504,15 +504,15 @@\n \n return refiner;\n }\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -395,14 +395,14 @@\n Far::TopologyRefiner * refiner =\n Far::TopologyRefinerFactory::Create(desc,\n Far::TopologyRefinerFactory::Options(type, options));\n \n return refiner;\n }\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_5_2.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_5_2.html", "unified_diff": "@@ -826,15 +826,15 @@\n \n return EXIT_SUCCESS;\n }\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -706,14 +706,14 @@\n }\n }\n delete baseRefinerPtr;\n \n return EXIT_SUCCESS;\n }\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/far_tutorial_5_3.html", "source2": "./usr/share/doc/opensubdiv/far_tutorial_5_3.html", "unified_diff": "@@ -745,15 +745,15 @@\n \n return EXIT_SUCCESS;\n }\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -629,14 +629,14 @@\n delete refinerPtr;\n delete patchTablePtr;\n delete limitStencilTablePtr;\n \n return EXIT_SUCCESS;\n }\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/getting_started.html", "source2": "./usr/share/doc/opensubdiv/getting_started.html", "unified_diff": "@@ -321,15 +321,15 @@\n \n \n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -144,14 +144,14 @@\n |./documentation/|The reStructuredText source files along with python scripts|\n |________________|that_generate_the_HTML_documentation_site._________________|\n | |Standalone regression tests and baseline data to help |\n |./regression/ |maintain the integrity of our APIs. If GPU SDKs are |\n | |detected, some tests will attempt to run computations on |\n |________________|those_GPUs.________________________________________________|\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/glevallimit.html", "source2": "./usr/share/doc/opensubdiv/glevallimit.html", "unified_diff": "@@ -237,15 +237,15 @@\n

SEE ALSO

\n

Other examples :\n glViewer, glFVarViewer, glEvalLimit, glStencilViewer, glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -102,14 +102,14 @@\n **** OPTIONS ****\n See the description of the common_comand_line_options for the subset of common\n options supported here.\n **** SEE ALSO ****\n Other examples : glViewer, glFVarViewer, glEvalLimit, glStencilViewer,\n glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/glfvarviewer.html", "source2": "./usr/share/doc/opensubdiv/glfvarviewer.html", "unified_diff": "@@ -227,15 +227,15 @@\n

SEE ALSO

\n

Other examples :\n glViewer, glFVarViewer, glEvalLimit, glStencilViewer, glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -89,14 +89,14 @@\n **** OPTIONS ****\n See the description of the common_comand_line_options for the subset of common\n options supported here.\n **** SEE ALSO ****\n Other examples : glViewer, glFVarViewer, glEvalLimit, glStencilViewer,\n glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/glpainttest.html", "source2": "./usr/share/doc/opensubdiv/glpainttest.html", "unified_diff": "@@ -229,15 +229,15 @@\n

SEE ALSO

\n

Other examples :\n glViewer, glFVarViewer, glEvalLimit, glStencilViewer, glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -88,14 +88,14 @@\n **** KEYBOARD CONTROLS ****\n c : use texture as color\n d : use texture as displacement\n **** SEE ALSO ****\n Other examples : glViewer, glFVarViewer, glEvalLimit, glStencilViewer,\n glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/glptexviewer.html", "source2": "./usr/share/doc/opensubdiv/glptexviewer.html", "unified_diff": "@@ -266,15 +266,15 @@\n

SEE ALSO

\n

Other examples :\n glViewer, glFVarViewer, glEvalLimit, glStencilViewer, glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -124,14 +124,14 @@\n +/- : increase / decrease tessellation rate\n r : reload and re-compile the shader files\n e : draw normals\n **** SEE ALSO ****\n Other examples : glViewer, glFVarViewer, glEvalLimit, glStencilViewer,\n glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/glsharetopology.html", "source2": "./usr/share/doc/opensubdiv/glsharetopology.html", "unified_diff": "@@ -218,15 +218,15 @@\n

SEE ALSO

\n

Other examples :\n glViewer, glFVarViewer, glEvalLimit, glStencilViewer, glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -84,14 +84,14 @@\n **** OPTIONS ****\n See the description of the common_comand_line_options for the subset of common\n options supported here.\n **** SEE ALSO ****\n Other examples : glViewer, glFVarViewer, glEvalLimit, glStencilViewer,\n glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/glstencilviewer.html", "source2": "./usr/share/doc/opensubdiv/glstencilviewer.html", "unified_diff": "@@ -220,15 +220,15 @@\n

SEE ALSO

\n

Other examples :\n glViewer, glFVarViewer, glEvalLimit, glStencilViewer, glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -85,14 +85,14 @@\n **** OPTIONS ****\n See the description of the common_comand_line_options for the subset of common\n options supported here.\n **** SEE ALSO ****\n Other examples : glViewer, glFVarViewer, glEvalLimit, glStencilViewer,\n glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/glviewer.html", "source2": "./usr/share/doc/opensubdiv/glviewer.html", "unified_diff": "@@ -220,15 +220,15 @@\n

SEE ALSO

\n

Other examples :\n glViewer, glFVarViewer, glEvalLimit, glStencilViewer, glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -85,14 +85,14 @@\n **** OPTIONS ****\n See the description of the common_comand_line_options for the subset of common\n options supported here.\n **** SEE ALSO ****\n Other examples : glViewer, glFVarViewer, glEvalLimit, glStencilViewer,\n glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/hbr_overview.html", "source2": "./usr/share/doc/opensubdiv/hbr_overview.html", "unified_diff": "@@ -289,15 +289,15 @@\n \n \n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -127,14 +127,14 @@\n o the BoundaryNone rule skips all boundary vertices\n (results are ''undefined'')\n o the AlwaysSharp rule does not apply\n * face-varying boundaries:\n o the BoundaryNone rule selects bilinear face-varying\n interpolation\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/hbr_tutorial_0.html", "source2": "./usr/share/doc/opensubdiv/hbr_tutorial_0.html", "unified_diff": "@@ -328,15 +328,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -212,14 +212,14 @@\n delete hmesh;\n delete catmark;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/hbr_tutorial_1.html", "source2": "./usr/share/doc/opensubdiv/hbr_tutorial_1.html", "unified_diff": "@@ -363,15 +363,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -252,14 +252,14 @@\n delete hmesh;\n delete catmark;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/hbr_tutorial_2.html", "source2": "./usr/share/doc/opensubdiv/hbr_tutorial_2.html", "unified_diff": "@@ -430,15 +430,15 @@\n \n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -321,14 +321,14 @@\n return hmesh;\n }\n \n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/hedits.html", "source2": "./usr/share/doc/opensubdiv/hedits.html", "unified_diff": "@@ -303,15 +303,15 @@\n \n \"images/hedit_example5.png\"\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -161,14 +161,14 @@\n * The \"hole\" operation introduces holes (missing faces) into the\n subdivision mesh at any level in the subdivision hierarchy. The\n faces will be deleted, and none of their children will appear\n (you cannot \"unhole\" a face if any ancestor is a \"hole\"). This\n operation takes no float or string arguments.\n [images/hedit_example5.png]\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/intro.html", "source2": "./usr/share/doc/opensubdiv/intro.html", "unified_diff": "@@ -312,15 +312,15 @@\n
Pixar Research:
\n
Pixar R&D Portal
\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -168,14 +168,14 @@\n **** Contributing ****\n For details on how to contribute to OpenSubdiv, see the page on Contributing\n ===============================================================================\n **** External Resources ****\n Pixar Research:\n Pixar_R&D_Portal\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/license.html", "source2": "./usr/share/doc/opensubdiv/license.html", "unified_diff": "@@ -222,15 +222,15 @@\n KIND, either express or implied. See the Apache License for the specific\n language governing permissions and limitations under the Apache License.\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -98,14 +98,14 @@\n \n Unless required by applicable law or agreed to in writing, software\n distributed under the Apache License with the above modification is\n distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n KIND, either express or implied. See the Apache License for the specific\n language governing permissions and limitations under the Apache License.\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/mod_notes.html", "source2": "./usr/share/doc/opensubdiv/mod_notes.html", "unified_diff": "@@ -331,15 +331,15 @@\n the modeling techniques he uses when modeling props and architecture sets for\n feature films.

\n \"https://graphics.pixar.com/opensubdiv/videothumbnails/subdiv_modeling_pixar2013.png\"\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -172,14 +172,14 @@\n ===============================================================================\n Ivo Kos, Modelling Technical Director at Pixar Animation Studios, shows some of\n the modeling techniques he uses when modeling props and architecture sets for\n feature films.\n [https://graphics.pixar.com/opensubdiv/videothumbnails/\n subdiv_modeling_pixar2013.png]\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/mtlptexviewer.html", "source2": "./usr/share/doc/opensubdiv/mtlptexviewer.html", "unified_diff": "@@ -223,15 +223,15 @@\n

SEE ALSO

\n

Other examples :\n glViewer, glFVarViewer, glEvalLimit, glStencilViewer, glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -87,14 +87,14 @@\n q : quit\n f : fit frame\n +/- : increase / decrease tessellation rate\n **** SEE ALSO ****\n Other examples : glViewer, glFVarViewer, glEvalLimit, glStencilViewer,\n glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/mtlviewer.html", "source2": "./usr/share/doc/opensubdiv/mtlviewer.html", "unified_diff": "@@ -245,15 +245,15 @@\n

SEE ALSO

\n

Other examples :\n glViewer, glFVarViewer, glEvalLimit, glStencilViewer, glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -91,14 +91,14 @@\n q : quit\n f : fit frame\n +/- : increase / decrease tessellation rate\n **** SEE ALSO ****\n Other examples : glViewer, glFVarViewer, glEvalLimit, glStencilViewer,\n glPtexViewer, glPaintTest, glShareTopology, dxViewer, dxPtexViewer,\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/osd_overview.html", "source2": "./usr/share/doc/opensubdiv/osd_overview.html", "unified_diff": "@@ -625,15 +625,15 @@\n as the drawing API, Osd::CudaGLVertexBuffer fits the case since it implements\n BindCudaBuffer() and BindVBO(). Again, clients can implement their own buffer\n class and pass it to the Evaluators.

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -229,14 +229,14 @@\n The buffers can use these methods as a trigger of interop. Osd provides a\n default implementation of interop buffer for most of the backend combinations.\n For example, if the client wants to use CUDA as a computation backend and use\n OpenGL as the drawing API, Osd::CudaGLVertexBuffer fits the case since it\n implements BindCudaBuffer() and BindVBO(). Again, clients can implement their\n own buffer class and pass it to the Evaluators.\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/osd_shader_interface.html", "source2": "./usr/share/doc/opensubdiv/osd_shader_interface.html", "unified_diff": "@@ -461,15 +461,15 @@\n \n \n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -277,14 +277,14 @@\n out vec4 tessOuterLo, out vec4 tessOuterHi);\n void OsdEvalPatchBezierTriangleTessLevels(\n vec3 cv[15],\n ivec3 patchParam,\n out vec4 tessLevelOuter, out vec2 tessLevelInner,\n out vec4 tessOuterLo, out vec4 tessOuterHi);\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/osd_tutorial_0.html", "source2": "./usr/share/doc/opensubdiv/osd_tutorial_0.html", "unified_diff": "@@ -331,15 +331,15 @@\n }\n \n //------------------------------------------------------------------------------\n \n \n
\n
\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -215,14 +215,14 @@\n \n return refiner;\n }\n \n //-----------------------------------------------------------------------------\n -\n ===============================================================================\n-Generated on: 2024-01-12 13:27 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/porting.html", "source2": "./usr/share/doc/opensubdiv/porting.html", "unified_diff": "@@ -740,15 +740,15 @@\n flag "CMAKE_INCDIR_BASE" when configuring cmake (i.e.\n -DCMAKE_INCDIR_BASE=include/opensubdiv3) and then including files from\n "opensubdiv3/..." in client code.

\n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -391,14 +391,14 @@\n build-time search paths.\n To support both OpenSubdiv 2.0 and 3.0 in your build environment, you can\n prefix the header install directory of OpenSubdiv 3.0. Do this using the build\n flag \"CMAKE_INCDIR_BASE\" when configuring cmake (i.e. -\n DCMAKE_INCDIR_BASE=include/opensubdiv3) and then including files from\n \"opensubdiv3/...\" in client code.\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/references.html", "source2": "./usr/share/doc/opensubdiv/references.html", "unified_diff": "@@ -280,15 +280,15 @@\n
http://doi.org/10.1016/0010-4485%2878%2990110-0
\n \n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -156,14 +156,14 @@\n http://doi.org/10.1145/166117.166121\n \n Recursively generated B-spline surfaces on arbitrary topological\n meshes\n Catmull, E.; Clark, J. Computer-Aided Design 10 (6) (1978)\n http://doi.org/10.1016/0010-4485%2878%2990110-0\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/release_30.html", "source2": "./usr/share/doc/opensubdiv/release_30.html", "unified_diff": "@@ -410,15 +410,15 @@\n and rules. See Subdivision Compatibility for a\n complete list.

\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -252,14 +252,14 @@\n Please see the Porting_Guide for help on how to port existing code written for\n OpenSubdiv 2.x to the new 3.0 release.\n ===============================================================================\n *** Subdivision Compatibility ***\n The 3.0 release has made some minor changes to the subdivision specification\n and rules. See Subdivision_Compatibility for a complete list.\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/release_31.html", "source2": "./usr/share/doc/opensubdiv/release_31.html", "unified_diff": "@@ -555,15 +555,15 @@\n \n \n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -285,14 +285,14 @@\n *** Bug Fixes ***\n * Fixed Ptex version parsing and compatibility issues\n * Fixed compatibility issues with VS2015\n * Fixed bug interpolating face-varying data with Bilinear scheme\n * Fixed bug with refinement using Chaikin creasing\n * Fixed bugs with HUD sliders in the example viewers\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/release_32.html", "source2": "./usr/share/doc/opensubdiv/release_32.html", "unified_diff": "@@ -343,15 +343,15 @@\n \n \n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -139,14 +139,14 @@\n * Corrected numerous spelling errors in doxygen comments\n * Updated glFVarViewer with improved error detection and command\n line parsing\n * Added option to build using MSVC with static CRT\n *** Bug Fixes ***\n * Fixed a double delete of GL program in Osd::GLComputeEvaluator\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/release_33.html", "source2": "./usr/share/doc/opensubdiv/release_33.html", "unified_diff": "@@ -369,15 +369,15 @@\n \n \n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -164,14 +164,14 @@\n **** Other Changes ****\n *** Improvements ***\n * Fixed several instances of local variable shadowing that could\n cause build warnings\n * Updated continuous-integration build scripts and added testing\n on macOS\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/release_34.html", "source2": "./usr/share/doc/opensubdiv/release_34.html", "unified_diff": "@@ -492,15 +492,15 @@\n \n \n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -257,14 +257,14 @@\n * Fixed Far::PrimvarRefiner internal limitFVar() prototype\n (GitHub #979)\n * Fixed Far::StencilTable append when base StencilTable empty\n (GitHub #982)\n * Patches around non-manifold vertices now free of cracks (GitHub\n #1013)\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/release_35.html", "source2": "./usr/share/doc/opensubdiv/release_35.html", "unified_diff": "@@ -375,15 +375,15 @@\n \n \n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -172,14 +172,14 @@\n * Fixed GLSL shader source to remove storage qualifiers from\n struct members (GitHub #1271)\n * Fixed use of CMake variables for Apple builds (GitHub #1235)\n * Fixed build errors when using OpenGL without GLFW (GitHub\n #1257)\n * Fixed links to embedded videos (GitHub #1231)\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/release_notes.html", "source2": "./usr/share/doc/opensubdiv/release_notes.html", "unified_diff": "@@ -720,15 +720,15 @@\n
\n

Previous 2.x Release Notes

\n

Previous releases

\n
\n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -452,14 +452,14 @@\n focus the API.\n * Interpolation of Vertex and Varying primvars in a single pass is no\n longer supported.\n * The Osd layer was largely refactored.\n **** Previous 2.x Release Notes ****\n Previous_releases\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/release_notes_2x.html", "source2": "./usr/share/doc/opensubdiv/release_notes_2x.html", "unified_diff": "@@ -888,15 +888,15 @@\n
\n

Release 1.0.0

\n

Oringal release:

\n
\n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -567,14 +567,14 @@\n **** Release 1.1.0 ****\n New Features\n * release initiated because of the switch to Git Flow\n ===============================================================================\n **** Release 1.0.0 ****\n Oringal release:\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/sdc_overview.html", "source2": "./usr/share/doc/opensubdiv/sdc_overview.html", "unified_diff": "@@ -501,15 +501,15 @@\n edge -- either the vertex opposite the edge for each triangle, or its center\n (which has no other use for Loop).

\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -341,14 +341,14 @@\n could be provided on the mask class, with an optional indication of the\n incident face topology for the irregular cases. The Loop scheme does not have\n \"face weights\", for a vertex-vertex mask, but for an edge-vertex mask it does\n require weights associated with the faces incident the edge -- either the\n vertex opposite the edge for each triangle, or its center (which has no other\n use for Loop).\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/subdivision_surfaces.html", "source2": "./usr/share/doc/opensubdiv/subdivision_surfaces.html", "unified_diff": "@@ -1225,15 +1225,15 @@\n at either boundary, it is impossible to join the meshes seamlessly

\n \n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -783,14 +783,14 @@\n Cylinder example :\n [images/smoothtriangles.png]\n This rule was empirically determined to make triangles subdivide more smoothly.\n However, this rule breaks the nice property that two separate meshes can be\n joined seamlessly by overlapping their boundaries; i.e. when there are\n triangles at either boundary, it is impossible to join the meshes seamlessly\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/tipuesearch/tipuesearch_content.js", "source2": "./usr/share/doc/opensubdiv/tipuesearch/tipuesearch_content.js", "unified_diff": null, "details": [{"source1": "js-beautify {}", "source2": "js-beautify {}", "unified_diff": "@@ -1,343 +1,343 @@\n var tipuesearch = {\n \"pages\": [{\n- \"title\": \"bfr_tutorial_3_2.cpp\",\n- \"text\": \"bfr_tutorial_3_2.cpp bfr_tutorial_3_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_3_2bfr_tutorial_3_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial is a variation of tutorials showing simple uniform tessellation. Rather than constructing and evaluating a Surface at a time, this tutorial shows how Surfaces can be created and saved for repeated use. A simple SurfaceCache class is created that creates and stores the Surface for each face, along with the patch points associated with it. The main tessellation function remains essentially the same, but here it access the Surfaces from the SurfaceCache rather than computing them locally. Note that while this example illustrated the retention of all Surfaces for a mesh, this behavior is not recommended. It does not scale well for large meshes and undermines the memory savings that transient use of Surfaces is designed to achieve. Rather than storing Surfaces for all faces, maintaining a priority queue for a fixed number may be a reasonable compromise. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include memory include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args This simple class creates and dispenses Surfaces for all faces of a mesh. It consists primarily of an array of simple structs entries for each face and a single array of patch points for all Surfaces created. There are many ways to create such a cache depending on requirements. This is a simple example, but the interface presents some options that are worth considering. A SurfaceCache is constructed here given the following - a reference to the SurfaceFactory - the cache could just as easily take a reference to the mesh and construct the SurfaceFactory internally - the position data for the mesh - this is needed to compute patch points for the Surfaces - if caching UVs or any other primvar, other data needs to be provided -- along with the interpolation type for that data vertex, face-varying, etc. - option to cache patch points - the cache could store the Surfaces only or also include their patch points - storing patch points takes more memory but will eliminate any preparation time for evaluation of the Surface - option to cache all surfaces - the benefits to caching simple linear or regular surfaces are minimal -- and may even be detrimental - so only caching non-linear irregular surfaces is an option worth considering The SurfaceCache implementation here provides the options noted above. But for simplicity, the actual usage of the SurfaceCache does not deal with the permutations of additional work that is necessary when the Surfaces or their patch points are not cached. class SurfaceCache public typedef Bfr Surface float Surface typedef Bfr RefinerSurfaceFactory SurfaceFactory public SurfaceCache SurfaceFactory const surfaceFactory , std vector float const meshPoints , bool cachePatchPoints true , bool cacheAllSurfaces true SurfaceCache delete SurfaceCache default Public methods to retrieved cached Surfaces and their pre-computed patch points bool FaceHasLimitSurface int face return _entries face . hasLimit Surface const GetSurface int face return _entries face . surface . get float const GetPatchPoints int face return getPatchPoints face private Simple struct to keep track of Surface and more for each face struct FaceEntry FaceEntry surface , hasLimit false , pointOffset -1 std unique_ptr Surface const surface bool hasLimit int pointOffset Non-const version to be used internally to aide assignment float getPatchPoints int face return _entries face . surface _points . empty _points . data _entries face . pointOffset 3 0 private std vector FaceEntry _entries std vector float _points SurfaceCache SurfaceCache SurfaceFactory const surfaceFactory , std vector float const meshPoints , bool cachePatchPoints , bool cacheAllSurfaces int numFaces surfaceFactory . GetNumFaces _entries . resize numFaces int numPointsInCache 0 for int face 0 face numFaces face Surface s surfaceFactory . CreateVertexSurface float face if s FaceEntry entry _entries face entry . hasLimit true if cacheAllSurfaces s - IsRegular s - IsLinear entry . surface . reset s entry . pointOffset numPointsInCache numPointsInCache s - GetNumPatchPoints else delete s if cachePatchPoints _points . resize numPointsInCache 3 for int face 0 face numFaces face float patchPoints getPatchPoints face if patchPoints GetSurface face - PreparePatchPoints meshPoints . data , 3 , patchPoints , 3 The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions Initialize a SurfaceCache to construct Surfaces for all faces. From this point forward the SurfaceFactory is no longer used to access Surfaces. Note also that usage below is specific to the options used to initialize the SurfaceCache bool cachePatchPoints true bool cacheAllSurfaces true SurfaceCache surfaceCache meshSurfaceFactory , meshVertexPositions , cachePatchPoints , cacheAllSurfaces As with previous tutorials, output data associated with the face can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. std vector float outCoords std vector float outPos , outDu , outDv std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Retrieve the Surface for this face when present if surfaceCache . FaceHasLimitSurface faceIndex continue Surface const faceSurface surfaceCache . GetSurface faceIndex Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern faceSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Retrieve the patch points for the Surface, then use them to evaluate output points for all identified coordinates float const facePatchPoints surfaceCache . GetPatchPoints faceIndex int pointSize 3 outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize for int i 0 , j 0 i numOutCoords i , j pointSize faceSurface . Evaluate outCoords i 2 , facePatchPoints , pointSize , outPos j , outDu j , outDv j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outFacets , tessFacetSize , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"osd_tutorial_0.cpp\",\n+ \"text\": \"osd_tutorial_0.cpp osd_tutorial_0.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsosdtutorial_0osd_tutorial_0.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial demonstrates the manipulation of Osd Evaluator and BufferDescriptor. include opensubdivfartopologyDescriptor.h include opensubdivfarstencilTableFactory.h include opensubdivosdcpuEvaluator.h include opensubdivosdcpuVertexBuffer.h include cstdio include cstring ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 24 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv static Far TopologyRefiner const createTopologyRefiner int maxlevel ------------------------------------------------------------------------------ int main int , char int maxlevel 2 , nCoarseVerts 0 , nRefinedVerts 0 Setup phase Far StencilTable const stencilTable NULL Setup FarStencilTable Far TopologyRefiner const refiner createTopologyRefiner maxlevel Setup a factory to create FarStencilTable for more details see Far tutorials Far StencilTableFactory Options options options . generateOffsets true options . generateIntermediateLevels false stencilTable Far StencilTableFactory Create refiner , options nCoarseVerts refiner - GetLevel 0 . GetNumVertices nRefinedVerts stencilTable - GetNumStencils We are done with Far cleanup table delete refiner Setup a buffer for vertex primvar data Osd CpuVertexBuffer vbuffer Osd CpuVertexBuffer Create 3 , nCoarseVerts nRefinedVerts Execution phase every frame Pack the control vertex data at the start of the vertex buffer and update every time control data changes vbuffer - UpdateData g_verts , 0 , nCoarseVerts Osd BufferDescriptor srcDesc 0 , 3 , 3 Osd BufferDescriptor dstDesc nCoarseVerts 3 , 3 , 3 Launch the computation Osd CpuEvaluator EvalStencils vbuffer , srcDesc , vbuffer , dstDesc , stencilTable Visualization with Maya print a MEL script that generates particles at the location of the refined vertices printf particle float const refinedVerts vbuffer - BindCpuBuffer 3 nCoarseVerts for int i 0 i nRefinedVerts i float const vert refinedVerts 3 i printf -p f f f n , vert 0 , vert 1 , vert 2 printf -c 1 n delete stencilTable delete vbuffer ------------------------------------------------------------------------------ static Far TopologyRefiner const createTopologyRefiner int maxlevel Populate a topology descriptor with our raw data typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options Uniformly refine the topology up to maxlevel refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel return refiner ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"bfr_tutorial_3_2.html\"\n+ \"loc\": \"osd_tutorial_0.html\"\n }, {\n- \"title\": \"bfr_tutorial_3_1.cpp\",\n- \"text\": \"bfr_tutorial_3_1.cpp bfr_tutorial_3_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_3_1bfr_tutorial_3_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial illustrates the definition of a custom subclass of BfrSurfaceFactory -- providing a class with the SurfaceFactory interface adapted to a connected mesh representation. The bulk of this code is therefore identical to a previous tutorial 1.3 which illustrates simple use of a BfrSurface factory. The only difference here lies in the explicit local definition of the subclass of BfrSurfaceFactory for FarTopologyRefiner -- named CustomSurfaceFactory in this case. include .customSurfaceFactory.h include opensubdivfartopologyRefiner.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag bool uv2xyzFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false , uv2xyzFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else if strcmp argv i , -uv2xyz uv2xyzFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , std vector float const meshFaceVaryingUVs , Args const options Use simpler local type names for the Surface and its factory typedef CustomSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing When dealing with face-varying data, an identifier is necessary when constructing Surfaces in order to distinguish the different face-varying data channels. To avoid repeatedly specifying that identifier when only one is present or of interest, it can be specified via the Options. bool meshHasUVs meshTopology . GetNumFVarChannels 0 SurfaceFactory Options surfaceOptions if meshHasUVs surfaceOptions . SetDefaultFVarID 0 SurfaceFactory surfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface posSurface Surface uvSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector float outUV std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces surfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surfaces for position and UVs of this face. There are two ways to do this -- both illustrated here Creating Surfaces for the different data interpolation types independently is clear and convenient, but considerable work may be duplicated in the construction process in the case of non-linear face-varying Surfaces. So unless it is known that face-varying interpolation is linear, use of InitSurfaces is generally preferred. Remember also that the face-varying identifier is omitted from the initialization methods here as it was previously assigned to the SurfaceFactoryOptions. In the absence of an assignment of the default FVarID to the Options, a failure to specify the FVarID here will result in failure. The cases below are expanded for illustration purposes, and validity of the resulting Surface is tested here, rather than the return value of initialization methods. bool createSurfacesTogether true if meshHasUVs surfaceFactory . InitVertexSurface faceIndex , posSurface else if createSurfacesTogether surfaceFactory . InitSurfaces faceIndex , posSurface , uvSurface else if surfaceFactory . InitVertexSurface faceIndex , posSurface surfaceFactory . InitFaceVaryingSurface faceIndex , uvSurface if posSurface . IsValid continue Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern posSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Evaluate vertex positions Resize patch point and output arrays int pointSize 3 facePatchPoints . resize posSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate patch point and output arrays posSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize posSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outPos j , outDu j , outDv j Evaluate face-varying UVs when present if meshHasUVs Resize patch point and output arrays - note reuse of the same patch point array as position int pointSize 2 facePatchPoints . resize uvSurface . GetNumPatchPoints pointSize outUV . resize numOutCoords pointSize Populate patch point and output arrays uvSurface . PreparePatchPoints meshFaceVaryingUVs . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize uvSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outUV j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex if meshHasUVs options . uv2xyzFlag objWriter . WriteVertexPositions outUV , 2 objWriter . WriteFaces outFacets , tessFacetSize , false , false else objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv if meshHasUVs objWriter . WriteVertexUVs outUV objWriter . WriteFaces outFacets , tessFacetSize , true , meshHasUVs Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , meshFVarUVs , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"bfr_tutorial_3_2.cpp\",\n+ \"text\": \"bfr_tutorial_3_2.cpp bfr_tutorial_3_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_3_2bfr_tutorial_3_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial is a variation of tutorials showing simple uniform tessellation. Rather than constructing and evaluating a Surface at a time, this tutorial shows how Surfaces can be created and saved for repeated use. A simple SurfaceCache class is created that creates and stores the Surface for each face, along with the patch points associated with it. The main tessellation function remains essentially the same, but here it access the Surfaces from the SurfaceCache rather than computing them locally. Note that while this example illustrated the retention of all Surfaces for a mesh, this behavior is not recommended. It does not scale well for large meshes and undermines the memory savings that transient use of Surfaces is designed to achieve. Rather than storing Surfaces for all faces, maintaining a priority queue for a fixed number may be a reasonable compromise. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include memory include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args This simple class creates and dispenses Surfaces for all faces of a mesh. It consists primarily of an array of simple structs entries for each face and a single array of patch points for all Surfaces created. There are many ways to create such a cache depending on requirements. This is a simple example, but the interface presents some options that are worth considering. A SurfaceCache is constructed here given the following - a reference to the SurfaceFactory - the cache could just as easily take a reference to the mesh and construct the SurfaceFactory internally - the position data for the mesh - this is needed to compute patch points for the Surfaces - if caching UVs or any other primvar, other data needs to be provided -- along with the interpolation type for that data vertex, face-varying, etc. - option to cache patch points - the cache could store the Surfaces only or also include their patch points - storing patch points takes more memory but will eliminate any preparation time for evaluation of the Surface - option to cache all surfaces - the benefits to caching simple linear or regular surfaces are minimal -- and may even be detrimental - so only caching non-linear irregular surfaces is an option worth considering The SurfaceCache implementation here provides the options noted above. But for simplicity, the actual usage of the SurfaceCache does not deal with the permutations of additional work that is necessary when the Surfaces or their patch points are not cached. class SurfaceCache public typedef Bfr Surface float Surface typedef Bfr RefinerSurfaceFactory SurfaceFactory public SurfaceCache SurfaceFactory const surfaceFactory , std vector float const meshPoints , bool cachePatchPoints true , bool cacheAllSurfaces true SurfaceCache delete SurfaceCache default Public methods to retrieved cached Surfaces and their pre-computed patch points bool FaceHasLimitSurface int face return _entries face . hasLimit Surface const GetSurface int face return _entries face . surface . get float const GetPatchPoints int face return getPatchPoints face private Simple struct to keep track of Surface and more for each face struct FaceEntry FaceEntry surface , hasLimit false , pointOffset -1 std unique_ptr Surface const surface bool hasLimit int pointOffset Non-const version to be used internally to aide assignment float getPatchPoints int face return _entries face . surface _points . empty _points . data _entries face . pointOffset 3 0 private std vector FaceEntry _entries std vector float _points SurfaceCache SurfaceCache SurfaceFactory const surfaceFactory , std vector float const meshPoints , bool cachePatchPoints , bool cacheAllSurfaces int numFaces surfaceFactory . GetNumFaces _entries . resize numFaces int numPointsInCache 0 for int face 0 face numFaces face Surface s surfaceFactory . CreateVertexSurface float face if s FaceEntry entry _entries face entry . hasLimit true if cacheAllSurfaces s - IsRegular s - IsLinear entry . surface . reset s entry . pointOffset numPointsInCache numPointsInCache s - GetNumPatchPoints else delete s if cachePatchPoints _points . resize numPointsInCache 3 for int face 0 face numFaces face float patchPoints getPatchPoints face if patchPoints GetSurface face - PreparePatchPoints meshPoints . data , 3 , patchPoints , 3 The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions Initialize a SurfaceCache to construct Surfaces for all faces. From this point forward the SurfaceFactory is no longer used to access Surfaces. Note also that usage below is specific to the options used to initialize the SurfaceCache bool cachePatchPoints true bool cacheAllSurfaces true SurfaceCache surfaceCache meshSurfaceFactory , meshVertexPositions , cachePatchPoints , cacheAllSurfaces As with previous tutorials, output data associated with the face can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. std vector float outCoords std vector float outPos , outDu , outDv std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Retrieve the Surface for this face when present if surfaceCache . FaceHasLimitSurface faceIndex continue Surface const faceSurface surfaceCache . GetSurface faceIndex Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern faceSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Retrieve the patch points for the Surface, then use them to evaluate output points for all identified coordinates float const facePatchPoints surfaceCache . GetPatchPoints faceIndex int pointSize 3 outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize for int i 0 , j 0 i numOutCoords i , j pointSize faceSurface . Evaluate outCoords i 2 , facePatchPoints , pointSize , outPos j , outDu j , outDv j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outFacets , tessFacetSize , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"bfr_tutorial_3_1.html\"\n+ \"loc\": \"bfr_tutorial_3_2.html\"\n }, {\n \"title\": \"bfr_tutorial_2_2.cpp\",\n- \"text\": \"bfr_tutorial_2_2.cpp bfr_tutorial_2_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_2_2bfr_tutorial_2_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial builds on others using the SurfaceFactory, Surface and Tessellation classes by using more of the functionality of the Tessellation class to construct a tessellation of the mesh that is topologically watertight, i.e. resulting points evaluated along shared edges or vertices are shared and not duplicated. Since Tessellation provides points around its boundary first, the evaluated points for shared vertices and edges are identified when constructed and reused when shared later. The boundary of the tessellation of a face is therefore a collection of shared points and methods of Tessellation help to remap the faces generated to the shared set of points. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio include cassert Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv using Far Index using Far IndexArray using Far ConstIndexArray Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args Simple local structs supporting shared points for vertices and edges namespace struct SharedVertex SharedVertex pointIndex -1 bool IsSet const return pointIndex 0 void Set int index pointIndex index int pointIndex struct SharedEdge SharedEdge pointIndex -1 , numPoints 0 bool IsSet const return pointIndex 0 void Set int index , int n pointIndex index , numPoints n int pointIndex int numPoints end namespace The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. This tessellation function differs from earlier tutorials in that it computes and reuses shared points at vertices and edges of the mesh. There are several ways to compute these shared points, and which is best depends on context. Dealing with shared data poses complications for threading in general, so computing all points for the vertices and edges up front may be preferred -- despite the fact that faces will be visited more than once first when generating potentially shared vertex or edge points, and later when generating any interior points. The loops for vertices and edges can be threaded and the indexing of the shared points is simpler. For the single-threaded case here, the faces are each processed in order and any shared points will be computed and used as needed. So each face is visited once and so each Surface initialized once but the bookkeeping to deal with indices of shared points becomes more complicated. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface faceSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Declare vectors to identify shared tessellation points at vertices and edges and their indices around the boundary of a face Far TopologyLevel const baseLevel meshTopology . GetLevel 0 std vector SharedVertex sharedVerts baseLevel . GetNumVertices std vector SharedEdge sharedEdges baseLevel . GetNumEdges std vector int tessBoundaryIndices Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numMeshPointsEvaluated 0 int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surface for this face -- if valid skipping holes and boundary faces in some rare cases if meshSurfaceFactory . InitVertexSurface faceIndex , faceSurface continue Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern faceSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Resize patch point and output arrays int pointSize 3 facePatchPoints . resize faceSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate the patch point array faceSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize Evaluate the sample points of the Tessellation First traverse the boundary of the face to determine whether to evaluate or share points on vertices and edges of the face. Both pre-existing and new boundary points are identified by index in an array for later use. The interior points are all trivially computed after the boundary is dealt with. Identify the boundary and interior coords and initialize the index array for the potentially shared boundary points int numBoundaryCoords tessPattern . GetNumBoundaryCoords int numInteriorCoords numOutCoords - numBoundaryCoords float const tessBoundaryCoords outCoords 0 float const tessInteriorCoords outCoords numBoundaryCoords 2 ConstIndexArray fVerts baseLevel . GetFaceVertices faceIndex ConstIndexArray fEdges baseLevel . GetFaceEdges faceIndex tessBoundaryIndices . resize numBoundaryCoords Walk around the face, inspecting each vertex and outgoing edge, and populating the index array of boundary points float patchPointData facePatchPoints . data int boundaryIndex 0 int numFacePointsEvaluated 0 for int i 0 i fVerts . size i Index vertIndex fVerts i Index edgeIndex fEdges i int edgeRate options . tessUniformRate Evaluateassign or retrieve the shared point for the vertex SharedVertex sharedVertex sharedVerts vertIndex if sharedVertex . IsSet Identify indices of the new shared point in both the mesh and face and increment their inventory int indexInMesh numMeshPointsEvaluated int indexInFace numFacePointsEvaluated sharedVertex . Set indexInMesh Evaluate new shared point and assign index to boundary float const uv tessBoundaryCoords boundaryIndex 2 int pIndex indexInFace pointSize faceSurface . Evaluate uv , patchPointData , pointSize , outPos pIndex , outDu pIndex , outDv pIndex tessBoundaryIndices boundaryIndex indexInMesh else Assign shared vertex point index to boundary tessBoundaryIndices boundaryIndex sharedVertex . pointIndex Evaluateassign or retrieve all shared points for the edge To keep this simple, assume the edge is manifold. So the second face sharing the edge has that edge in the opposite direction in its boundary relative to the first face -- making it necessary to reverse the order of shared points for the boundary of the second face. To support a non-manifold edge, all subsequent faces that share the assigned shared edge must determine if their orientation of that edge is reversed relative to the first face for which the shared edge points were evaluated. So a little more book-keeping andor inspection is required. if edgeRate 1 int pointsPerEdge edgeRate - 1 SharedEdge sharedEdge sharedEdges edgeIndex if sharedEdge . IsSet Identify indices of the new shared points in both the mesh and face and increment their inventory int nextInMesh numMeshPointsEvaluated int nextInFace numFacePointsEvaluated numFacePointsEvaluated pointsPerEdge numMeshPointsEvaluated pointsPerEdge sharedEdge . Set nextInMesh , pointsPerEdge Evaluate shared points and assign indices to boundary float const uv tessBoundaryCoords boundaryIndex 2 for int j 0 j pointsPerEdge j , uv 2 int pIndex nextInFace pointSize faceSurface . Evaluate uv , patchPointData , pointSize , outPos pIndex , outDu pIndex , outDv pIndex tessBoundaryIndices boundaryIndex nextInMesh else See note above on simplification for manifold edges assert baseLevel . IsEdgeNonManifold edgeIndex Assign shared points to boundary in reverse order int nextInMesh sharedEdge . pointIndex pointsPerEdge - 1 for int j 0 j pointsPerEdge j tessBoundaryIndices boundaryIndex nextInMesh -- Evaluate any interior points unique to this face -- appending them to those shared points computed above for the boundary if numInteriorCoords float const uv tessInteriorCoords int iLast numFacePointsEvaluated numInteriorCoords for int i numFacePointsEvaluated i iLast i , uv 2 int pIndex i pointSize faceSurface . Evaluate uv , patchPointData , pointSize , outPos pIndex , outDu pIndex , outDv pIndex numFacePointsEvaluated numInteriorCoords numMeshPointsEvaluated numInteriorCoords Remember to trimresize the arrays storing evaluation results for new points to reflect the size actually populated. outPos . resize numFacePointsEvaluated pointSize outDu . resize numFacePointsEvaluated pointSize outDv . resize numFacePointsEvaluated pointSize Identify the faces of the Tessellation Note that the coordinate indices used by the facets are local to the face i.e. they range from 0..N-1, where N is the number of coordinates in the pattern and so need to be offset when writing to Obj format. For more advanced use, the coordinates associated with the boundary and interior of the pattern are distinguishable so that those on the boundary can be easily remapped to refer to shared edge or corner points, while those in the interior can be separately offset or similarly remapped. So transform the indices of the facets here as needed using the indices of shared boundary points assembled above and a suitable offset for the new interior points added int tessInteriorOffset numMeshPointsEvaluated - numOutCoords int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , tessBoundaryIndices . data , tessInteriorOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outFacets , tessFacetSize , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"bfr_tutorial_2_2.cpp bfr_tutorial_2_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_2_2bfr_tutorial_2_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial builds on others using the SurfaceFactory, Surface and Tessellation classes by using more of the functionality of the Tessellation class to construct a tessellation of the mesh that is topologically watertight, i.e. resulting points evaluated along shared edges or vertices are shared and not duplicated. Since Tessellation provides points around its boundary first, the evaluated points for shared vertices and edges are identified when constructed and reused when shared later. The boundary of the tessellation of a face is therefore a collection of shared points and methods of Tessellation help to remap the faces generated to the shared set of points. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio include cassert Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv using Far Index using Far IndexArray using Far ConstIndexArray Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args Simple local structs supporting shared points for vertices and edges namespace struct SharedVertex SharedVertex pointIndex -1 bool IsSet const return pointIndex 0 void Set int index pointIndex index int pointIndex struct SharedEdge SharedEdge pointIndex -1 , numPoints 0 bool IsSet const return pointIndex 0 void Set int index , int n pointIndex index , numPoints n int pointIndex int numPoints end namespace The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. This tessellation function differs from earlier tutorials in that it computes and reuses shared points at vertices and edges of the mesh. There are several ways to compute these shared points, and which is best depends on context. Dealing with shared data poses complications for threading in general, so computing all points for the vertices and edges up front may be preferred -- despite the fact that faces will be visited more than once first when generating potentially shared vertex or edge points, and later when generating any interior points. The loops for vertices and edges can be threaded and the indexing of the shared points is simpler. For the single-threaded case here, the faces are each processed in order and any shared points will be computed and used as needed. So each face is visited once and so each Surface initialized once but the bookkeeping to deal with indices of shared points becomes more complicated. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface faceSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Declare vectors to identify shared tessellation points at vertices and edges and their indices around the boundary of a face Far TopologyLevel const baseLevel meshTopology . GetLevel 0 std vector SharedVertex sharedVerts baseLevel . GetNumVertices std vector SharedEdge sharedEdges baseLevel . GetNumEdges std vector int tessBoundaryIndices Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numMeshPointsEvaluated 0 int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surface for this face -- if valid skipping holes and boundary faces in some rare cases if meshSurfaceFactory . InitVertexSurface faceIndex , faceSurface continue Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern faceSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Resize patch point and output arrays int pointSize 3 facePatchPoints . resize faceSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate the patch point array faceSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize Evaluate the sample points of the Tessellation First traverse the boundary of the face to determine whether to evaluate or share points on vertices and edges of the face. Both pre-existing and new boundary points are identified by index in an array for later use. The interior points are all trivially computed after the boundary is dealt with. Identify the boundary and interior coords and initialize the index array for the potentially shared boundary points int numBoundaryCoords tessPattern . GetNumBoundaryCoords int numInteriorCoords numOutCoords - numBoundaryCoords float const tessBoundaryCoords outCoords 0 float const tessInteriorCoords outCoords numBoundaryCoords 2 ConstIndexArray fVerts baseLevel . GetFaceVertices faceIndex ConstIndexArray fEdges baseLevel . GetFaceEdges faceIndex tessBoundaryIndices . resize numBoundaryCoords Walk around the face, inspecting each vertex and outgoing edge, and populating the index array of boundary points float patchPointData facePatchPoints . data int boundaryIndex 0 int numFacePointsEvaluated 0 for int i 0 i fVerts . size i Index vertIndex fVerts i Index edgeIndex fEdges i int edgeRate options . tessUniformRate Evaluateassign or retrieve the shared point for the vertex SharedVertex sharedVertex sharedVerts vertIndex if sharedVertex . IsSet Identify indices of the new shared point in both the mesh and face and increment their inventory int indexInMesh numMeshPointsEvaluated int indexInFace numFacePointsEvaluated sharedVertex . Set indexInMesh Evaluate new shared point and assign index to boundary float const uv tessBoundaryCoords boundaryIndex 2 int pIndex indexInFace pointSize faceSurface . Evaluate uv , patchPointData , pointSize , outPos pIndex , outDu pIndex , outDv pIndex tessBoundaryIndices boundaryIndex indexInMesh else Assign shared vertex point index to boundary tessBoundaryIndices boundaryIndex sharedVertex . pointIndex Evaluateassign or retrieve all shared points for the edge To keep this simple, assume the edge is manifold. So the second face sharing the edge has that edge in the opposite direction in its boundary relative to the first face -- making it necessary to reverse the order of shared points for the boundary of the second face. To support a non-manifold edge, all subsequent faces that share the assigned shared edge must determine if their orientation of that edge is reversed relative to the first face for which the shared edge points were evaluated. So a little more book-keeping andor inspection is required. if edgeRate 1 int pointsPerEdge edgeRate - 1 SharedEdge sharedEdge sharedEdges edgeIndex if sharedEdge . IsSet Identify indices of the new shared points in both the mesh and face and increment their inventory int nextInMesh numMeshPointsEvaluated int nextInFace numFacePointsEvaluated numFacePointsEvaluated pointsPerEdge numMeshPointsEvaluated pointsPerEdge sharedEdge . Set nextInMesh , pointsPerEdge Evaluate shared points and assign indices to boundary float const uv tessBoundaryCoords boundaryIndex 2 for int j 0 j pointsPerEdge j , uv 2 int pIndex nextInFace pointSize faceSurface . Evaluate uv , patchPointData , pointSize , outPos pIndex , outDu pIndex , outDv pIndex tessBoundaryIndices boundaryIndex nextInMesh else See note above on simplification for manifold edges assert baseLevel . IsEdgeNonManifold edgeIndex Assign shared points to boundary in reverse order int nextInMesh sharedEdge . pointIndex pointsPerEdge - 1 for int j 0 j pointsPerEdge j tessBoundaryIndices boundaryIndex nextInMesh -- Evaluate any interior points unique to this face -- appending them to those shared points computed above for the boundary if numInteriorCoords float const uv tessInteriorCoords int iLast numFacePointsEvaluated numInteriorCoords for int i numFacePointsEvaluated i iLast i , uv 2 int pIndex i pointSize faceSurface . Evaluate uv , patchPointData , pointSize , outPos pIndex , outDu pIndex , outDv pIndex numFacePointsEvaluated numInteriorCoords numMeshPointsEvaluated numInteriorCoords Remember to trimresize the arrays storing evaluation results for new points to reflect the size actually populated. outPos . resize numFacePointsEvaluated pointSize outDu . resize numFacePointsEvaluated pointSize outDv . resize numFacePointsEvaluated pointSize Identify the faces of the Tessellation Note that the coordinate indices used by the facets are local to the face i.e. they range from 0..N-1, where N is the number of coordinates in the pattern and so need to be offset when writing to Obj format. For more advanced use, the coordinates associated with the boundary and interior of the pattern are distinguishable so that those on the boundary can be easily remapped to refer to shared edge or corner points, while those in the interior can be separately offset or similarly remapped. So transform the indices of the facets here as needed using the indices of shared boundary points assembled above and a suitable offset for the new interior points added int tessInteriorOffset numMeshPointsEvaluated - numOutCoords int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , tessBoundaryIndices . data , tessInteriorOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outFacets , tessFacetSize , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"bfr_tutorial_2_2.html\"\n }, {\n- \"title\": \"osd_tutorial_0.cpp\",\n- \"text\": \"osd_tutorial_0.cpp osd_tutorial_0.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsosdtutorial_0osd_tutorial_0.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial demonstrates the manipulation of Osd Evaluator and BufferDescriptor. include opensubdivfartopologyDescriptor.h include opensubdivfarstencilTableFactory.h include opensubdivosdcpuEvaluator.h include opensubdivosdcpuVertexBuffer.h include cstdio include cstring ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 24 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv static Far TopologyRefiner const createTopologyRefiner int maxlevel ------------------------------------------------------------------------------ int main int , char int maxlevel 2 , nCoarseVerts 0 , nRefinedVerts 0 Setup phase Far StencilTable const stencilTable NULL Setup FarStencilTable Far TopologyRefiner const refiner createTopologyRefiner maxlevel Setup a factory to create FarStencilTable for more details see Far tutorials Far StencilTableFactory Options options options . generateOffsets true options . generateIntermediateLevels false stencilTable Far StencilTableFactory Create refiner , options nCoarseVerts refiner - GetLevel 0 . GetNumVertices nRefinedVerts stencilTable - GetNumStencils We are done with Far cleanup table delete refiner Setup a buffer for vertex primvar data Osd CpuVertexBuffer vbuffer Osd CpuVertexBuffer Create 3 , nCoarseVerts nRefinedVerts Execution phase every frame Pack the control vertex data at the start of the vertex buffer and update every time control data changes vbuffer - UpdateData g_verts , 0 , nCoarseVerts Osd BufferDescriptor srcDesc 0 , 3 , 3 Osd BufferDescriptor dstDesc nCoarseVerts 3 , 3 , 3 Launch the computation Osd CpuEvaluator EvalStencils vbuffer , srcDesc , vbuffer , dstDesc , stencilTable Visualization with Maya print a MEL script that generates particles at the location of the refined vertices printf particle float const refinedVerts vbuffer - BindCpuBuffer 3 nCoarseVerts for int i 0 i nRefinedVerts i float const vert refinedVerts 3 i printf -p f f f n , vert 0 , vert 1 , vert 2 printf -c 1 n delete stencilTable delete vbuffer ------------------------------------------------------------------------------ static Far TopologyRefiner const createTopologyRefiner int maxlevel Populate a topology descriptor with our raw data typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options Uniformly refine the topology up to maxlevel refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel return refiner ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"bfr_tutorial_3_1.cpp\",\n+ \"text\": \"bfr_tutorial_3_1.cpp bfr_tutorial_3_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_3_1bfr_tutorial_3_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial illustrates the definition of a custom subclass of BfrSurfaceFactory -- providing a class with the SurfaceFactory interface adapted to a connected mesh representation. The bulk of this code is therefore identical to a previous tutorial 1.3 which illustrates simple use of a BfrSurface factory. The only difference here lies in the explicit local definition of the subclass of BfrSurfaceFactory for FarTopologyRefiner -- named CustomSurfaceFactory in this case. include .customSurfaceFactory.h include opensubdivfartopologyRefiner.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag bool uv2xyzFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false , uv2xyzFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else if strcmp argv i , -uv2xyz uv2xyzFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , std vector float const meshFaceVaryingUVs , Args const options Use simpler local type names for the Surface and its factory typedef CustomSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing When dealing with face-varying data, an identifier is necessary when constructing Surfaces in order to distinguish the different face-varying data channels. To avoid repeatedly specifying that identifier when only one is present or of interest, it can be specified via the Options. bool meshHasUVs meshTopology . GetNumFVarChannels 0 SurfaceFactory Options surfaceOptions if meshHasUVs surfaceOptions . SetDefaultFVarID 0 SurfaceFactory surfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface posSurface Surface uvSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector float outUV std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces surfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surfaces for position and UVs of this face. There are two ways to do this -- both illustrated here Creating Surfaces for the different data interpolation types independently is clear and convenient, but considerable work may be duplicated in the construction process in the case of non-linear face-varying Surfaces. So unless it is known that face-varying interpolation is linear, use of InitSurfaces is generally preferred. Remember also that the face-varying identifier is omitted from the initialization methods here as it was previously assigned to the SurfaceFactoryOptions. In the absence of an assignment of the default FVarID to the Options, a failure to specify the FVarID here will result in failure. The cases below are expanded for illustration purposes, and validity of the resulting Surface is tested here, rather than the return value of initialization methods. bool createSurfacesTogether true if meshHasUVs surfaceFactory . InitVertexSurface faceIndex , posSurface else if createSurfacesTogether surfaceFactory . InitSurfaces faceIndex , posSurface , uvSurface else if surfaceFactory . InitVertexSurface faceIndex , posSurface surfaceFactory . InitFaceVaryingSurface faceIndex , uvSurface if posSurface . IsValid continue Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern posSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Evaluate vertex positions Resize patch point and output arrays int pointSize 3 facePatchPoints . resize posSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate patch point and output arrays posSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize posSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outPos j , outDu j , outDv j Evaluate face-varying UVs when present if meshHasUVs Resize patch point and output arrays - note reuse of the same patch point array as position int pointSize 2 facePatchPoints . resize uvSurface . GetNumPatchPoints pointSize outUV . resize numOutCoords pointSize Populate patch point and output arrays uvSurface . PreparePatchPoints meshFaceVaryingUVs . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize uvSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outUV j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex if meshHasUVs options . uv2xyzFlag objWriter . WriteVertexPositions outUV , 2 objWriter . WriteFaces outFacets , tessFacetSize , false , false else objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv if meshHasUVs objWriter . WriteVertexUVs outUV objWriter . WriteFaces outFacets , tessFacetSize , true , meshHasUVs Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , meshFVarUVs , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"osd_tutorial_0.html\"\n+ \"loc\": \"bfr_tutorial_3_1.html\"\n+ }, {\n+ \"title\": \"bfr_tutorial_1_5.cpp\",\n+ \"text\": \"bfr_tutorial_1_5.cpp bfr_tutorial_1_5.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_1_5bfr_tutorial_1_5.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial is an alternative to an earlier tutorial that showed uniform tessellation. This version differs by evaluating the points of the tessellation using limit stencils instead of the standard Surface evaluation methods. Limit stencils factor the evaluation into a set of coefficients for each control point affecting the Surface. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface faceSurface std vector float faceControlPoints std vector float limitStencils std vector float outCoords std vector float outPos , outDu , outDv std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surface for this face -- if valid skipping holes and boundary faces in some rare cases if meshSurfaceFactory . InitVertexSurface faceIndex , faceSurface continue Resize stencils and control point arrays based on the number of control points for the Surface int numControlPoints faceSurface . GetNumControlPoints limitStencils . resize 3 numControlPoints float pStencil limitStencils . data float duStencil limitStencils . data numControlPoints float dvStencil limitStencils . data numControlPoints 2 Limit stencils can be applied using the control points in a local array or directy from the mesh. Both are shown here, so if using the local array, resize and populate it bool gatherControlPoints true if gatherControlPoints faceControlPoints . resize numControlPoints 3 faceSurface . GatherControlPoints meshVertexPositions . data , 3 , faceControlPoints . data , 3 Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern faceSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Evaluate and apply stencils to compute points of the tessellation outPos . resize numOutCoords 3 outDu . resize numOutCoords 3 outDv . resize numOutCoords 3 for int i 0 i numOutCoords i float const uv outCoords . data i 2 faceSurface . EvaluateStencil uv , pStencil , duStencil , dvStencil float p outPos . data i 3 float du outDu . data i 3 float dv outDv . data i 3 if gatherControlPoints float const controlPoints faceControlPoints . data faceSurface . ApplyStencil pStencil , controlPoints , 3 , p faceSurface . ApplyStencil duStencil , controlPoints , 3 , du faceSurface . ApplyStencil dvStencil , controlPoints , 3 , dv else float const meshPoints meshVertexPositions . data faceSurface . ApplyStencilFromMesh pStencil , meshPoints , 3 , p faceSurface . ApplyStencilFromMesh duStencil , meshPoints , 3 , du faceSurface . ApplyStencilFromMesh dvStencil , meshPoints , 3 , dv Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outFacets , tessFacetSize , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"tags\": \"\",\n+ \"loc\": \"bfr_tutorial_1_5.html\"\n }, {\n \"title\": \"bfr_tutorial_2_1.cpp\",\n- \"text\": \"bfr_tutorial_2_1.cpp bfr_tutorial_2_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_2_1bfr_tutorial_2_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial builds on the previous tutorial that makes use of the SurfaceFactory, Surface and Tessellation classes by illustrating the use of non-uniform tessellation parameters with Tessellation. Tessellation rates for the edges of a face are determined by a length associated with each edge. That length may be computed using either the control hull or the limit surface. The length of a tessellation interval is required and will be inferred if not explicitly specified as a command line option. The tessellation rate for an edge is computed as its length divided by the length of the tessellation interval. A maximum tessellation rate is imposed to prevent accidental unbounded tessellation, but can easily be raised as needed. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio include cmath include algorithm Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType float tessInterval int tessRateMax bool useHullFlag bool tessQuadsFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessInterval 0.0f , tessRateMax 10 , useHullFlag false , tessQuadsFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -length if i argc tessInterval float atof argv i else if strcmp argv i , -max if i argc tessRateMax atoi argv i else if strcmp argv i , -hull useHullFlag true else if strcmp argv i , -quads tessQuadsFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args Local trivial functions for simple edge length calculations and the determination of associated tessellation rates inline float EdgeLength float const v0 , float const v1 float dv 3 dv 0 std abs v0 0 - v1 0 dv 1 std abs v0 1 - v1 1 dv 2 std abs v0 2 - v1 2 return std sqrt dv 0 dv 0 dv 1 dv 1 dv 2 dv 2 float FindLongestEdge Far TopologyRefiner const mesh , std vector float const vertPos , int pointSize float maxLength 0.0f int numEdges mesh . GetLevel 0 . GetNumEdges for int i 0 i numEdges i Far ConstIndexArray edgeVerts mesh . GetLevel 0 . GetEdgeVertices i float edgeLength EdgeLength vertPos edgeVerts 0 pointSize , vertPos edgeVerts 1 pointSize maxLength std max maxLength , edgeLength return maxLength void GetEdgeTessRates std vector float const vertPos , int pointSize , Args const options , int edgeRates int numEdges int vertPos . size pointSize for int i 0 i numEdges i int j i 1 numEdges float edgeLength EdgeLength vertPos i pointSize , vertPos j pointSize edgeRates i 1 int edgeLength options . tessInterval edgeRates i std min edgeRates i , options . tessRateMax The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface faceSurface std vector float facePatchPoints std vector int faceTessRates std vector float outCoords std vector float outPos , outDu , outDv std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surface for this face -- if valid skipping holes and boundary faces in some rare cases if meshSurfaceFactory . InitVertexSurface faceIndex , faceSurface continue Prepare the Surface patch points first as it may be evaluated to determine suitable edge-rates for Tessellation int pointSize 3 facePatchPoints . resize faceSurface . GetNumPatchPoints pointSize faceSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize For each of the N edges of the face, a tessellation rate is determined to initialize a non-uniform Tessellation pattern. Many metrics are possible -- some based on the geometry itself size, curvature, others dependent on viewpoint screen space size, center of view, etc. and many more. Simple techniques are chosen here for illustration and can easily be replaced. Here two methods are shown using lengths between the corners of the face -- the first using the vertex positions of the face and the second using points evaluated at the corners of its limit surface. Use of the control hull is more efficient avoiding the evaluation but may prove less effective in some cases though both estimates have their limitations. int N faceSurface . GetFaceSize Use the output array temporarily to hold the N positions outPos . resize N pointSize if options . useHullFlag Far ConstIndexArray verts meshTopology . GetLevel 0 . GetFaceVertices faceIndex for int i 0 , j 0 i N i , j pointSize float const vPos meshVertexPositions verts i pointSize outPos j vPos 0 outPos j 1 vPos 1 outPos j 2 vPos 2 else Bfr Parameterization faceParam faceSurface . GetParameterization for int i 0 , j 0 i N i , j pointSize float uv 2 faceParam . GetVertexCoord i , uv faceSurface . Evaluate uv , facePatchPoints . data , pointSize , outPos j faceTessRates . resize N GetEdgeTessRates outPos , pointSize , options , faceTessRates . data Declare a non-uniform Tessellation using the rates for each edge and identify coordinates of the points to evaluate Additional interior rates can be optionally provided 2 for quads, 1 for others but will be inferred in their absence. Bfr Tessellation tessPattern faceSurface . GetParameterization , N , faceTessRates . data , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Resize the output arrays and evaluate outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize for int i 0 , j 0 i numOutCoords i , j pointSize faceSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outPos j , outDu j , outDv j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outFacets , tessFacetSize , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE If no interval length was specified, set one by finding the longest edge of the mesh and dividing it by the maximum tessellation rate if args . tessInterval 0.0f args . tessInterval FindLongestEdge meshTopology , meshVtxPositions , 3 float args . tessRateMax tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"bfr_tutorial_2_1.cpp bfr_tutorial_2_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_2_1bfr_tutorial_2_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial builds on the previous tutorial that makes use of the SurfaceFactory, Surface and Tessellation classes by illustrating the use of non-uniform tessellation parameters with Tessellation. Tessellation rates for the edges of a face are determined by a length associated with each edge. That length may be computed using either the control hull or the limit surface. The length of a tessellation interval is required and will be inferred if not explicitly specified as a command line option. The tessellation rate for an edge is computed as its length divided by the length of the tessellation interval. A maximum tessellation rate is imposed to prevent accidental unbounded tessellation, but can easily be raised as needed. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio include cmath include algorithm Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType float tessInterval int tessRateMax bool useHullFlag bool tessQuadsFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessInterval 0.0f , tessRateMax 10 , useHullFlag false , tessQuadsFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -length if i argc tessInterval float atof argv i else if strcmp argv i , -max if i argc tessRateMax atoi argv i else if strcmp argv i , -hull useHullFlag true else if strcmp argv i , -quads tessQuadsFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args Local trivial functions for simple edge length calculations and the determination of associated tessellation rates inline float EdgeLength float const v0 , float const v1 float dv 3 dv 0 std abs v0 0 - v1 0 dv 1 std abs v0 1 - v1 1 dv 2 std abs v0 2 - v1 2 return std sqrt dv 0 dv 0 dv 1 dv 1 dv 2 dv 2 float FindLongestEdge Far TopologyRefiner const mesh , std vector float const vertPos , int pointSize float maxLength 0.0f int numEdges mesh . GetLevel 0 . GetNumEdges for int i 0 i numEdges i Far ConstIndexArray edgeVerts mesh . GetLevel 0 . GetEdgeVertices i float edgeLength EdgeLength vertPos edgeVerts 0 pointSize , vertPos edgeVerts 1 pointSize maxLength std max maxLength , edgeLength return maxLength void GetEdgeTessRates std vector float const vertPos , int pointSize , Args const options , int edgeRates int numEdges int vertPos . size pointSize for int i 0 i numEdges i int j i 1 numEdges float edgeLength EdgeLength vertPos i pointSize , vertPos j pointSize edgeRates i 1 int edgeLength options . tessInterval edgeRates i std min edgeRates i , options . tessRateMax The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface faceSurface std vector float facePatchPoints std vector int faceTessRates std vector float outCoords std vector float outPos , outDu , outDv std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surface for this face -- if valid skipping holes and boundary faces in some rare cases if meshSurfaceFactory . InitVertexSurface faceIndex , faceSurface continue Prepare the Surface patch points first as it may be evaluated to determine suitable edge-rates for Tessellation int pointSize 3 facePatchPoints . resize faceSurface . GetNumPatchPoints pointSize faceSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize For each of the N edges of the face, a tessellation rate is determined to initialize a non-uniform Tessellation pattern. Many metrics are possible -- some based on the geometry itself size, curvature, others dependent on viewpoint screen space size, center of view, etc. and many more. Simple techniques are chosen here for illustration and can easily be replaced. Here two methods are shown using lengths between the corners of the face -- the first using the vertex positions of the face and the second using points evaluated at the corners of its limit surface. Use of the control hull is more efficient avoiding the evaluation but may prove less effective in some cases though both estimates have their limitations. int N faceSurface . GetFaceSize Use the output array temporarily to hold the N positions outPos . resize N pointSize if options . useHullFlag Far ConstIndexArray verts meshTopology . GetLevel 0 . GetFaceVertices faceIndex for int i 0 , j 0 i N i , j pointSize float const vPos meshVertexPositions verts i pointSize outPos j vPos 0 outPos j 1 vPos 1 outPos j 2 vPos 2 else Bfr Parameterization faceParam faceSurface . GetParameterization for int i 0 , j 0 i N i , j pointSize float uv 2 faceParam . GetVertexCoord i , uv faceSurface . Evaluate uv , facePatchPoints . data , pointSize , outPos j faceTessRates . resize N GetEdgeTessRates outPos , pointSize , options , faceTessRates . data Declare a non-uniform Tessellation using the rates for each edge and identify coordinates of the points to evaluate Additional interior rates can be optionally provided 2 for quads, 1 for others but will be inferred in their absence. Bfr Tessellation tessPattern faceSurface . GetParameterization , N , faceTessRates . data , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Resize the output arrays and evaluate outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize for int i 0 , j 0 i numOutCoords i , j pointSize faceSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outPos j , outDu j , outDv j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outFacets , tessFacetSize , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE If no interval length was specified, set one by finding the longest edge of the mesh and dividing it by the maximum tessellation rate if args . tessInterval 0.0f args . tessInterval FindLongestEdge meshTopology , meshVtxPositions , 3 float args . tessRateMax tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"bfr_tutorial_2_1.html\"\n }, {\n+ \"title\": \"far_tutorial_5_3.cpp\",\n+ \"text\": \"far_tutorial_5_3.cpp far_tutorial_5_3.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_5_3far_tutorial_5_3.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to use a FarLimitStenciTable to repeatedly and efficiently evaluate a set of points and optionally derivatives on the limit surface. A LimitStencilTable derives from StencilTable but is specialized to factor the evaluation of limit positions and derivatives into stencils. This allows a set of limit properties to be efficiently recomputed in response to changes to the vertices of the base mesh. Constructing the different kinds of StencilTables can have a high cost, so whether that cost is worth it will depend on your usage e.g. if points are only computed once, using stencil tables is typically not worth the added cost. Any points on the limit surface can be identified for evaluation. In this example we create a crude tessellation similar to tutorial_5_2. The midpoint of each face and points near the corners of the face are evaluated and a triangle fan connects them. include ......regressioncommonarg_utils.h include ......regressioncommonfar_utils.h include opensubdivfartopologyDescriptor.h include opensubdivfarpatchTableFactory.h include opensubdivfarstencilTableFactory.h include opensubdivfarptexIndices.h include cassert include cstdio include cstring include fstream include sstream using namespace OpenSubdiv using Far Index Global utilities in this namespace are not relevant to the tutorial. They simply serve to construct some default geometry to be processed in the form of a TopologyRefiner and vector of vertex positions. namespace Simple structs for x,y,z position and a 3-tuple for the set of vertices of a triangle struct Pos Pos Pos float x , float y , float z p 0 x , p 1 y , p 2 z Pos operator Pos const op const return Pos p 0 op . p 0 , p 1 op . p 1 , p 2 op . p 2 Clear and AddWithWeight required for interpolation void Clear void 0 p 0 p 1 p 2 0.0f void AddWithWeight Pos const src , float weight p 0 weight src . p 0 p 1 weight src . p 1 p 2 weight src . p 2 float p 3 typedef std vector Pos PosVector struct Tri Tri Tri int a , int b , int c v 0 a , v 1 b , v 2 c int v 3 typedef std vector Tri TriVector Functions to populate the topology and geometry arrays a simple shape whose positions may be transformed void createCube std vector int vertsPerFace , std vector Index faceVertsPerFace , std vector Pos positionsPerVert Local topology and position of a cube centered at origin static float const cubePositions 8 3 -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f static int const cubeFaceVerts 6 4 0 , 3 , 2 , 1 , 4 , 5 , 6 , 7 , 0 , 4 , 7 , 3 , 1 , 2 , 6 , 5 , 0 , 1 , 5 , 4 , 3 , 7 , 6 , 2 Initialize verts-per-face and face-vertices for each face vertsPerFace . resize 6 faceVertsPerFace . resize 24 for int i 0 i 6 i vertsPerFace i 4 for int j 0 j 4 j faceVertsPerFace i 4 j cubeFaceVerts i j Initialize vertex positions positionsPerVert . resize 8 for int i 0 i 8 i float const p cubePositions i positionsPerVert i Pos p 0 , p 1 , p 2 Create a TopologyRefiner from default geometry created above Far TopologyRefiner createTopologyRefinerDefault PosVector posVector std vector int topVertsPerFace std vector Index topFaceVerts createCube topVertsPerFace , topFaceVerts , posVector typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_AND_CORNER Descriptor desc desc . numVertices int posVector . size desc . numFaces int topVertsPerFace . size desc . numVertsPerFace topVertsPerFace 0 desc . vertIndicesPerFace topFaceVerts 0 Instantiate a FarTopologyRefiner from the descriptor. Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options assert refiner return refiner Create a TopologyRefiner from a specified Obj file geometry created internally Far TopologyRefiner createTopologyRefinerFromObj std string const objFileName , Sdc SchemeType schemeType , PosVector posVector const char filename objFileName . c_str const Shape shape 0 std ifstream ifs filename if ifs std stringstream ss ss ifs . rdbuf ifs . close std string shapeString ss . str shape Shape parseObj shapeString . c_str , ConvertSdcTypeToShapeScheme schemeType , false if shape 0 fprintf stderr , Error Cannot create Shape from .obj file s n , filename return 0 else fprintf stderr , Error Cannot open .obj file s n , filename return 0 Sdc SchemeType sdcType GetSdcType shape Sdc Options sdcOptions GetSdcOptions shape Far TopologyRefiner refiner Far TopologyRefinerFactory Shape Create shape , Far TopologyRefinerFactory Shape Options sdcType , sdcOptions if refiner 0 fprintf stderr , Error Unable to construct TopologyRefiner from .obj file s n , filename return 0 int numVertices refiner - GetNumVerticesTotal posVector . resize numVertices std memcpy posVector 0 . p 0 , shape - verts 0 , numVertices 3 sizeof float delete shape return refiner Simple function to export an Obj file for the limit points -- which provides a simple tessllation similar to tutorial_5_2. int writeToObj Far TopologyLevel const baseLevel , std vector Pos const vertexPositions , int nextObjVertexIndex for size_t i 0 i vertexPositions . size i float const p vertexPositions i . p printf v f f f n , p 0 , p 1 , p 2 Connect the sequences of limit points center followed by corners into triangle fans for each base face for int i 0 i baseLevel . GetNumFaces i int faceSize baseLevel . GetFaceVertices i . size int vCenter nextObjVertexIndex 1 int vCorner vCenter 1 for int k 0 k faceSize k printf f d d d n , vCenter , vCorner k , vCorner k 1 faceSize nextObjVertexIndex faceSize 1 return nextObjVertexIndex end namespace Command line arguments parsed to provide run-time options class Args public std string inputObjFile Sdc SchemeType schemeType int maxPatchDepth int numPoses Pos poseOffset bool deriv1Flag bool noPatchesFlag bool noOutputFlag public Args int argc , char argv inputObjFile , schemeType Sdc SCHEME_CATMARK , maxPatchDepth 3 , numPoses 0 , poseOffset 1.0f , 0.0f , 0.0f , deriv1Flag false , noPatchesFlag false , noOutputFlag false Parse and assign standard arguments and Obj files ArgOptions args args . Parse argc , argv maxPatchDepth args . GetLevel schemeType ConvertShapeSchemeToSdcType args . GetDefaultScheme const std vector const char objFiles args . GetObjFiles if objFiles . empty for size_t i 1 i objFiles . size i fprintf stderr , Warning .obj file s ignored n , objFiles i inputObjFile std string objFiles 0 Parse remaining arguments specific to this example const std vector const char rargs args . GetRemainingArgs for size_t i 0 i rargs . size i if strcmp rargs i , -d1 deriv1Flag true else if strcmp rargs i , -nopatches noPatchesFlag true else if strcmp rargs i , -poses if i rargs . size numPoses atoi rargs i else if strcmp rargs i , -offset if i rargs . size poseOffset . p 0 float atof rargs i if i rargs . size poseOffset . p 1 float atof rargs i if i rargs . size poseOffset . p 2 float atof rargs i else if strcmp rargs i , -nooutput noOutputFlag true else fprintf stderr , Warning Argument s ignored n , rargs i private Args Assemble the set of locations for the limit points. The resulting vector of LocationArrays can contain arbitrary locations on the limit surface -- with multiple locations for the same patch grouped into a single array. In this case, for each base face, coordinates for the center and its corners are specified -- from which we will construct a triangle fan providing a crude tessellation similar to tutorial_5_2. typedef Far LimitStencilTableFactory LocationArray LocationArray int assembleLimitPointLocations Far TopologyRefiner const refiner , std vector LocationArray locations Coordinates for the center of the face and its corners slightly inset. Unlike most of the public interface for patches, the LocationArray refers to parameteric coordinates as s,t, so that convention will be followed here. Note that the s,t coordinates in a LocationArray are referred to by reference. The memory holding these s,t values must persist while the LimitStencilTable is constructed -- the arrays here are declared as static for that purpose. static float const quadSCoords 5 0.5f , 0.05f , 0.95f , 0.95f , 0.05f static float const quadTCoords 5 0.5f , 0.05f , 0.05f , 0.95f , 0.95f static float const triSCoords 4 0.33f , 0.05f , 0.95f , 0.05f static float const triTCoords 4 0.33f , 0.05f , 0.00f , 0.95f static float const irregSCoords 2 1.0f , 0.05f static float const irregTCoords 2 1.0f , 0.05f Since these are references to patches to be evaluated, we require use of the Ptex indices to identify the top-most parameterized patch, which is essential to dealing with non-quad faces in the case of Catmark. Far TopologyLevel const baseLevel refiner . GetLevel 0 Far PtexIndices basePtexIndices refiner int regFaceSize Sdc SchemeTypeTraits GetRegularFaceSize refiner . GetSchemeType For each base face, simply refer to the s,t arrays for regular quad and triangular patches with a single LocationArray. Otherwise, for irregular faces, the corners of the face come from different patches and so must be referenced in separate LocationArrays. locations . clear int numLimitPoints 0 for int i 0 i baseLevel . GetNumFaces i int baseFaceSize baseLevel . GetFaceVertices i . size int basePtexId basePtexIndices . GetFaceId i bool faceIsRegular baseFaceSize regFaceSize if faceIsRegular All coordinates are on the same top-level patch LocationArray loc loc . ptexIdx basePtexId loc . numLocations baseFaceSize 1 if baseFaceSize 4 loc . s quadSCoords loc . t quadTCoords else loc . s triSCoords loc . t triTCoords locations . push_back loc else Center coordinate is on the first sub-patch while those on near the corners are on each successive sub-patch LocationArray loc loc . numLocations 1 for int j 0 j baseFaceSize j bool isPerimeter j 0 loc . ptexIdx basePtexId isPerimeter j -1 0 loc . s irregSCoords isPerimeter loc . t irregTCoords isPerimeter locations . push_back loc numLimitPoints baseFaceSize 1 return numLimitPoints Load command line arguments and geometry, build the LimitStencilTable for a set of points on the limit surface and compute those points for several orientations of the mesh int main int argc , char argv Args args argc , argv Create or load the base geometry command line arguments allow a .obj file to be specified, providing a TopologyRefiner and a set of base vertex positions to work with std vector Pos basePositions Far TopologyRefiner refinerPtr args . inputObjFile . empty createTopologyRefinerDefault basePositions createTopologyRefinerFromObj args . inputObjFile , args . schemeType , basePositions assert refinerPtr Far TopologyRefiner refiner refinerPtr Far TopologyLevel const baseLevel refiner . GetLevel 0 Use of LimitStencilTable requires either explicit or implicit use of a PatchTable. A PatchTable is not required to construct a LimitStencilTable -- one will be constructed internally for use and discarded -- but explicit construction is recommended to control the many legacy options for PatchTable, rather than relying on internal defaults. Adaptive refinement is required in both cases to indicate the accuracy of the patches. Note that if a TopologyRefiner and PatchTable are not used for any other purpose than computing the limit points, that specifying the subset of faces containing those limit points in the adaptive refinement and PatchTable construction can avoid unnecessary overhead. Far PatchTable patchTablePtr 0 if args . noPatchesFlag refiner . RefineAdaptive Far TopologyRefiner AdaptiveOptions args . maxPatchDepth else Far PatchTableFactory Options patchOptions args . maxPatchDepth patchOptions . useInfSharpPatch true patchOptions . generateLegacySharpCornerPatches false patchOptions . generateVaryingTables false patchOptions . generateFVarTables false patchOptions . endCapType Far PatchTableFactory Options ENDCAP_GREGORY_BASIS refiner . RefineAdaptive patchOptions . GetRefineAdaptiveOptions patchTablePtr Far PatchTableFactory Create refiner , patchOptions assert patchTablePtr Assemble the set of locations for the limit points. For each base face, coordinates for the center and its corners are specified -- from which we will construct a triangle fan providing a crude tessellation similar to tutorial_5_2. std vector LocationArray locations int numLimitPoints assembleLimitPointLocations refiner , locations Construct a LimitStencilTable from the refiner, patch table optional and the collection of limit point locations. Stencils can optionally be created for computing dervatives -- the default is to compute 1st derivative stencils, so be sure to disable that if not necessary Far LimitStencilTableFactory Options limitOptions limitOptions . generate1stDerivatives args . deriv1Flag Far LimitStencilTable const limitStencilTablePtr Far LimitStencilTableFactory Create refiner , locations , 0 , optional StencilTable for the refined points patchTablePtr , optional PatchTable limitOptions assert limitStencilTablePtr Far LimitStencilTable const limitStencilTable limitStencilTablePtr Apply the constructed LimitStencilTable to compute limit positions from the base level vertex positions. This is trivial if computing all positions in one invokation. The UpdateValues method and those for derivatives are overloaded to optionally accept a subrange of indices to distribute the computation std vector Pos limitPositions numLimitPoints limitStencilTable . UpdateValues basePositions , limitPositions Call with the optional subrange limitStencilTable . UpdateValues basePositions , limitPositions , 0 , numLimitPoints 2 limitStencilTable . UpdateValues basePositions , limitPositions , numLimitPoints 2 1 , numLimitPoints Write vertices and faces in Obj format for the original limit points int objVertCount 0 if args . noOutputFlag printf g base_mesh n objVertCount writeToObj baseLevel , limitPositions , objVertCount Recompute the limit points and output faces for different poses of the original mesh -- in this case simply translated. Also optionally compute 1st derivatives though they are not used here std vector Pos posePositions basePositions std vector Pos limitDu args . deriv1Flag numLimitPoints 0 std vector Pos limitDv args . deriv1Flag numLimitPoints 0 for int i 0 i args . numPoses i Trivially transform the base vertex positions and re-compute for size_t j 0 j basePositions . size j posePositions j posePositions j args . poseOffset limitStencilTable . UpdateValues posePositions , limitPositions if args . deriv1Flag limitStencilTable . UpdateDerivs posePositions , limitDu , limitDv if args . noOutputFlag printf n g pose_d n , i objVertCount writeToObj baseLevel , limitPositions , objVertCount delete refinerPtr delete patchTablePtr delete limitStencilTablePtr return EXIT_SUCCESS Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"tags\": \"\",\n+ \"loc\": \"far_tutorial_5_3.html\"\n+ }, {\n \"title\": \"bfr_tutorial_1_4.cpp\",\n- \"text\": \"bfr_tutorial_1_4.cpp bfr_tutorial_1_4.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_1_4bfr_tutorial_1_4.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial builds on the previous tutorial that makes use of the SurfaceFactory, Surface and Tessellation classes for evaluating and tessellating the limit surface of faces of a mesh by illustrating how the presence of additional data in the mesh arrays is handled. As in the previous tutorial, vertex positions and face-varying UVs are provided with the mesh to be evaluated. But here an additional color is interleaved with the position in the vertex data of the mesh and a third component is added to face-varying UV data making it u,v,w. To evaluate the position and 2D UVs while avoiding the color and unused third UV coordinate, the SurfacePointDescriptor class is used to describe the size and stride of the desired data to be evaluated in the arrays of mesh data. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag bool uv2xyzFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false , uv2xyzFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else if strcmp argv i , -uv2xyz uv2xyzFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVtxData , int vtxDataSize , std vector float const meshFVarData , int fvarDataSize , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface typedef Surface PointDescriptor SurfacePoint Identify the source positions and UVs within more general data arrays for the mesh. If position andor UV are not at the start of the vtx andor fvar data, simply offset the head of the array here accordingly bool meshHasUVs meshTopology . GetNumFVarChannels 0 float const meshPosData meshVtxData . data SurfacePoint meshPosPoint 3 , vtxDataSize float const meshUVData meshHasUVs meshFVarData . data 0 SurfacePoint meshUVPoint 2 , fvarDataSize Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing When dealing with face-varying data, an identifier is necessary when constructing Surfaces in order to distinguish the different face-varying data channels. To avoid repeatedly specifying that identifier when only one is present or of interest, it can be specified via the Options. SurfaceFactory Options surfaceOptions if meshHasUVs surfaceOptions . SetDefaultFVarID 0 SurfaceFactory surfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface posSurface Surface uvSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector float outUV std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces surfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surfaces for position and UVs of this face. There are two ways to do this -- both illustrated here Creating Surfaces for the different data interpolation types independently is clear and convenient, but considerable work may be duplicated in the construction process in the case of non-linear face-varying Surfaces. So unless it is known that face-varying interpolation is linear, use of InitSurfaces is generally preferred. Remember also that the face-varying identifier is omitted from the initialization methods here as it was previously assigned to the SurfaceFactoryOptions. In the absence of an assignment of the default FVarID to the Options, a failure to specify the FVarID here will result in failure. The cases below are expanded for illustration purposes, and validity of the resulting Surface is tested here, rather than the return value of initialization methods. bool createSurfacesTogether true if meshHasUVs surfaceFactory . InitVertexSurface faceIndex , posSurface else if createSurfacesTogether surfaceFactory . InitSurfaces faceIndex , posSurface , uvSurface else if surfaceFactory . InitVertexSurface faceIndex , posSurface surfaceFactory . InitFaceVaryingSurface faceIndex , uvSurface if posSurface . IsValid continue Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern posSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Evaluate vertex positions Resize patch point and output arrays int pointSize meshPosPoint . size facePatchPoints . resize posSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate patch point and output arrays float patchPosData facePatchPoints . data SurfacePoint patchPosPoint pointSize posSurface . PreparePatchPoints meshPosData , meshPosPoint , patchPosData , patchPosPoint for int i 0 , j 0 i numOutCoords i , j pointSize posSurface . Evaluate outCoords i 2 , patchPosData , patchPosPoint , outPos j , outDu j , outDv j Evaluate face-varying UVs when present if meshHasUVs Resize patch point and output arrays - note reuse of the same patch point array as position int pointSize meshUVPoint . size facePatchPoints . resize uvSurface . GetNumPatchPoints pointSize outUV . resize numOutCoords pointSize Populate patch point and output arrays float patchUVData facePatchPoints . data SurfacePoint patchUVPoint pointSize uvSurface . PreparePatchPoints meshUVData , meshUVPoint , patchUVData , patchUVPoint for int i 0 , j 0 i numOutCoords i , j pointSize uvSurface . Evaluate outCoords i 2 , patchUVData , patchUVPoint , outUV j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex if meshHasUVs options . uv2xyzFlag objWriter . WriteVertexPositions outUV , 2 objWriter . WriteFaces outFacets , tessFacetSize , false , false else objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv if meshHasUVs objWriter . WriteVertexUVs outUV objWriter . WriteFaces outFacets , tessFacetSize , true , meshHasUVs Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE Expand the loaded position and UV arrays to include additional data initialized with -1 for distinction, e.g. add a 4-tuple for RGBA color to the vertex data and add a third field w to the face-varying data int numPos int meshVtxPositions . size 3 int vtxSize 7 std vector float vtxData numPos vtxSize , -1.0f for int i 0 i numPos i vtxData i vtxSize meshVtxPositions i 3 vtxData i vtxSize 1 meshVtxPositions i 3 1 vtxData i vtxSize 2 meshVtxPositions i 3 2 int numUVs int meshFVarUVs . size 2 int fvarSize 3 std vector float fvarData numUVs fvarSize , -1.0f for int i 0 i numUVs i fvarData i fvarSize meshFVarUVs i 2 fvarData i fvarSize 1 meshFVarUVs i 2 1 Pass the expanded data arrays along with their respective strides tessellateToObj meshTopology , vtxData , vtxSize , fvarData , fvarSize , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"bfr_tutorial_1_4.cpp bfr_tutorial_1_4.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_1_4bfr_tutorial_1_4.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial builds on the previous tutorial that makes use of the SurfaceFactory, Surface and Tessellation classes for evaluating and tessellating the limit surface of faces of a mesh by illustrating how the presence of additional data in the mesh arrays is handled. As in the previous tutorial, vertex positions and face-varying UVs are provided with the mesh to be evaluated. But here an additional color is interleaved with the position in the vertex data of the mesh and a third component is added to face-varying UV data making it u,v,w. To evaluate the position and 2D UVs while avoiding the color and unused third UV coordinate, the SurfacePointDescriptor class is used to describe the size and stride of the desired data to be evaluated in the arrays of mesh data. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag bool uv2xyzFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false , uv2xyzFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else if strcmp argv i , -uv2xyz uv2xyzFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVtxData , int vtxDataSize , std vector float const meshFVarData , int fvarDataSize , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface typedef Surface PointDescriptor SurfacePoint Identify the source positions and UVs within more general data arrays for the mesh. If position andor UV are not at the start of the vtx andor fvar data, simply offset the head of the array here accordingly bool meshHasUVs meshTopology . GetNumFVarChannels 0 float const meshPosData meshVtxData . data SurfacePoint meshPosPoint 3 , vtxDataSize float const meshUVData meshHasUVs meshFVarData . data 0 SurfacePoint meshUVPoint 2 , fvarDataSize Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing When dealing with face-varying data, an identifier is necessary when constructing Surfaces in order to distinguish the different face-varying data channels. To avoid repeatedly specifying that identifier when only one is present or of interest, it can be specified via the Options. SurfaceFactory Options surfaceOptions if meshHasUVs surfaceOptions . SetDefaultFVarID 0 SurfaceFactory surfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface posSurface Surface uvSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector float outUV std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces surfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surfaces for position and UVs of this face. There are two ways to do this -- both illustrated here Creating Surfaces for the different data interpolation types independently is clear and convenient, but considerable work may be duplicated in the construction process in the case of non-linear face-varying Surfaces. So unless it is known that face-varying interpolation is linear, use of InitSurfaces is generally preferred. Remember also that the face-varying identifier is omitted from the initialization methods here as it was previously assigned to the SurfaceFactoryOptions. In the absence of an assignment of the default FVarID to the Options, a failure to specify the FVarID here will result in failure. The cases below are expanded for illustration purposes, and validity of the resulting Surface is tested here, rather than the return value of initialization methods. bool createSurfacesTogether true if meshHasUVs surfaceFactory . InitVertexSurface faceIndex , posSurface else if createSurfacesTogether surfaceFactory . InitSurfaces faceIndex , posSurface , uvSurface else if surfaceFactory . InitVertexSurface faceIndex , posSurface surfaceFactory . InitFaceVaryingSurface faceIndex , uvSurface if posSurface . IsValid continue Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern posSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Evaluate vertex positions Resize patch point and output arrays int pointSize meshPosPoint . size facePatchPoints . resize posSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate patch point and output arrays float patchPosData facePatchPoints . data SurfacePoint patchPosPoint pointSize posSurface . PreparePatchPoints meshPosData , meshPosPoint , patchPosData , patchPosPoint for int i 0 , j 0 i numOutCoords i , j pointSize posSurface . Evaluate outCoords i 2 , patchPosData , patchPosPoint , outPos j , outDu j , outDv j Evaluate face-varying UVs when present if meshHasUVs Resize patch point and output arrays - note reuse of the same patch point array as position int pointSize meshUVPoint . size facePatchPoints . resize uvSurface . GetNumPatchPoints pointSize outUV . resize numOutCoords pointSize Populate patch point and output arrays float patchUVData facePatchPoints . data SurfacePoint patchUVPoint pointSize uvSurface . PreparePatchPoints meshUVData , meshUVPoint , patchUVData , patchUVPoint for int i 0 , j 0 i numOutCoords i , j pointSize uvSurface . Evaluate outCoords i 2 , patchUVData , patchUVPoint , outUV j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex if meshHasUVs options . uv2xyzFlag objWriter . WriteVertexPositions outUV , 2 objWriter . WriteFaces outFacets , tessFacetSize , false , false else objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv if meshHasUVs objWriter . WriteVertexUVs outUV objWriter . WriteFaces outFacets , tessFacetSize , true , meshHasUVs Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE Expand the loaded position and UV arrays to include additional data initialized with -1 for distinction, e.g. add a 4-tuple for RGBA color to the vertex data and add a third field w to the face-varying data int numPos int meshVtxPositions . size 3 int vtxSize 7 std vector float vtxData numPos vtxSize , -1.0f for int i 0 i numPos i vtxData i vtxSize meshVtxPositions i 3 vtxData i vtxSize 1 meshVtxPositions i 3 1 vtxData i vtxSize 2 meshVtxPositions i 3 2 int numUVs int meshFVarUVs . size 2 int fvarSize 3 std vector float fvarData numUVs fvarSize , -1.0f for int i 0 i numUVs i fvarData i fvarSize meshFVarUVs i 2 fvarData i fvarSize 1 meshFVarUVs i 2 1 Pass the expanded data arrays along with their respective strides tessellateToObj meshTopology , vtxData , vtxSize , fvarData , fvarSize , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"bfr_tutorial_1_4.html\"\n }, {\n- \"title\": \"bfr_tutorial_1_5.cpp\",\n- \"text\": \"bfr_tutorial_1_5.cpp bfr_tutorial_1_5.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_1_5bfr_tutorial_1_5.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial is an alternative to an earlier tutorial that showed uniform tessellation. This version differs by evaluating the points of the tessellation using limit stencils instead of the standard Surface evaluation methods. Limit stencils factor the evaluation into a set of coefficients for each control point affecting the Surface. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface faceSurface std vector float faceControlPoints std vector float limitStencils std vector float outCoords std vector float outPos , outDu , outDv std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surface for this face -- if valid skipping holes and boundary faces in some rare cases if meshSurfaceFactory . InitVertexSurface faceIndex , faceSurface continue Resize stencils and control point arrays based on the number of control points for the Surface int numControlPoints faceSurface . GetNumControlPoints limitStencils . resize 3 numControlPoints float pStencil limitStencils . data float duStencil limitStencils . data numControlPoints float dvStencil limitStencils . data numControlPoints 2 Limit stencils can be applied using the control points in a local array or directy from the mesh. Both are shown here, so if using the local array, resize and populate it bool gatherControlPoints true if gatherControlPoints faceControlPoints . resize numControlPoints 3 faceSurface . GatherControlPoints meshVertexPositions . data , 3 , faceControlPoints . data , 3 Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern faceSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Evaluate and apply stencils to compute points of the tessellation outPos . resize numOutCoords 3 outDu . resize numOutCoords 3 outDv . resize numOutCoords 3 for int i 0 i numOutCoords i float const uv outCoords . data i 2 faceSurface . EvaluateStencil uv , pStencil , duStencil , dvStencil float p outPos . data i 3 float du outDu . data i 3 float dv outDv . data i 3 if gatherControlPoints float const controlPoints faceControlPoints . data faceSurface . ApplyStencil pStencil , controlPoints , 3 , p faceSurface . ApplyStencil duStencil , controlPoints , 3 , du faceSurface . ApplyStencil dvStencil , controlPoints , 3 , dv else float const meshPoints meshVertexPositions . data faceSurface . ApplyStencilFromMesh pStencil , meshPoints , 3 , p faceSurface . ApplyStencilFromMesh duStencil , meshPoints , 3 , du faceSurface . ApplyStencilFromMesh dvStencil , meshPoints , 3 , dv Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outFacets , tessFacetSize , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"far_tutorial_5_2.cpp\",\n+ \"text\": \"far_tutorial_5_2.cpp far_tutorial_5_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_5_2far_tutorial_5_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to manage the limit surface of a potentially large mesh by creating groups of patches for selected faces of the mesh. Familiarity with construction and evaluation of a PatchTable is assumed see tutorial_5_1. When the patches for a mesh do not need to be retained for further use, e.g. when simply computing points for a tessellation, the time and space required to construct a single large PatchTable can be considerable. By constructing, evaluating and discarding smaller PatchTables for subsets of the mesh, the high transient memory cost can be avoided when computed serially. When computed in parallel, there may be little memory savings, but the construction time can then be distributed. This tutorial creates simple geometry currently a lattice of cubes that can be expanded in complexity with a simple multiplier. The collection of faces are then divided into a specified number of groups from which patches will be constructed and evaluated. A simple tessellation a triangle fan around the midpoint of each face is then written in Obj format to the standard output. include ......regressioncommonarg_utils.h include ......regressioncommonfar_utils.h include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include opensubdivfarpatchTableFactory.h include opensubdivfarpatchMap.h include opensubdivfarptexIndices.h include cassert include cstdio include cstring include fstream include sstream using namespace OpenSubdiv using Far Index Global utilities in this namespace are not relevant to the tutorial. They simply serve to construct some default geometry to be processed in the form of a TopologyRefiner and vector of vertex positions. namespace Simple structs for x,y,z position and a 3-tuple for the set of vertices of a triangle struct Pos Pos Pos float x , float y , float z p 0 x , p 1 y , p 2 z Pos operator Pos const op const return Pos p 0 op . p 0 , p 1 op . p 1 , p 2 op . p 2 Clear and AddWithWeight required for interpolation void Clear void 0 p 0 p 1 p 2 0.0f void AddWithWeight Pos const src , float weight p 0 weight src . p 0 p 1 weight src . p 1 p 2 weight src . p 2 float p 3 typedef std vector Pos PosVector struct Tri Tri Tri int a , int b , int c v 0 a , v 1 b , v 2 c int v 3 typedef std vector Tri TriVector Functions to populate the topology and geometry arrays with simple shapes that we can multiply to increase complexity void appendDefaultPrimitive Pos const origin , std vector int vertsPerFace , std vector Index faceVerts , std vector Pos positionsPerVert Local topology and position of a cube centered at origin static float const cubePositions 8 3 -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f static int const cubeFaceVerts 6 4 0 , 3 , 2 , 1 , 4 , 5 , 6 , 7 , 0 , 4 , 7 , 3 , 1 , 2 , 6 , 5 , 0 , 1 , 5 , 4 , 3 , 7 , 6 , 2 Identify the next vertex before appending vertex positions int baseVertex int positionsPerVert . size for int i 0 i 8 i float const p cubePositions i positionsPerVert . push_back origin Pos p 0 , p 1 , p 2 Append number of verts-per-face and face-vertices for each face for int i 0 i 6 i vertsPerFace . push_back 4 for int j 0 j 4 j faceVerts . push_back baseVertex cubeFaceVerts i j void createDefaultGeometry int multiplier , std vector int vertsPerFace , std vector Index faceVerts , std vector Pos positionsPerVert Default primitive is currently a cube int const vertsPerPrimitive 8 int const facesPerPrimitive 6 int const faceVertsPerPrimitive 24 int nPrimitives multiplier multiplier multiplier positionsPerVert . reserve nPrimitives vertsPerPrimitive vertsPerFace . reserve nPrimitives facesPerPrimitive faceVerts . reserve nPrimitives faceVertsPerPrimitive for int x 0 x multiplier x for int y 0 y multiplier y for int z 0 z multiplier z appendDefaultPrimitive Pos float x 2.0f , float y 2.0f , float z 2.0f , vertsPerFace , faceVerts , positionsPerVert Create a TopologyRefiner from default geometry created above Far TopologyRefiner createTopologyRefinerDefault int multiplier , PosVector posVector std vector int topVertsPerFace std vector Index topFaceVerts createDefaultGeometry multiplier , topVertsPerFace , topFaceVerts , posVector typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_AND_CORNER Descriptor desc desc . numVertices int posVector . size desc . numFaces int topVertsPerFace . size desc . numVertsPerFace topVertsPerFace 0 desc . vertIndicesPerFace topFaceVerts 0 Instantiate a FarTopologyRefiner from the descriptor. Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options if refiner 0 exit EXIT_FAILURE bool dumpDefaultGeometryToObj false if dumpDefaultGeometryToObj int nVerts int posVector . size for int i 0 i nVerts i float const p posVector i . p printf v f f f n , p 0 , p 1 , p 2 int const fVerts topFaceVerts 0 int nFaces int topVertsPerFace . size for int i 0 i nFaces i printf f for int j 0 j topVertsPerFace i j printf d , 1 fVerts printf n exit EXIT_SUCCESS return refiner Create a TopologyRefiner from a specified Obj file geometry created internally Far TopologyRefiner createTopologyRefinerFromObj std string const objFileName , Sdc SchemeType schemeType , PosVector posVector const char filename objFileName . c_str const Shape shape 0 std ifstream ifs filename if ifs std stringstream ss ss ifs . rdbuf ifs . close std string shapeString ss . str shape Shape parseObj shapeString . c_str , ConvertSdcTypeToShapeScheme schemeType , false if shape 0 fprintf stderr , Error Cannot create Shape from .obj file s n , filename return 0 else fprintf stderr , Error Cannot open .obj file s n , filename return 0 Sdc SchemeType sdcType GetSdcType shape Sdc Options sdcOptions GetSdcOptions shape Far TopologyRefiner refiner Far TopologyRefinerFactory Shape Create shape , Far TopologyRefinerFactory Shape Options sdcType , sdcOptions if refiner 0 fprintf stderr , Error Unable to construct TopologyRefiner from .obj file s n , filename return 0 int numVertices refiner - GetNumVerticesTotal posVector . resize numVertices std memcpy posVector 0 . p 0 , shape - verts 0 , numVertices 3 sizeof float delete shape return refiner end namespace The PatchGroup bundles objects used to create and evaluate a sparse set of patches. Its construction creates a PatchTable and all other objects necessary to evaluate patches associated with the specified subset of faces provided. A simple method to tessellate a specified face is provided. Note that, since the data buffers for the base level and refined levels are separate we want to avoid copying primvar data for the base level of a potentially large mesh, that patch evaluation needs to account for the separation when combining control points. struct PatchGroup PatchGroup Far PatchTableFactory Options patchOptions , Far TopologyRefiner const baseRefinerArg , Far PtexIndices const basePtexIndicesArg , std vector Pos const basePositionsArg , std vector Index const baseFacesArg PatchGroup void TessellateBaseFace int face , PosVector tessPoints , TriVector tessTris const Const reference members Far TopologyRefiner const baseRefiner Far PtexIndices const basePtexIndices std vector Pos const basePositions std vector Index const baseFaces Members constructed to evaluate patches Far PatchTable patchTable Far PatchMap patchMap int patchFaceSize std vector Pos localPositions PatchGroup PatchGroup Far PatchTableFactory Options patchOptions , Far TopologyRefiner const baseRefinerArg , Far PtexIndices const basePtexIndicesArg , std vector Pos const basePositionsArg , std vector Index const baseFacesArg baseRefiner baseRefinerArg , basePtexIndices basePtexIndicesArg , basePositions basePositionsArg , baseFaces baseFacesArg Create a local refiner sharing the base level, apply adaptive refinement to the given subset of base faces, and construct a patch table and its associated map for the same set of faces Far ConstIndexArray groupFaces baseFaces 0 , int baseFaces . size Far TopologyRefiner localRefiner Far TopologyRefinerFactory Far TopologyDescriptor Create baseRefiner localRefiner - RefineAdaptive patchOptions . GetRefineAdaptiveOptions , groupFaces patchTable Far PatchTableFactory Create localRefiner , patchOptions , groupFaces patchMap new Far PatchMap patchTable patchFaceSize Sdc SchemeTypeTraits GetRegularFaceSize baseRefiner . GetSchemeType Compute the number of refined and local points needed to evaluate the patches, allocate and interpolate. This varies from tutorial_5_1 in that the primvar buffer for the base vertices is separate from the refined vertices and local patch points which must also be accounted for when evaluating the patches. int nBaseVertices localRefiner - GetLevel 0 . GetNumVertices int nRefinedVertices localRefiner - GetNumVerticesTotal - nBaseVertices int nLocalPoints patchTable - GetNumLocalPoints localPositions . resize nRefinedVertices nLocalPoints if nRefinedVertices Far PrimvarRefiner primvarRefiner localRefiner Pos const src basePositions 0 Pos dst localPositions 0 for int level 1 level localRefiner - GetNumLevels level primvarRefiner . Interpolate level , src , dst src dst dst localRefiner - GetLevel level . GetNumVertices if nLocalPoints patchTable - GetLocalPointStencilTable - UpdateValues basePositions 0 , nBaseVertices , localPositions 0 , localPositions nRefinedVertices delete localRefiner PatchGroup PatchGroup delete patchTable delete patchMap void PatchGroup TessellateBaseFace int face , PosVector tessPoints , TriVector tessTris const Tesselate the face with points at the midpoint of the face and at each corner, and triangles connecting the midpoint to each edge. Irregular faces require an aribrary number of corners points, but all are at the origin of the child face of the irregular base face float const quadPoints 5 2 0.5f , 0.5f , 0.0f , 0.0f , 1.0f , 0.0f , 1.0f , 1.0f , 0.0f , 1.0f float const triPoints 4 2 0.5f , 0.5f , 0.0f , 0.0f , 1.0f , 0.0f , 0.0f , 1.0f float const irregPoints 4 2 1.0f , 1.0f , 0.0f , 0.0f Determine the topology of the given base face and the resulting tessellation points and faces to generate int baseFace baseFaces face int faceSize baseRefiner . GetLevel 0 . GetFaceVertices baseFace . size bool faceIsIrregular faceSize patchFaceSize int nTessPoints faceSize 1 int nTessFaces faceSize tessPoints . resize nTessPoints tessTris . resize nTessFaces Compute the mid and corner points -- remember that for an irregular face, we must reference the individual ptex faces for each corner int ptexFace basePtexIndices . GetFaceId baseFace int numBaseVerts int basePositions . size for int i 0 i nTessPoints i Choose the s,t coordinate from the fixed tessellation float const st faceIsIrregular irregPoints i 0 faceSize 4 quadPoints i triPoints i Locate the patch corresponding to the face ptex idx and s,t and evaluate int patchFace ptexFace if faceIsIrregular i 0 patchFace i - 1 Far PatchTable PatchHandle const handle patchMap - FindPatch patchFace , st 0 , st 1 assert handle float pWeights 20 patchTable - EvaluateBasis handle , st 0 , st 1 , pWeights Identify the patch cvs and combine with the evaluated weights -- remember to distinguish cvs in the base level Far ConstIndexArray cvIndices patchTable - GetPatchVertices handle Pos pos tessPoints i pos . Clear for int cv 0 cv cvIndices . size cv int cvIndex cvIndices cv if cvIndex numBaseVerts pos . AddWithWeight basePositions cvIndex , pWeights cv else pos . AddWithWeight localPositions cvIndex - numBaseVerts , pWeights cv Assign triangles connecting the midpoint of the base face to the points computed at the ends of each of its edges for int i 0 i nTessFaces i tessTris i Tri 0 , 1 i , 1 i 1 faceSize Command line arguments parsed to provide run-time options class Args public std string inputObjFile Sdc SchemeType schemeType int geoMultiplier int maxPatchDepth int numPatchGroups bool noTessFlag bool noOutputFlag public Args int argc , char argv inputObjFile , schemeType Sdc SCHEME_CATMARK , geoMultiplier 10 , maxPatchDepth 3 , numPatchGroups 10 , noTessFlag false , noOutputFlag false Parse and assign standard arguments and Obj files ArgOptions args args . Parse argc , argv maxPatchDepth args . GetLevel schemeType ConvertShapeSchemeToSdcType args . GetDefaultScheme const std vector const char objFiles args . GetObjFiles if objFiles . empty for size_t i 1 i objFiles . size i fprintf stderr , Warning .obj file s ignored n , objFiles i inputObjFile std string objFiles 0 Parse remaining arguments specific to this example const std vector const char rargs args . GetRemainingArgs for size_t i 0 i rargs . size i if strcmp rargs i , -groups if i rargs . size numPatchGroups atoi rargs i else if strcmp rargs i , -mult if i rargs . size geoMultiplier atoi rargs i else if strcmp rargs i , -notess noTessFlag true else if strcmp rargs i , -nooutput noOutputFlag true else fprintf stderr , Warning Argument s ignored n , rargs i private Args Load command line arguments and geometry, then divide the mesh into groups of faces from which to create and tessellate patches int main int argc , char argv Args args argc , argv Create or load the base geometry command line arguments allow a .obj file to be specified. In addition to the TopologyRefiner and set of positions for the base vertices, a set of PtexIndices is also required to evaluate patches, so build it here once for use elsewhere std vector Pos basePositions Far TopologyRefiner baseRefinerPtr args . inputObjFile . empty createTopologyRefinerDefault args . geoMultiplier , basePositions createTopologyRefinerFromObj args . inputObjFile , args . schemeType , basePositions assert baseRefinerPtr Far TopologyRefiner baseRefiner baseRefinerPtr Far PtexIndices basePtexIndices baseRefiner Determine the sizes of the patch groups specified -- there will be two sizes that differ by one to account for unequal division int numBaseFaces baseRefiner . GetNumFacesTotal int numPatchGroups args . numPatchGroups if numPatchGroups numBaseFaces numPatchGroups numBaseFaces else if numPatchGroups 1 numPatchGroups 1 int lesserGroupSize numBaseFaces numPatchGroups int numLargerGroups numBaseFaces - numPatchGroups lesserGroupSize Define the options used to construct the patches for each group. Unless suppressed, a tessellation in Obj format will also be printed to standard output, so keep track of the vertex indices. Far PatchTableFactory Options patchOptions args . maxPatchDepth patchOptions . generateVaryingTables false patchOptions . shareEndCapPatchPoints false patchOptions . endCapType Far PatchTableFactory Options ENDCAP_GREGORY_BASIS int objVertCount 0 PosVector tessPoints TriVector tessFaces for int i 0 i numPatchGroups i Initialize a vector with a group of base faces from which to create and evaluate patches Index minFace i lesserGroupSize std min i , numLargerGroups Index maxFace minFace lesserGroupSize i numLargerGroups std vector Far Index baseFaces maxFace - minFace for int face minFace face maxFace face baseFaces face - minFace face Declare a PatchGroup and tessellate its base faces -- generating vertices and faces in Obj format to standard output PatchGroup patchGroup patchOptions , baseRefiner , basePtexIndices , basePositions , baseFaces if args . noTessFlag continue if args . noOutputFlag printf g patchGroup_d n , i for int j 0 j int baseFaces . size j patchGroup . TessellateBaseFace j , tessPoints , tessFaces if args . noOutputFlag int nVerts int tessPoints . size for int k 0 k nVerts k float const p tessPoints k . p printf v f f f n , p 0 , p 1 , p 2 int nTris int tessFaces . size int vBase 1 objVertCount for int k 0 k nTris k int const v tessFaces k . v printf f d d d n , vBase v 0 , vBase v 1 , vBase v 2 objVertCount nVerts delete baseRefinerPtr return EXIT_SUCCESS Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"bfr_tutorial_1_5.html\"\n+ \"loc\": \"far_tutorial_5_2.html\"\n }, {\n \"title\": \"bfr_tutorial_1_3.cpp\",\n- \"text\": \"bfr_tutorial_1_3.cpp bfr_tutorial_1_3.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_1_3bfr_tutorial_1_3.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial builds on the previous tutorial that makes use of the SurfaceFactory, Surface and Tessellation classes for evaluating and tessellating the limit surface of faces of a mesh by adding support for the evaluation of face-varying UVs. If UVs exist in the given mesh, they will be evaluated and included with the vertex positions and normals previously illustrated as part of the tessellation written to the Obj file. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag bool uv2xyzFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false , uv2xyzFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else if strcmp argv i , -uv2xyz uv2xyzFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , std vector float const meshFaceVaryingUVs , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing When dealing with face-varying data, an identifier is necessary when constructing Surfaces in order to distinguish the different face-varying data channels. To avoid repeatedly specifying that identifier when only one is present or of interest, it can be specified via the Options. bool meshHasUVs meshTopology . GetNumFVarChannels 0 SurfaceFactory Options surfaceOptions if meshHasUVs surfaceOptions . SetDefaultFVarID 0 SurfaceFactory surfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface posSurface Surface uvSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector float outUV std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces surfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surfaces for position and UVs of this face. There are two ways to do this -- both illustrated here Creating Surfaces for the different data interpolation types independently is clear and convenient, but considerable work may be duplicated in the construction process in the case of non-linear face-varying Surfaces. So unless it is known that face-varying interpolation is linear, use of InitSurfaces is generally preferred. Remember also that the face-varying identifier is omitted from the initialization methods here as it was previously assigned to the SurfaceFactoryOptions. In the absence of an assignment of the default FVarID to the Options, a failure to specify the FVarID here will result in failure. The cases below are expanded for illustration purposes, and validity of the resulting Surface is tested here, rather than the return value of initialization methods. bool createSurfacesTogether true if meshHasUVs surfaceFactory . InitVertexSurface faceIndex , posSurface else if createSurfacesTogether surfaceFactory . InitSurfaces faceIndex , posSurface , uvSurface else if surfaceFactory . InitVertexSurface faceIndex , posSurface surfaceFactory . InitFaceVaryingSurface faceIndex , uvSurface if posSurface . IsValid continue Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern posSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Evaluate vertex positions Resize patch point and output arrays int pointSize 3 facePatchPoints . resize posSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate patch point and output arrays posSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize posSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outPos j , outDu j , outDv j Evaluate face-varying UVs when present if meshHasUVs Resize patch point and output arrays - note reuse of the same patch point array as position int pointSize 2 facePatchPoints . resize uvSurface . GetNumPatchPoints pointSize outUV . resize numOutCoords pointSize Populate patch point and output arrays uvSurface . PreparePatchPoints meshFaceVaryingUVs . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize uvSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outUV j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex if meshHasUVs options . uv2xyzFlag objWriter . WriteVertexPositions outUV , 2 objWriter . WriteFaces outFacets , tessFacetSize , false , false else objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv if meshHasUVs objWriter . WriteVertexUVs outUV objWriter . WriteFaces outFacets , tessFacetSize , true , meshHasUVs Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , meshFVarUVs , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"bfr_tutorial_1_3.cpp bfr_tutorial_1_3.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_1_3bfr_tutorial_1_3.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial builds on the previous tutorial that makes use of the SurfaceFactory, Surface and Tessellation classes for evaluating and tessellating the limit surface of faces of a mesh by adding support for the evaluation of face-varying UVs. If UVs exist in the given mesh, they will be evaluated and included with the vertex positions and normals previously illustrated as part of the tessellation written to the Obj file. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag bool uv2xyzFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false , uv2xyzFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else if strcmp argv i , -uv2xyz uv2xyzFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , std vector float const meshFaceVaryingUVs , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing When dealing with face-varying data, an identifier is necessary when constructing Surfaces in order to distinguish the different face-varying data channels. To avoid repeatedly specifying that identifier when only one is present or of interest, it can be specified via the Options. bool meshHasUVs meshTopology . GetNumFVarChannels 0 SurfaceFactory Options surfaceOptions if meshHasUVs surfaceOptions . SetDefaultFVarID 0 SurfaceFactory surfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface posSurface Surface uvSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector float outUV std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces surfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surfaces for position and UVs of this face. There are two ways to do this -- both illustrated here Creating Surfaces for the different data interpolation types independently is clear and convenient, but considerable work may be duplicated in the construction process in the case of non-linear face-varying Surfaces. So unless it is known that face-varying interpolation is linear, use of InitSurfaces is generally preferred. Remember also that the face-varying identifier is omitted from the initialization methods here as it was previously assigned to the SurfaceFactoryOptions. In the absence of an assignment of the default FVarID to the Options, a failure to specify the FVarID here will result in failure. The cases below are expanded for illustration purposes, and validity of the resulting Surface is tested here, rather than the return value of initialization methods. bool createSurfacesTogether true if meshHasUVs surfaceFactory . InitVertexSurface faceIndex , posSurface else if createSurfacesTogether surfaceFactory . InitSurfaces faceIndex , posSurface , uvSurface else if surfaceFactory . InitVertexSurface faceIndex , posSurface surfaceFactory . InitFaceVaryingSurface faceIndex , uvSurface if posSurface . IsValid continue Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern posSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Evaluate vertex positions Resize patch point and output arrays int pointSize 3 facePatchPoints . resize posSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate patch point and output arrays posSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize posSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outPos j , outDu j , outDv j Evaluate face-varying UVs when present if meshHasUVs Resize patch point and output arrays - note reuse of the same patch point array as position int pointSize 2 facePatchPoints . resize uvSurface . GetNumPatchPoints pointSize outUV . resize numOutCoords pointSize Populate patch point and output arrays uvSurface . PreparePatchPoints meshFaceVaryingUVs . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize uvSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outUV j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex if meshHasUVs options . uv2xyzFlag objWriter . WriteVertexPositions outUV , 2 objWriter . WriteFaces outFacets , tessFacetSize , false , false else objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv if meshHasUVs objWriter . WriteVertexUVs outUV objWriter . WriteFaces outFacets , tessFacetSize , true , meshHasUVs Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , meshFVarUVs , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"bfr_tutorial_1_3.html\"\n }, {\n- \"title\": \"far_tutorial_5_2.cpp\",\n- \"text\": \"far_tutorial_5_2.cpp far_tutorial_5_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_5_2far_tutorial_5_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to manage the limit surface of a potentially large mesh by creating groups of patches for selected faces of the mesh. Familiarity with construction and evaluation of a PatchTable is assumed see tutorial_5_1. When the patches for a mesh do not need to be retained for further use, e.g. when simply computing points for a tessellation, the time and space required to construct a single large PatchTable can be considerable. By constructing, evaluating and discarding smaller PatchTables for subsets of the mesh, the high transient memory cost can be avoided when computed serially. When computed in parallel, there may be little memory savings, but the construction time can then be distributed. This tutorial creates simple geometry currently a lattice of cubes that can be expanded in complexity with a simple multiplier. The collection of faces are then divided into a specified number of groups from which patches will be constructed and evaluated. A simple tessellation a triangle fan around the midpoint of each face is then written in Obj format to the standard output. include ......regressioncommonarg_utils.h include ......regressioncommonfar_utils.h include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include opensubdivfarpatchTableFactory.h include opensubdivfarpatchMap.h include opensubdivfarptexIndices.h include cassert include cstdio include cstring include fstream include sstream using namespace OpenSubdiv using Far Index Global utilities in this namespace are not relevant to the tutorial. They simply serve to construct some default geometry to be processed in the form of a TopologyRefiner and vector of vertex positions. namespace Simple structs for x,y,z position and a 3-tuple for the set of vertices of a triangle struct Pos Pos Pos float x , float y , float z p 0 x , p 1 y , p 2 z Pos operator Pos const op const return Pos p 0 op . p 0 , p 1 op . p 1 , p 2 op . p 2 Clear and AddWithWeight required for interpolation void Clear void 0 p 0 p 1 p 2 0.0f void AddWithWeight Pos const src , float weight p 0 weight src . p 0 p 1 weight src . p 1 p 2 weight src . p 2 float p 3 typedef std vector Pos PosVector struct Tri Tri Tri int a , int b , int c v 0 a , v 1 b , v 2 c int v 3 typedef std vector Tri TriVector Functions to populate the topology and geometry arrays with simple shapes that we can multiply to increase complexity void appendDefaultPrimitive Pos const origin , std vector int vertsPerFace , std vector Index faceVerts , std vector Pos positionsPerVert Local topology and position of a cube centered at origin static float const cubePositions 8 3 -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f static int const cubeFaceVerts 6 4 0 , 3 , 2 , 1 , 4 , 5 , 6 , 7 , 0 , 4 , 7 , 3 , 1 , 2 , 6 , 5 , 0 , 1 , 5 , 4 , 3 , 7 , 6 , 2 Identify the next vertex before appending vertex positions int baseVertex int positionsPerVert . size for int i 0 i 8 i float const p cubePositions i positionsPerVert . push_back origin Pos p 0 , p 1 , p 2 Append number of verts-per-face and face-vertices for each face for int i 0 i 6 i vertsPerFace . push_back 4 for int j 0 j 4 j faceVerts . push_back baseVertex cubeFaceVerts i j void createDefaultGeometry int multiplier , std vector int vertsPerFace , std vector Index faceVerts , std vector Pos positionsPerVert Default primitive is currently a cube int const vertsPerPrimitive 8 int const facesPerPrimitive 6 int const faceVertsPerPrimitive 24 int nPrimitives multiplier multiplier multiplier positionsPerVert . reserve nPrimitives vertsPerPrimitive vertsPerFace . reserve nPrimitives facesPerPrimitive faceVerts . reserve nPrimitives faceVertsPerPrimitive for int x 0 x multiplier x for int y 0 y multiplier y for int z 0 z multiplier z appendDefaultPrimitive Pos float x 2.0f , float y 2.0f , float z 2.0f , vertsPerFace , faceVerts , positionsPerVert Create a TopologyRefiner from default geometry created above Far TopologyRefiner createTopologyRefinerDefault int multiplier , PosVector posVector std vector int topVertsPerFace std vector Index topFaceVerts createDefaultGeometry multiplier , topVertsPerFace , topFaceVerts , posVector typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_AND_CORNER Descriptor desc desc . numVertices int posVector . size desc . numFaces int topVertsPerFace . size desc . numVertsPerFace topVertsPerFace 0 desc . vertIndicesPerFace topFaceVerts 0 Instantiate a FarTopologyRefiner from the descriptor. Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options if refiner 0 exit EXIT_FAILURE bool dumpDefaultGeometryToObj false if dumpDefaultGeometryToObj int nVerts int posVector . size for int i 0 i nVerts i float const p posVector i . p printf v f f f n , p 0 , p 1 , p 2 int const fVerts topFaceVerts 0 int nFaces int topVertsPerFace . size for int i 0 i nFaces i printf f for int j 0 j topVertsPerFace i j printf d , 1 fVerts printf n exit EXIT_SUCCESS return refiner Create a TopologyRefiner from a specified Obj file geometry created internally Far TopologyRefiner createTopologyRefinerFromObj std string const objFileName , Sdc SchemeType schemeType , PosVector posVector const char filename objFileName . c_str const Shape shape 0 std ifstream ifs filename if ifs std stringstream ss ss ifs . rdbuf ifs . close std string shapeString ss . str shape Shape parseObj shapeString . c_str , ConvertSdcTypeToShapeScheme schemeType , false if shape 0 fprintf stderr , Error Cannot create Shape from .obj file s n , filename return 0 else fprintf stderr , Error Cannot open .obj file s n , filename return 0 Sdc SchemeType sdcType GetSdcType shape Sdc Options sdcOptions GetSdcOptions shape Far TopologyRefiner refiner Far TopologyRefinerFactory Shape Create shape , Far TopologyRefinerFactory Shape Options sdcType , sdcOptions if refiner 0 fprintf stderr , Error Unable to construct TopologyRefiner from .obj file s n , filename return 0 int numVertices refiner - GetNumVerticesTotal posVector . resize numVertices std memcpy posVector 0 . p 0 , shape - verts 0 , numVertices 3 sizeof float delete shape return refiner end namespace The PatchGroup bundles objects used to create and evaluate a sparse set of patches. Its construction creates a PatchTable and all other objects necessary to evaluate patches associated with the specified subset of faces provided. A simple method to tessellate a specified face is provided. Note that, since the data buffers for the base level and refined levels are separate we want to avoid copying primvar data for the base level of a potentially large mesh, that patch evaluation needs to account for the separation when combining control points. struct PatchGroup PatchGroup Far PatchTableFactory Options patchOptions , Far TopologyRefiner const baseRefinerArg , Far PtexIndices const basePtexIndicesArg , std vector Pos const basePositionsArg , std vector Index const baseFacesArg PatchGroup void TessellateBaseFace int face , PosVector tessPoints , TriVector tessTris const Const reference members Far TopologyRefiner const baseRefiner Far PtexIndices const basePtexIndices std vector Pos const basePositions std vector Index const baseFaces Members constructed to evaluate patches Far PatchTable patchTable Far PatchMap patchMap int patchFaceSize std vector Pos localPositions PatchGroup PatchGroup Far PatchTableFactory Options patchOptions , Far TopologyRefiner const baseRefinerArg , Far PtexIndices const basePtexIndicesArg , std vector Pos const basePositionsArg , std vector Index const baseFacesArg baseRefiner baseRefinerArg , basePtexIndices basePtexIndicesArg , basePositions basePositionsArg , baseFaces baseFacesArg Create a local refiner sharing the base level, apply adaptive refinement to the given subset of base faces, and construct a patch table and its associated map for the same set of faces Far ConstIndexArray groupFaces baseFaces 0 , int baseFaces . size Far TopologyRefiner localRefiner Far TopologyRefinerFactory Far TopologyDescriptor Create baseRefiner localRefiner - RefineAdaptive patchOptions . GetRefineAdaptiveOptions , groupFaces patchTable Far PatchTableFactory Create localRefiner , patchOptions , groupFaces patchMap new Far PatchMap patchTable patchFaceSize Sdc SchemeTypeTraits GetRegularFaceSize baseRefiner . GetSchemeType Compute the number of refined and local points needed to evaluate the patches, allocate and interpolate. This varies from tutorial_5_1 in that the primvar buffer for the base vertices is separate from the refined vertices and local patch points which must also be accounted for when evaluating the patches. int nBaseVertices localRefiner - GetLevel 0 . GetNumVertices int nRefinedVertices localRefiner - GetNumVerticesTotal - nBaseVertices int nLocalPoints patchTable - GetNumLocalPoints localPositions . resize nRefinedVertices nLocalPoints if nRefinedVertices Far PrimvarRefiner primvarRefiner localRefiner Pos const src basePositions 0 Pos dst localPositions 0 for int level 1 level localRefiner - GetNumLevels level primvarRefiner . Interpolate level , src , dst src dst dst localRefiner - GetLevel level . GetNumVertices if nLocalPoints patchTable - GetLocalPointStencilTable - UpdateValues basePositions 0 , nBaseVertices , localPositions 0 , localPositions nRefinedVertices delete localRefiner PatchGroup PatchGroup delete patchTable delete patchMap void PatchGroup TessellateBaseFace int face , PosVector tessPoints , TriVector tessTris const Tesselate the face with points at the midpoint of the face and at each corner, and triangles connecting the midpoint to each edge. Irregular faces require an aribrary number of corners points, but all are at the origin of the child face of the irregular base face float const quadPoints 5 2 0.5f , 0.5f , 0.0f , 0.0f , 1.0f , 0.0f , 1.0f , 1.0f , 0.0f , 1.0f float const triPoints 4 2 0.5f , 0.5f , 0.0f , 0.0f , 1.0f , 0.0f , 0.0f , 1.0f float const irregPoints 4 2 1.0f , 1.0f , 0.0f , 0.0f Determine the topology of the given base face and the resulting tessellation points and faces to generate int baseFace baseFaces face int faceSize baseRefiner . GetLevel 0 . GetFaceVertices baseFace . size bool faceIsIrregular faceSize patchFaceSize int nTessPoints faceSize 1 int nTessFaces faceSize tessPoints . resize nTessPoints tessTris . resize nTessFaces Compute the mid and corner points -- remember that for an irregular face, we must reference the individual ptex faces for each corner int ptexFace basePtexIndices . GetFaceId baseFace int numBaseVerts int basePositions . size for int i 0 i nTessPoints i Choose the s,t coordinate from the fixed tessellation float const st faceIsIrregular irregPoints i 0 faceSize 4 quadPoints i triPoints i Locate the patch corresponding to the face ptex idx and s,t and evaluate int patchFace ptexFace if faceIsIrregular i 0 patchFace i - 1 Far PatchTable PatchHandle const handle patchMap - FindPatch patchFace , st 0 , st 1 assert handle float pWeights 20 patchTable - EvaluateBasis handle , st 0 , st 1 , pWeights Identify the patch cvs and combine with the evaluated weights -- remember to distinguish cvs in the base level Far ConstIndexArray cvIndices patchTable - GetPatchVertices handle Pos pos tessPoints i pos . Clear for int cv 0 cv cvIndices . size cv int cvIndex cvIndices cv if cvIndex numBaseVerts pos . AddWithWeight basePositions cvIndex , pWeights cv else pos . AddWithWeight localPositions cvIndex - numBaseVerts , pWeights cv Assign triangles connecting the midpoint of the base face to the points computed at the ends of each of its edges for int i 0 i nTessFaces i tessTris i Tri 0 , 1 i , 1 i 1 faceSize Command line arguments parsed to provide run-time options class Args public std string inputObjFile Sdc SchemeType schemeType int geoMultiplier int maxPatchDepth int numPatchGroups bool noTessFlag bool noOutputFlag public Args int argc , char argv inputObjFile , schemeType Sdc SCHEME_CATMARK , geoMultiplier 10 , maxPatchDepth 3 , numPatchGroups 10 , noTessFlag false , noOutputFlag false Parse and assign standard arguments and Obj files ArgOptions args args . Parse argc , argv maxPatchDepth args . GetLevel schemeType ConvertShapeSchemeToSdcType args . GetDefaultScheme const std vector const char objFiles args . GetObjFiles if objFiles . empty for size_t i 1 i objFiles . size i fprintf stderr , Warning .obj file s ignored n , objFiles i inputObjFile std string objFiles 0 Parse remaining arguments specific to this example const std vector const char rargs args . GetRemainingArgs for size_t i 0 i rargs . size i if strcmp rargs i , -groups if i rargs . size numPatchGroups atoi rargs i else if strcmp rargs i , -mult if i rargs . size geoMultiplier atoi rargs i else if strcmp rargs i , -notess noTessFlag true else if strcmp rargs i , -nooutput noOutputFlag true else fprintf stderr , Warning Argument s ignored n , rargs i private Args Load command line arguments and geometry, then divide the mesh into groups of faces from which to create and tessellate patches int main int argc , char argv Args args argc , argv Create or load the base geometry command line arguments allow a .obj file to be specified. In addition to the TopologyRefiner and set of positions for the base vertices, a set of PtexIndices is also required to evaluate patches, so build it here once for use elsewhere std vector Pos basePositions Far TopologyRefiner baseRefinerPtr args . inputObjFile . empty createTopologyRefinerDefault args . geoMultiplier , basePositions createTopologyRefinerFromObj args . inputObjFile , args . schemeType , basePositions assert baseRefinerPtr Far TopologyRefiner baseRefiner baseRefinerPtr Far PtexIndices basePtexIndices baseRefiner Determine the sizes of the patch groups specified -- there will be two sizes that differ by one to account for unequal division int numBaseFaces baseRefiner . GetNumFacesTotal int numPatchGroups args . numPatchGroups if numPatchGroups numBaseFaces numPatchGroups numBaseFaces else if numPatchGroups 1 numPatchGroups 1 int lesserGroupSize numBaseFaces numPatchGroups int numLargerGroups numBaseFaces - numPatchGroups lesserGroupSize Define the options used to construct the patches for each group. Unless suppressed, a tessellation in Obj format will also be printed to standard output, so keep track of the vertex indices. Far PatchTableFactory Options patchOptions args . maxPatchDepth patchOptions . generateVaryingTables false patchOptions . shareEndCapPatchPoints false patchOptions . endCapType Far PatchTableFactory Options ENDCAP_GREGORY_BASIS int objVertCount 0 PosVector tessPoints TriVector tessFaces for int i 0 i numPatchGroups i Initialize a vector with a group of base faces from which to create and evaluate patches Index minFace i lesserGroupSize std min i , numLargerGroups Index maxFace minFace lesserGroupSize i numLargerGroups std vector Far Index baseFaces maxFace - minFace for int face minFace face maxFace face baseFaces face - minFace face Declare a PatchGroup and tessellate its base faces -- generating vertices and faces in Obj format to standard output PatchGroup patchGroup patchOptions , baseRefiner , basePtexIndices , basePositions , baseFaces if args . noTessFlag continue if args . noOutputFlag printf g patchGroup_d n , i for int j 0 j int baseFaces . size j patchGroup . TessellateBaseFace j , tessPoints , tessFaces if args . noOutputFlag int nVerts int tessPoints . size for int k 0 k nVerts k float const p tessPoints k . p printf v f f f n , p 0 , p 1 , p 2 int nTris int tessFaces . size int vBase 1 objVertCount for int k 0 k nTris k int const v tessFaces k . v printf f d d d n , vBase v 0 , vBase v 1 , vBase v 2 objVertCount nVerts delete baseRefinerPtr return EXIT_SUCCESS Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n- \"tags\": \"\",\n- \"loc\": \"far_tutorial_5_2.html\"\n- }, {\n \"title\": \"bfr_tutorial_1_1.cpp\",\n- \"text\": \"bfr_tutorial_1_1.cpp bfr_tutorial_1_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_1_1bfr_tutorial_1_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial illustrates the use of the SurfaceFactory, Surface and Parameterization classes for creating and evaluating the limit surface associated with each base face of a mesh. Following the creation of a connected mesh for a shape using a FarTopologyRefiner, as illustrated in Far tutorials, an instance of a SurfaceFactory is declared to process its faces. Each face of the mesh is evaluated and tessellated independently with a simple triangle fan, with results written out in Obj format for inspection. These classes make it simple to evaluate and tessellate all faces quads, tris or others while supporting the full set of subdivision options. While a triangle fan may be a trivial tessellation and so not very useful later examples using the Tessellation class provide more useful results with the same simplicity. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface faceSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector int outTriangles Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surface for this face -- if valid skipping holes and boundary faces in some rare cases if meshSurfaceFactory . InitVertexSurface faceIndex , faceSurface continue Get the Parameterization of the Surface and use it to identify coordinates for evaluation -- in this case, at the vertices and center of the face to create a fan of triangles Bfr Parameterization faceParam faceSurface . GetParameterization int faceSize faceParam . GetFaceSize int numOutCoords faceSize 1 outCoords . resize numOutCoords 2 for int i 0 i faceSize i faceParam . GetVertexCoord i , outCoords i 2 faceParam . GetCenterCoord outCoords faceSize 2 Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Resize patch point and output arrays int pointSize 3 facePatchPoints . resize faceSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate patch point and output arrays faceSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize faceSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outPos j , outDu j , outDv j Identify the faces of the tessellation, i.e. the triangle fan connecting points at the vertices to the center last point Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. int objVertexIndexOffset objWriter . GetNumVertices outTriangles . resize faceSize 3 int outTriangle outTriangles . data for int i 0 i faceSize i , outTriangle 3 outTriangle 0 objVertexIndexOffset i outTriangle 1 objVertexIndexOffset i 1 faceSize outTriangle 2 objVertexIndexOffset faceSize Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outTriangles , 3 , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"bfr_tutorial_1_1.cpp bfr_tutorial_1_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_1_1bfr_tutorial_1_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial illustrates the use of the SurfaceFactory, Surface and Parameterization classes for creating and evaluating the limit surface associated with each base face of a mesh. Following the creation of a connected mesh for a shape using a FarTopologyRefiner, as illustrated in Far tutorials, an instance of a SurfaceFactory is declared to process its faces. Each face of the mesh is evaluated and tessellated independently with a simple triangle fan, with results written out in Obj format for inspection. These classes make it simple to evaluate and tessellate all faces quads, tris or others while supporting the full set of subdivision options. While a triangle fan may be a trivial tessellation and so not very useful later examples using the Tessellation class provide more useful results with the same simplicity. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface faceSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector int outTriangles Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surface for this face -- if valid skipping holes and boundary faces in some rare cases if meshSurfaceFactory . InitVertexSurface faceIndex , faceSurface continue Get the Parameterization of the Surface and use it to identify coordinates for evaluation -- in this case, at the vertices and center of the face to create a fan of triangles Bfr Parameterization faceParam faceSurface . GetParameterization int faceSize faceParam . GetFaceSize int numOutCoords faceSize 1 outCoords . resize numOutCoords 2 for int i 0 i faceSize i faceParam . GetVertexCoord i , outCoords i 2 faceParam . GetCenterCoord outCoords faceSize 2 Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Resize patch point and output arrays int pointSize 3 facePatchPoints . resize faceSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate patch point and output arrays faceSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize faceSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outPos j , outDu j , outDv j Identify the faces of the tessellation, i.e. the triangle fan connecting points at the vertices to the center last point Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. int objVertexIndexOffset objWriter . GetNumVertices outTriangles . resize faceSize 3 int outTriangle outTriangles . data for int i 0 i faceSize i , outTriangle 3 outTriangle 0 objVertexIndexOffset i outTriangle 1 objVertexIndexOffset i 1 faceSize outTriangle 2 objVertexIndexOffset faceSize Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outTriangles , 3 , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"bfr_tutorial_1_1.html\"\n }, {\n \"title\": \"bfr_tutorial_1_2.cpp\",\n- \"text\": \"bfr_tutorial_1_2.cpp bfr_tutorial_1_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_1_2bfr_tutorial_1_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial builds on the previous tutorial that makes use of the SurfaceFactory and Surface for evaluating the limit surface of faces by using the Tessellation class to determine the points to evaluate and the faces that connect them. The Tessellation class replaces the explicit determination of points and faces for the triangle fan of the previous example. Given a uniform tessellation rate via a command line option, Tessellation returns the set of coordinates to evaluate, and separately returns the faces that connect them. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface faceSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surface for this face -- if valid skipping holes and boundary faces in some rare cases if meshSurfaceFactory . InitVertexSurface faceIndex , faceSurface continue Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern faceSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Resize patch point and output arrays int pointSize 3 facePatchPoints . resize faceSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate patch point and output arrays faceSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize faceSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outPos j , outDu j , outDv j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outFacets , tessFacetSize , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"bfr_tutorial_1_2.cpp bfr_tutorial_1_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsbfrtutorial_1_2bfr_tutorial_1_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial builds on the previous tutorial that makes use of the SurfaceFactory and Surface for evaluating the limit surface of faces by using the Tessellation class to determine the points to evaluate and the faces that connect them. The Tessellation class replaces the explicit determination of points and faces for the triangle fan of the previous example. Given a uniform tessellation rate via a command line option, Tessellation returns the set of coordinates to evaluate, and separately returns the faces that connect them. include opensubdivfartopologyRefiner.h include opensubdivbfrrefinerSurfaceFactory.h include opensubdivbfrsurface.h include opensubdivbfrtessellation.h include vector include string include cstring include cstdio Local headers with support for this tutorial in namespace tutorial include .meshLoader.h include .objWriter.h using namespace OpenSubdiv Simple command line arguments to provide input and run-time options class Args public std string inputObjFile std string outputObjFile Sdc SchemeType schemeType int tessUniformRate bool tessQuadsFlag public Args int argc , char argv inputObjFile , outputObjFile , schemeType Sdc SCHEME_CATMARK , tessUniformRate 5 , tessQuadsFlag false for int i 1 i argc i if strstr argv i , .obj if inputObjFile . empty inputObjFile std string argv i else fprintf stderr , Warning Extra Obj file s ignored n , argv i else if strcmp argv i , -o if i argc outputObjFile std string argv i else if strcmp argv i , -bilinear schemeType Sdc SCHEME_BILINEAR else if strcmp argv i , -catmark schemeType Sdc SCHEME_CATMARK else if strcmp argv i , -loop schemeType Sdc SCHEME_LOOP else if strcmp argv i , -res if i argc tessUniformRate atoi argv i else if strcmp argv i , -quads tessQuadsFlag true else fprintf stderr , Warning Unrecognized argument s ignored n , argv i private Args The main tessellation function given a mesh and vertex positions, tessellate each face -- writing results in Obj format. void tessellateToObj Far TopologyRefiner const meshTopology , std vector float const meshVertexPositions , Args const options Use simpler local type names for the Surface and its factory typedef Bfr RefinerSurfaceFactory SurfaceFactory typedef Bfr Surface float Surface Initialize the SurfaceFactory for the given base mesh very low cost in terms of both time and space and tessellate each face independently i.e. no shared vertices Note that the SurfaceFactory is not thread-safe by default due to use of an internal cache. Creating a separate instance of the SurfaceFactory for each thread is one way to safely parallelize this loop. Another preferred is to assign a thread-safe cache to the single instance. First declare any evaluation options when initializing though none are used in this simple case SurfaceFactory Options surfaceOptions SurfaceFactory meshSurfaceFactory meshTopology , surfaceOptions The Surface to be constructed and evaluated for each face -- as well as the intermediate and output data associated with it -- can be declared in the scope local to each face. But since dynamic memory is involved with these variables, it is preferred to declare them outside that loop to preserve and reuse that dynamic memory. Surface faceSurface std vector float facePatchPoints std vector float outCoords std vector float outPos , outDu , outDv std vector int outFacets Assign Tessellation Options applied for all faces. Tessellations allow the creating of either 3- or 4-sided faces -- both of which are supported here via a command line option int const tessFacetSize 3 options . tessQuadsFlag Bfr Tessellation Options tessOptions tessOptions . SetFacetSize tessFacetSize tessOptions . PreserveQuads options . tessQuadsFlag Process each face, writing the output of each in Obj format tutorial ObjWriter objWriter options . outputObjFile int numFaces meshSurfaceFactory . GetNumFaces for int faceIndex 0 faceIndex numFaces faceIndex Initialize the Surface for this face -- if valid skipping holes and boundary faces in some rare cases if meshSurfaceFactory . InitVertexSurface faceIndex , faceSurface continue Declare a simple uniform Tessellation for the Parameterization of this face and identify coordinates of the points to evaluate Bfr Tessellation tessPattern faceSurface . GetParameterization , options . tessUniformRate , tessOptions int numOutCoords tessPattern . GetNumCoords outCoords . resize numOutCoords 2 tessPattern . GetCoords outCoords . data Prepare the patch points for the Surface, then use them to evaluate output points for all identified coordinates Resize patch point and output arrays int pointSize 3 facePatchPoints . resize faceSurface . GetNumPatchPoints pointSize outPos . resize numOutCoords pointSize outDu . resize numOutCoords pointSize outDv . resize numOutCoords pointSize Populate patch point and output arrays faceSurface . PreparePatchPoints meshVertexPositions . data , pointSize , facePatchPoints . data , pointSize for int i 0 , j 0 i numOutCoords i , j pointSize faceSurface . Evaluate outCoords i 2 , facePatchPoints . data , pointSize , outPos j , outDu j , outDv j Identify the faces of the Tessellation Note the need to offset vertex indices for the output faces -- using the number of vertices generated prior to this face. One of several Tessellation methods to transform the facet indices simply translates all indices by the desired offset. int objVertexIndexOffset objWriter . GetNumVertices int numFacets tessPattern . GetNumFacets outFacets . resize numFacets tessFacetSize tessPattern . GetFacets outFacets . data tessPattern . TransformFacetCoordIndices outFacets . data , objVertexIndexOffset Write the evaluated points and faces connecting them as Obj objWriter . WriteGroupName baseFace_ , faceIndex objWriter . WriteVertexPositions outPos objWriter . WriteVertexNormals outDu , outDv objWriter . WriteFaces outFacets , tessFacetSize , true , false Load command line arguments, specified or default geometry and process int main int argc , char argv Args args argc , argv Far TopologyRefiner meshTopology 0 std vector float meshVtxPositions std vector float meshFVarUVs meshTopology tutorial createTopologyRefiner args . inputObjFile , args . schemeType , meshVtxPositions , meshFVarUVs if meshTopology 0 return EXIT_FAILURE tessellateToObj meshTopology , meshVtxPositions , args delete meshTopology return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"bfr_tutorial_1_2.html\"\n }, {\n \"title\": \"far_tutorial_5_1.cpp\",\n- \"text\": \"far_tutorial_5_1.cpp far_tutorial_5_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_5_1far_tutorial_5_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to interpolate surface limits at arbitrary parametric locations using feature adaptive FarPatchTables. The evaluation of the limit surface at arbitrary locations requires the adaptive isolation of topological features. This process converts the input polygonal control cage into a collection of bi-cubic patches. We can then evaluate the patches at random parametric locations and obtain analytical positions and tangents on the limit surface. The results are dumped into a MEL script that draws streak particle systems that show the tangent and bi-tangent at the random samples locations. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include opensubdivfarpatchTableFactory.h include opensubdivfarpatchMap.h include opensubdivfarptexIndices.h include cassert include cstdio include cstring include cfloat using namespace OpenSubdiv typedef double Real pyramid geometry from catmark_pyramid_crease0.h static int const g_nverts 5 static Real const g_verts 24 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 2.0f , 0.0f , 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 0.0f , static int const g_vertsperface 5 3 , 3 , 3 , 3 , 4 static int const g_nfaces 5 static int const g_faceverts 16 0 , 1 , 2 , 0 , 2 , 3 , 0 , 3 , 4 , 0 , 4 , 1 , 4 , 3 , 2 , 1 static int const g_ncreases 4 static int const g_creaseverts 8 4 , 3 , 3 , 2 , 2 , 1 , 1 , 4 static float const g_creaseweights 4 3.0f , 3.0f , 3.0f , 3.0f Creates a FarTopologyRefiner from the pyramid shape above static Far TopologyRefiner createTopologyRefiner ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex void Clear void 0 point 0 point 1 point 2 0.0f void AddWithWeight Vertex const src , Real weight point 0 weight src . point 0 point 1 weight src . point 1 point 2 weight src . point 2 Real point 3 ------------------------------------------------------------------------------ Limit frame container implementation -- this interface is not strictly required but follows a similar pattern to Vertex. struct LimitFrame void Clear void 0 point 0 point 1 point 2 0.0f deriv1 0 deriv1 1 deriv1 2 0.0f deriv2 0 deriv2 1 deriv2 2 0.0f void AddWithWeight Vertex const src , Real weight , Real d1Weight , Real d2Weight point 0 weight src . point 0 point 1 weight src . point 1 point 2 weight src . point 2 deriv1 0 d1Weight src . point 0 deriv1 1 d1Weight src . point 1 deriv1 2 d1Weight src . point 2 deriv2 0 d2Weight src . point 0 deriv2 1 d2Weight src . point 1 deriv2 2 d2Weight src . point 2 Real point 3 , deriv1 3 , deriv2 3 ------------------------------------------------------------------------------ int main int , char Generate a FarTopologyRefiner see tutorial_1_1 for details. Far TopologyRefiner refiner createTopologyRefiner Patches are constructed from adaptively refined faces, but the processes of constructing the PatchTable and of applying adaptive refinement have historically been separate. Adaptive refinement is applied purely to satisfy the needs of the desired PatchTable, so options associated with adaptive refinement should be derived from those specified for the PatchTable. This is not a strict requirement, but it will avoid problems arising from specifyingcoordinating the two independently especially when dealing with face-varying patches. Initialize options for the PatchTable Choose patches adaptively refined to level 3 since the sharpest crease in the shape is 3.0f in g_creaseweights, and include the inf-sharp crease option just to illustrate the need to syncronize options. int maxPatchLevel 3 Far PatchTableFactory Options patchOptions maxPatchLevel patchOptions . SetPatchPrecision Real patchOptions . useInfSharpPatch true patchOptions . generateVaryingTables false patchOptions . endCapType Far PatchTableFactory Options ENDCAP_GREGORY_BASIS Initialize corresonding options for adaptive refinement Far TopologyRefiner AdaptiveOptions adaptiveOptions maxPatchLevel bool assignAdaptiveOptionsExplicitly false if assignAdaptiveOptionsExplicitly adaptiveOptions . useInfSharpPatch true else Be sure patch options were intialized with the desired max level adaptiveOptions patchOptions . GetRefineAdaptiveOptions assert adaptiveOptions . useInfSharpPatch patchOptions . useInfSharpPatch Apply adaptive refinement and construct the associated PatchTable to evaluate the limit surface refiner - RefineAdaptive adaptiveOptions Far PatchTable const patchTable Far PatchTableFactory Create refiner , patchOptions Compute the total number of points we need to evaluate the PatchTable. Approximations at irregular or extraordinary features require the use of additional points associated with the patches that are referred to as local points i.e. local to the PatchTable. int nRefinerVertices refiner - GetNumVerticesTotal int nLocalPoints patchTable - GetNumLocalPoints Create a buffer to hold the position of the refined verts and local points, then copy the coarse positions at the beginning. std vector Vertex verts nRefinerVertices nLocalPoints std memcpy verts 0 , g_verts , g_nverts 3 sizeof Real Adaptive refinement may result in fewer levels than the max specified. int nRefinedLevels refiner - GetNumLevels Interpolate vertex primvar data they are the control vertices of the limit patches see tutorial_1_1 for details Far PrimvarRefinerReal Real primvarRefiner refiner Vertex src verts 0 for int level 1 level nRefinedLevels level Vertex dst src refiner - GetLevel level -1 . GetNumVertices primvarRefiner . Interpolate level , src , dst src dst Evaluate local points from interpolated vertex primvars. if nLocalPoints patchTable - GetLocalPointStencilTable Real - UpdateValues verts 0 , verts nRefinerVertices Create a FarPatchMap to help locating patches in the table Far PatchMap patchmap patchTable Create a FarPtexIndices to help find indices of ptex faces. Far PtexIndices ptexIndices refiner Generate random samples on each ptex face int nsamplesPerFace 200 , nfaces ptexIndices . GetNumFaces std vector LimitFrame samples nsamplesPerFace nfaces srand static_cast int 2147483647 Real pWeights 20 , dsWeights 20 , dtWeights 20 for int face 0 , count 0 face nfaces face for int sample 0 sample nsamplesPerFace sample , count Real s Real rand Real RAND_MAX , t Real rand Real RAND_MAX Locate the patch corresponding to the face ptex idx and s,t Far PatchTable PatchHandle const handle patchmap . FindPatch face , s , t assert handle Evaluate the patch weights, identify the CVs and compute the limit frame patchTable - EvaluateBasis handle , s , t , pWeights , dsWeights , dtWeights Far ConstIndexArray cvs patchTable - GetPatchVertices handle LimitFrame dst samples count dst . Clear for int cv 0 cv cvs . size cv dst . AddWithWeight verts cvs cv , pWeights cv , dsWeights cv , dtWeights cv Visualization with Maya print a MEL script that generates particles at the location of the limit vertices int nsamples int samples . size printf file -f -new n Output particle positions for the tangent printf particle -n deriv1 for int sample 0 sample nsamples sample Real const pos samples sample . point printf -p f f f n , pos 0 , pos 1 , pos 2 printf n Set per-particle direction using the limit tangent display as Streak printf setAttr deriv1.particleRenderType 6 n printf setAttr deriv1.velocity -type vectorArray d , nsamples for int sample 0 sample nsamples sample Real const tan1 samples sample . deriv1 printf f f f n , tan1 0 , tan1 1 , tan1 2 printf n Output particle positions for the bi-tangent printf particle -n deriv2 for int sample 0 sample nsamples sample Real const pos samples sample . point printf -p f f f n , pos 0 , pos 1 , pos 2 printf n printf setAttr deriv2.particleRenderType 6 n printf setAttr deriv2.velocity -type vectorArray d , nsamples for int sample 0 sample nsamples sample Real const tan2 samples sample . deriv2 printf f f f n , tan2 0 , tan2 1 , tan2 2 printf n Exercise to the reader cross tangent bi-tangent for limit surface normal... Force Maya DAG update to see the result in the viewport printf currentTime -edit currentTime -q n printf select deriv1Shape deriv2Shape n delete refiner delete patchTable return EXIT_SUCCESS ------------------------------------------------------------------------------ static Far TopologyRefiner createTopologyRefiner typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_faceverts desc . numCreases g_ncreases desc . creaseVertexIndexPairs g_creaseverts desc . creaseWeights g_creaseweights Instantiate a FarTopologyRefiner from the descriptor. Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options return refiner Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"far_tutorial_5_1.cpp far_tutorial_5_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_5_1far_tutorial_5_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to interpolate surface limits at arbitrary parametric locations using feature adaptive FarPatchTables. The evaluation of the limit surface at arbitrary locations requires the adaptive isolation of topological features. This process converts the input polygonal control cage into a collection of bi-cubic patches. We can then evaluate the patches at random parametric locations and obtain analytical positions and tangents on the limit surface. The results are dumped into a MEL script that draws streak particle systems that show the tangent and bi-tangent at the random samples locations. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include opensubdivfarpatchTableFactory.h include opensubdivfarpatchMap.h include opensubdivfarptexIndices.h include cassert include cstdio include cstring include cfloat using namespace OpenSubdiv typedef double Real pyramid geometry from catmark_pyramid_crease0.h static int const g_nverts 5 static Real const g_verts 24 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 2.0f , 0.0f , 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 0.0f , static int const g_vertsperface 5 3 , 3 , 3 , 3 , 4 static int const g_nfaces 5 static int const g_faceverts 16 0 , 1 , 2 , 0 , 2 , 3 , 0 , 3 , 4 , 0 , 4 , 1 , 4 , 3 , 2 , 1 static int const g_ncreases 4 static int const g_creaseverts 8 4 , 3 , 3 , 2 , 2 , 1 , 1 , 4 static float const g_creaseweights 4 3.0f , 3.0f , 3.0f , 3.0f Creates a FarTopologyRefiner from the pyramid shape above static Far TopologyRefiner createTopologyRefiner ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex void Clear void 0 point 0 point 1 point 2 0.0f void AddWithWeight Vertex const src , Real weight point 0 weight src . point 0 point 1 weight src . point 1 point 2 weight src . point 2 Real point 3 ------------------------------------------------------------------------------ Limit frame container implementation -- this interface is not strictly required but follows a similar pattern to Vertex. struct LimitFrame void Clear void 0 point 0 point 1 point 2 0.0f deriv1 0 deriv1 1 deriv1 2 0.0f deriv2 0 deriv2 1 deriv2 2 0.0f void AddWithWeight Vertex const src , Real weight , Real d1Weight , Real d2Weight point 0 weight src . point 0 point 1 weight src . point 1 point 2 weight src . point 2 deriv1 0 d1Weight src . point 0 deriv1 1 d1Weight src . point 1 deriv1 2 d1Weight src . point 2 deriv2 0 d2Weight src . point 0 deriv2 1 d2Weight src . point 1 deriv2 2 d2Weight src . point 2 Real point 3 , deriv1 3 , deriv2 3 ------------------------------------------------------------------------------ int main int , char Generate a FarTopologyRefiner see tutorial_1_1 for details. Far TopologyRefiner refiner createTopologyRefiner Patches are constructed from adaptively refined faces, but the processes of constructing the PatchTable and of applying adaptive refinement have historically been separate. Adaptive refinement is applied purely to satisfy the needs of the desired PatchTable, so options associated with adaptive refinement should be derived from those specified for the PatchTable. This is not a strict requirement, but it will avoid problems arising from specifyingcoordinating the two independently especially when dealing with face-varying patches. Initialize options for the PatchTable Choose patches adaptively refined to level 3 since the sharpest crease in the shape is 3.0f in g_creaseweights, and include the inf-sharp crease option just to illustrate the need to syncronize options. int maxPatchLevel 3 Far PatchTableFactory Options patchOptions maxPatchLevel patchOptions . SetPatchPrecision Real patchOptions . useInfSharpPatch true patchOptions . generateVaryingTables false patchOptions . endCapType Far PatchTableFactory Options ENDCAP_GREGORY_BASIS Initialize corresonding options for adaptive refinement Far TopologyRefiner AdaptiveOptions adaptiveOptions maxPatchLevel bool assignAdaptiveOptionsExplicitly false if assignAdaptiveOptionsExplicitly adaptiveOptions . useInfSharpPatch true else Be sure patch options were intialized with the desired max level adaptiveOptions patchOptions . GetRefineAdaptiveOptions assert adaptiveOptions . useInfSharpPatch patchOptions . useInfSharpPatch Apply adaptive refinement and construct the associated PatchTable to evaluate the limit surface refiner - RefineAdaptive adaptiveOptions Far PatchTable const patchTable Far PatchTableFactory Create refiner , patchOptions Compute the total number of points we need to evaluate the PatchTable. Approximations at irregular or extraordinary features require the use of additional points associated with the patches that are referred to as local points i.e. local to the PatchTable. int nRefinerVertices refiner - GetNumVerticesTotal int nLocalPoints patchTable - GetNumLocalPoints Create a buffer to hold the position of the refined verts and local points, then copy the coarse positions at the beginning. std vector Vertex verts nRefinerVertices nLocalPoints std memcpy verts 0 , g_verts , g_nverts 3 sizeof Real Adaptive refinement may result in fewer levels than the max specified. int nRefinedLevels refiner - GetNumLevels Interpolate vertex primvar data they are the control vertices of the limit patches see tutorial_1_1 for details Far PrimvarRefinerReal Real primvarRefiner refiner Vertex src verts 0 for int level 1 level nRefinedLevels level Vertex dst src refiner - GetLevel level -1 . GetNumVertices primvarRefiner . Interpolate level , src , dst src dst Evaluate local points from interpolated vertex primvars. if nLocalPoints patchTable - GetLocalPointStencilTable Real - UpdateValues verts 0 , verts nRefinerVertices Create a FarPatchMap to help locating patches in the table Far PatchMap patchmap patchTable Create a FarPtexIndices to help find indices of ptex faces. Far PtexIndices ptexIndices refiner Generate random samples on each ptex face int nsamplesPerFace 200 , nfaces ptexIndices . GetNumFaces std vector LimitFrame samples nsamplesPerFace nfaces srand static_cast int 2147483647 Real pWeights 20 , dsWeights 20 , dtWeights 20 for int face 0 , count 0 face nfaces face for int sample 0 sample nsamplesPerFace sample , count Real s Real rand Real RAND_MAX , t Real rand Real RAND_MAX Locate the patch corresponding to the face ptex idx and s,t Far PatchTable PatchHandle const handle patchmap . FindPatch face , s , t assert handle Evaluate the patch weights, identify the CVs and compute the limit frame patchTable - EvaluateBasis handle , s , t , pWeights , dsWeights , dtWeights Far ConstIndexArray cvs patchTable - GetPatchVertices handle LimitFrame dst samples count dst . Clear for int cv 0 cv cvs . size cv dst . AddWithWeight verts cvs cv , pWeights cv , dsWeights cv , dtWeights cv Visualization with Maya print a MEL script that generates particles at the location of the limit vertices int nsamples int samples . size printf file -f -new n Output particle positions for the tangent printf particle -n deriv1 for int sample 0 sample nsamples sample Real const pos samples sample . point printf -p f f f n , pos 0 , pos 1 , pos 2 printf n Set per-particle direction using the limit tangent display as Streak printf setAttr deriv1.particleRenderType 6 n printf setAttr deriv1.velocity -type vectorArray d , nsamples for int sample 0 sample nsamples sample Real const tan1 samples sample . deriv1 printf f f f n , tan1 0 , tan1 1 , tan1 2 printf n Output particle positions for the bi-tangent printf particle -n deriv2 for int sample 0 sample nsamples sample Real const pos samples sample . point printf -p f f f n , pos 0 , pos 1 , pos 2 printf n printf setAttr deriv2.particleRenderType 6 n printf setAttr deriv2.velocity -type vectorArray d , nsamples for int sample 0 sample nsamples sample Real const tan2 samples sample . deriv2 printf f f f n , tan2 0 , tan2 1 , tan2 2 printf n Exercise to the reader cross tangent bi-tangent for limit surface normal... Force Maya DAG update to see the result in the viewport printf currentTime -edit currentTime -q n printf select deriv1Shape deriv2Shape n delete refiner delete patchTable return EXIT_SUCCESS ------------------------------------------------------------------------------ static Far TopologyRefiner createTopologyRefiner typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_faceverts desc . numCreases g_ncreases desc . creaseVertexIndexPairs g_creaseverts desc . creaseWeights g_creaseweights Instantiate a FarTopologyRefiner from the descriptor. Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options return refiner Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"far_tutorial_5_1.html\"\n }, {\n \"title\": \"far_tutorial_4_3.cpp\",\n- \"text\": \"far_tutorial_4_3.cpp far_tutorial_4_3.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_4_3far_tutorial_4_3.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to create and manipulate table of cascading stencils. We initialize a FarTopologyRefiner with a cube and apply uniform refinement. We then use a FarStencilTableFactory to generate a stencil table. We set the factory Options to not factorize intermediate levels, thus giving a table of cascading stencils. We then apply the stencils to the vertex position primvar data, and insert a hierarchical edit at level 1. This edit is smoothed by the application of the subsequent stencil cascades. The results are dumped into an OBJ file that shows the intermediate levels of refinement of the original cube. include opensubdivfartopologyDescriptor.h include opensubdivfarstencilTable.h include opensubdivfarstencilTableFactory.h include cstdio include cstring ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z float const GetPosition const return _position float GetPosition return _position private float _position 3 ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 24 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv static Far TopologyRefiner createTopologyRefiner ------------------------------------------------------------------------------ int main int , char Generate a FarTopologyRefiner see tutorial_1_1 for details. Far TopologyRefiner refiner createTopologyRefiner Uniformly refine the topology up to maxlevel. int maxlevel 4 refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Use the FarStencilTable factory to create cascading stencil table note we want stencils for each refinement level cascade mode is achieved by setting factorizeIntermediateLevels to false Far StencilTableFactory Options options options . generateIntermediateLevels true options . factorizeIntermediateLevels false options . generateOffsets true Far StencilTable const stencilTable Far StencilTableFactory Create refiner , options std vector Vertex vertexBuffer refiner - GetNumVerticesTotal - g_nverts Vertex destVerts vertexBuffer 0 int start 0 , end 0 stencil batches for each level of subdivision for int level 0 level maxlevel level int nverts refiner - GetLevel level 1 . GetNumVertices Vertex const srcVerts reinterpret_cast Vertex g_verts if level 0 srcVerts vertexBuffer start start end end nverts stencilTable - UpdateValues srcVerts , destVerts , start , end apply 2 hierarchical edits on level 1 vertices if level 1 float pos destVerts start 5 . GetPosition pos 1 0.5f pos destVerts start 20 . GetPosition pos 0 0.25f Output OBJ of the highest level refined ----------- Vertex verts vertexBuffer 0 Print vertex positions for int level 1 , firstvert 0 level maxlevel level Far TopologyLevel const refLevel refiner - GetLevel level printf g level_d n , level int nverts refLevel . GetNumVertices for int vert 0 vert nverts vert float const pos verts vert . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 verts nverts Print faces for int face 0 face refLevel . GetNumFaces face Far ConstIndexArray fverts refLevel . GetFaceVertices face all refined Catmark faces should be quads assert fverts . size 4 printf f for int vert 0 vert fverts . size vert printf d , fverts vert firstvert 1 OBJ uses 1-based arrays... printf n firstvert nverts delete refiner delete stencilTable return EXIT_SUCCESS ------------------------------------------------------------------------------ static Far TopologyRefiner createTopologyRefiner Populate a topology descriptor with our raw data. typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor. return Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"far_tutorial_4_3.cpp far_tutorial_4_3.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_4_3far_tutorial_4_3.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to create and manipulate table of cascading stencils. We initialize a FarTopologyRefiner with a cube and apply uniform refinement. We then use a FarStencilTableFactory to generate a stencil table. We set the factory Options to not factorize intermediate levels, thus giving a table of cascading stencils. We then apply the stencils to the vertex position primvar data, and insert a hierarchical edit at level 1. This edit is smoothed by the application of the subsequent stencil cascades. The results are dumped into an OBJ file that shows the intermediate levels of refinement of the original cube. include opensubdivfartopologyDescriptor.h include opensubdivfarstencilTable.h include opensubdivfarstencilTableFactory.h include cstdio include cstring ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z float const GetPosition const return _position float GetPosition return _position private float _position 3 ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 24 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv static Far TopologyRefiner createTopologyRefiner ------------------------------------------------------------------------------ int main int , char Generate a FarTopologyRefiner see tutorial_1_1 for details. Far TopologyRefiner refiner createTopologyRefiner Uniformly refine the topology up to maxlevel. int maxlevel 4 refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Use the FarStencilTable factory to create cascading stencil table note we want stencils for each refinement level cascade mode is achieved by setting factorizeIntermediateLevels to false Far StencilTableFactory Options options options . generateIntermediateLevels true options . factorizeIntermediateLevels false options . generateOffsets true Far StencilTable const stencilTable Far StencilTableFactory Create refiner , options std vector Vertex vertexBuffer refiner - GetNumVerticesTotal - g_nverts Vertex destVerts vertexBuffer 0 int start 0 , end 0 stencil batches for each level of subdivision for int level 0 level maxlevel level int nverts refiner - GetLevel level 1 . GetNumVertices Vertex const srcVerts reinterpret_cast Vertex g_verts if level 0 srcVerts vertexBuffer start start end end nverts stencilTable - UpdateValues srcVerts , destVerts , start , end apply 2 hierarchical edits on level 1 vertices if level 1 float pos destVerts start 5 . GetPosition pos 1 0.5f pos destVerts start 20 . GetPosition pos 0 0.25f Output OBJ of the highest level refined ----------- Vertex verts vertexBuffer 0 Print vertex positions for int level 1 , firstvert 0 level maxlevel level Far TopologyLevel const refLevel refiner - GetLevel level printf g level_d n , level int nverts refLevel . GetNumVertices for int vert 0 vert nverts vert float const pos verts vert . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 verts nverts Print faces for int face 0 face refLevel . GetNumFaces face Far ConstIndexArray fverts refLevel . GetFaceVertices face all refined Catmark faces should be quads assert fverts . size 4 printf f for int vert 0 vert fverts . size vert printf d , fverts vert firstvert 1 OBJ uses 1-based arrays... printf n firstvert nverts delete refiner delete stencilTable return EXIT_SUCCESS ------------------------------------------------------------------------------ static Far TopologyRefiner createTopologyRefiner Populate a topology descriptor with our raw data. typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor. return Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"far_tutorial_4_3.html\"\n }, {\n- \"title\": \"far_tutorial_5_3.cpp\",\n- \"text\": \"far_tutorial_5_3.cpp far_tutorial_5_3.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_5_3far_tutorial_5_3.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to use a FarLimitStenciTable to repeatedly and efficiently evaluate a set of points and optionally derivatives on the limit surface. A LimitStencilTable derives from StencilTable but is specialized to factor the evaluation of limit positions and derivatives into stencils. This allows a set of limit properties to be efficiently recomputed in response to changes to the vertices of the base mesh. Constructing the different kinds of StencilTables can have a high cost, so whether that cost is worth it will depend on your usage e.g. if points are only computed once, using stencil tables is typically not worth the added cost. Any points on the limit surface can be identified for evaluation. In this example we create a crude tessellation similar to tutorial_5_2. The midpoint of each face and points near the corners of the face are evaluated and a triangle fan connects them. include ......regressioncommonarg_utils.h include ......regressioncommonfar_utils.h include opensubdivfartopologyDescriptor.h include opensubdivfarpatchTableFactory.h include opensubdivfarstencilTableFactory.h include opensubdivfarptexIndices.h include cassert include cstdio include cstring include fstream include sstream using namespace OpenSubdiv using Far Index Global utilities in this namespace are not relevant to the tutorial. They simply serve to construct some default geometry to be processed in the form of a TopologyRefiner and vector of vertex positions. namespace Simple structs for x,y,z position and a 3-tuple for the set of vertices of a triangle struct Pos Pos Pos float x , float y , float z p 0 x , p 1 y , p 2 z Pos operator Pos const op const return Pos p 0 op . p 0 , p 1 op . p 1 , p 2 op . p 2 Clear and AddWithWeight required for interpolation void Clear void 0 p 0 p 1 p 2 0.0f void AddWithWeight Pos const src , float weight p 0 weight src . p 0 p 1 weight src . p 1 p 2 weight src . p 2 float p 3 typedef std vector Pos PosVector struct Tri Tri Tri int a , int b , int c v 0 a , v 1 b , v 2 c int v 3 typedef std vector Tri TriVector Functions to populate the topology and geometry arrays a simple shape whose positions may be transformed void createCube std vector int vertsPerFace , std vector Index faceVertsPerFace , std vector Pos positionsPerVert Local topology and position of a cube centered at origin static float const cubePositions 8 3 -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f static int const cubeFaceVerts 6 4 0 , 3 , 2 , 1 , 4 , 5 , 6 , 7 , 0 , 4 , 7 , 3 , 1 , 2 , 6 , 5 , 0 , 1 , 5 , 4 , 3 , 7 , 6 , 2 Initialize verts-per-face and face-vertices for each face vertsPerFace . resize 6 faceVertsPerFace . resize 24 for int i 0 i 6 i vertsPerFace i 4 for int j 0 j 4 j faceVertsPerFace i 4 j cubeFaceVerts i j Initialize vertex positions positionsPerVert . resize 8 for int i 0 i 8 i float const p cubePositions i positionsPerVert i Pos p 0 , p 1 , p 2 Create a TopologyRefiner from default geometry created above Far TopologyRefiner createTopologyRefinerDefault PosVector posVector std vector int topVertsPerFace std vector Index topFaceVerts createCube topVertsPerFace , topFaceVerts , posVector typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_AND_CORNER Descriptor desc desc . numVertices int posVector . size desc . numFaces int topVertsPerFace . size desc . numVertsPerFace topVertsPerFace 0 desc . vertIndicesPerFace topFaceVerts 0 Instantiate a FarTopologyRefiner from the descriptor. Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options assert refiner return refiner Create a TopologyRefiner from a specified Obj file geometry created internally Far TopologyRefiner createTopologyRefinerFromObj std string const objFileName , Sdc SchemeType schemeType , PosVector posVector const char filename objFileName . c_str const Shape shape 0 std ifstream ifs filename if ifs std stringstream ss ss ifs . rdbuf ifs . close std string shapeString ss . str shape Shape parseObj shapeString . c_str , ConvertSdcTypeToShapeScheme schemeType , false if shape 0 fprintf stderr , Error Cannot create Shape from .obj file s n , filename return 0 else fprintf stderr , Error Cannot open .obj file s n , filename return 0 Sdc SchemeType sdcType GetSdcType shape Sdc Options sdcOptions GetSdcOptions shape Far TopologyRefiner refiner Far TopologyRefinerFactory Shape Create shape , Far TopologyRefinerFactory Shape Options sdcType , sdcOptions if refiner 0 fprintf stderr , Error Unable to construct TopologyRefiner from .obj file s n , filename return 0 int numVertices refiner - GetNumVerticesTotal posVector . resize numVertices std memcpy posVector 0 . p 0 , shape - verts 0 , numVertices 3 sizeof float delete shape return refiner Simple function to export an Obj file for the limit points -- which provides a simple tessllation similar to tutorial_5_2. int writeToObj Far TopologyLevel const baseLevel , std vector Pos const vertexPositions , int nextObjVertexIndex for size_t i 0 i vertexPositions . size i float const p vertexPositions i . p printf v f f f n , p 0 , p 1 , p 2 Connect the sequences of limit points center followed by corners into triangle fans for each base face for int i 0 i baseLevel . GetNumFaces i int faceSize baseLevel . GetFaceVertices i . size int vCenter nextObjVertexIndex 1 int vCorner vCenter 1 for int k 0 k faceSize k printf f d d d n , vCenter , vCorner k , vCorner k 1 faceSize nextObjVertexIndex faceSize 1 return nextObjVertexIndex end namespace Command line arguments parsed to provide run-time options class Args public std string inputObjFile Sdc SchemeType schemeType int maxPatchDepth int numPoses Pos poseOffset bool deriv1Flag bool noPatchesFlag bool noOutputFlag public Args int argc , char argv inputObjFile , schemeType Sdc SCHEME_CATMARK , maxPatchDepth 3 , numPoses 0 , poseOffset 1.0f , 0.0f , 0.0f , deriv1Flag false , noPatchesFlag false , noOutputFlag false Parse and assign standard arguments and Obj files ArgOptions args args . Parse argc , argv maxPatchDepth args . GetLevel schemeType ConvertShapeSchemeToSdcType args . GetDefaultScheme const std vector const char objFiles args . GetObjFiles if objFiles . empty for size_t i 1 i objFiles . size i fprintf stderr , Warning .obj file s ignored n , objFiles i inputObjFile std string objFiles 0 Parse remaining arguments specific to this example const std vector const char rargs args . GetRemainingArgs for size_t i 0 i rargs . size i if strcmp rargs i , -d1 deriv1Flag true else if strcmp rargs i , -nopatches noPatchesFlag true else if strcmp rargs i , -poses if i rargs . size numPoses atoi rargs i else if strcmp rargs i , -offset if i rargs . size poseOffset . p 0 float atof rargs i if i rargs . size poseOffset . p 1 float atof rargs i if i rargs . size poseOffset . p 2 float atof rargs i else if strcmp rargs i , -nooutput noOutputFlag true else fprintf stderr , Warning Argument s ignored n , rargs i private Args Assemble the set of locations for the limit points. The resulting vector of LocationArrays can contain arbitrary locations on the limit surface -- with multiple locations for the same patch grouped into a single array. In this case, for each base face, coordinates for the center and its corners are specified -- from which we will construct a triangle fan providing a crude tessellation similar to tutorial_5_2. typedef Far LimitStencilTableFactory LocationArray LocationArray int assembleLimitPointLocations Far TopologyRefiner const refiner , std vector LocationArray locations Coordinates for the center of the face and its corners slightly inset. Unlike most of the public interface for patches, the LocationArray refers to parameteric coordinates as s,t, so that convention will be followed here. Note that the s,t coordinates in a LocationArray are referred to by reference. The memory holding these s,t values must persist while the LimitStencilTable is constructed -- the arrays here are declared as static for that purpose. static float const quadSCoords 5 0.5f , 0.05f , 0.95f , 0.95f , 0.05f static float const quadTCoords 5 0.5f , 0.05f , 0.05f , 0.95f , 0.95f static float const triSCoords 4 0.33f , 0.05f , 0.95f , 0.05f static float const triTCoords 4 0.33f , 0.05f , 0.00f , 0.95f static float const irregSCoords 2 1.0f , 0.05f static float const irregTCoords 2 1.0f , 0.05f Since these are references to patches to be evaluated, we require use of the Ptex indices to identify the top-most parameterized patch, which is essential to dealing with non-quad faces in the case of Catmark. Far TopologyLevel const baseLevel refiner . GetLevel 0 Far PtexIndices basePtexIndices refiner int regFaceSize Sdc SchemeTypeTraits GetRegularFaceSize refiner . GetSchemeType For each base face, simply refer to the s,t arrays for regular quad and triangular patches with a single LocationArray. Otherwise, for irregular faces, the corners of the face come from different patches and so must be referenced in separate LocationArrays. locations . clear int numLimitPoints 0 for int i 0 i baseLevel . GetNumFaces i int baseFaceSize baseLevel . GetFaceVertices i . size int basePtexId basePtexIndices . GetFaceId i bool faceIsRegular baseFaceSize regFaceSize if faceIsRegular All coordinates are on the same top-level patch LocationArray loc loc . ptexIdx basePtexId loc . numLocations baseFaceSize 1 if baseFaceSize 4 loc . s quadSCoords loc . t quadTCoords else loc . s triSCoords loc . t triTCoords locations . push_back loc else Center coordinate is on the first sub-patch while those on near the corners are on each successive sub-patch LocationArray loc loc . numLocations 1 for int j 0 j baseFaceSize j bool isPerimeter j 0 loc . ptexIdx basePtexId isPerimeter j -1 0 loc . s irregSCoords isPerimeter loc . t irregTCoords isPerimeter locations . push_back loc numLimitPoints baseFaceSize 1 return numLimitPoints Load command line arguments and geometry, build the LimitStencilTable for a set of points on the limit surface and compute those points for several orientations of the mesh int main int argc , char argv Args args argc , argv Create or load the base geometry command line arguments allow a .obj file to be specified, providing a TopologyRefiner and a set of base vertex positions to work with std vector Pos basePositions Far TopologyRefiner refinerPtr args . inputObjFile . empty createTopologyRefinerDefault basePositions createTopologyRefinerFromObj args . inputObjFile , args . schemeType , basePositions assert refinerPtr Far TopologyRefiner refiner refinerPtr Far TopologyLevel const baseLevel refiner . GetLevel 0 Use of LimitStencilTable requires either explicit or implicit use of a PatchTable. A PatchTable is not required to construct a LimitStencilTable -- one will be constructed internally for use and discarded -- but explicit construction is recommended to control the many legacy options for PatchTable, rather than relying on internal defaults. Adaptive refinement is required in both cases to indicate the accuracy of the patches. Note that if a TopologyRefiner and PatchTable are not used for any other purpose than computing the limit points, that specifying the subset of faces containing those limit points in the adaptive refinement and PatchTable construction can avoid unnecessary overhead. Far PatchTable patchTablePtr 0 if args . noPatchesFlag refiner . RefineAdaptive Far TopologyRefiner AdaptiveOptions args . maxPatchDepth else Far PatchTableFactory Options patchOptions args . maxPatchDepth patchOptions . useInfSharpPatch true patchOptions . generateLegacySharpCornerPatches false patchOptions . generateVaryingTables false patchOptions . generateFVarTables false patchOptions . endCapType Far PatchTableFactory Options ENDCAP_GREGORY_BASIS refiner . RefineAdaptive patchOptions . GetRefineAdaptiveOptions patchTablePtr Far PatchTableFactory Create refiner , patchOptions assert patchTablePtr Assemble the set of locations for the limit points. For each base face, coordinates for the center and its corners are specified -- from which we will construct a triangle fan providing a crude tessellation similar to tutorial_5_2. std vector LocationArray locations int numLimitPoints assembleLimitPointLocations refiner , locations Construct a LimitStencilTable from the refiner, patch table optional and the collection of limit point locations. Stencils can optionally be created for computing dervatives -- the default is to compute 1st derivative stencils, so be sure to disable that if not necessary Far LimitStencilTableFactory Options limitOptions limitOptions . generate1stDerivatives args . deriv1Flag Far LimitStencilTable const limitStencilTablePtr Far LimitStencilTableFactory Create refiner , locations , 0 , optional StencilTable for the refined points patchTablePtr , optional PatchTable limitOptions assert limitStencilTablePtr Far LimitStencilTable const limitStencilTable limitStencilTablePtr Apply the constructed LimitStencilTable to compute limit positions from the base level vertex positions. This is trivial if computing all positions in one invokation. The UpdateValues method and those for derivatives are overloaded to optionally accept a subrange of indices to distribute the computation std vector Pos limitPositions numLimitPoints limitStencilTable . UpdateValues basePositions , limitPositions Call with the optional subrange limitStencilTable . UpdateValues basePositions , limitPositions , 0 , numLimitPoints 2 limitStencilTable . UpdateValues basePositions , limitPositions , numLimitPoints 2 1 , numLimitPoints Write vertices and faces in Obj format for the original limit points int objVertCount 0 if args . noOutputFlag printf g base_mesh n objVertCount writeToObj baseLevel , limitPositions , objVertCount Recompute the limit points and output faces for different poses of the original mesh -- in this case simply translated. Also optionally compute 1st derivatives though they are not used here std vector Pos posePositions basePositions std vector Pos limitDu args . deriv1Flag numLimitPoints 0 std vector Pos limitDv args . deriv1Flag numLimitPoints 0 for int i 0 i args . numPoses i Trivially transform the base vertex positions and re-compute for size_t j 0 j basePositions . size j posePositions j posePositions j args . poseOffset limitStencilTable . UpdateValues posePositions , limitPositions if args . deriv1Flag limitStencilTable . UpdateDerivs posePositions , limitDu , limitDv if args . noOutputFlag printf n g pose_d n , i objVertCount writeToObj baseLevel , limitPositions , objVertCount delete refinerPtr delete patchTablePtr delete limitStencilTablePtr return EXIT_SUCCESS Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n- \"tags\": \"\",\n- \"loc\": \"far_tutorial_5_3.html\"\n- }, {\n- \"title\": \"far_tutorial_3_1.cpp\",\n- \"text\": \"far_tutorial_3_1.cpp far_tutorial_3_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_3_1far_tutorial_3_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to interface a high-level topology representation with Far for better efficiency. In tutorial 0, we showed how to instantiate topology from a simple face-vertex list. Here we will show how to take advantage of more complex data structures. Many client applications that manipulate geometry use advanced data structures such as half-edge, quad-edge or winged-edge in order to represent complex topological relationships beyond the usual face-vertex lists. We can take advantage of this information. Far provides an advanced interface that allows such a client application to communicate advanced component relationships directly and avoid having Far rebuilding them redundantly. include opensubdivfartopologyRefinerFactory.h include opensubdivfarprimvarRefiner.h include cstdio ------------------------------------------------------------------------------ using namespace OpenSubdiv ------------------------------------------------------------------------------ For this tutorial, we provide the complete topological representation of a simple pyramid. In our case, we store it as a simple sequence of integers, with the understanding that client-code would provide a fully implemented data-structure such as quad-edges or winged-edges. Pyramid geometry from catmark_pyramid.h - extended for this tutorial static int g_nverts 5 , g_nedges 8 , g_nfaces 5 vertex positions static float g_verts 5 3 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 2.0f , 0.0f , 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 0.0f number of vertices in each face static int g_facenverts 5 3 , 3 , 3 , 3 , 4 index of face vertices static int g_faceverts 16 0 , 1 , 2 , 0 , 2 , 3 , 0 , 3 , 4 , 0 , 4 , 1 , 4 , 3 , 2 , 1 index of edge vertices 2 per edge static int g_edgeverts 16 0 , 1 , 1 , 2 , 2 , 0 , 2 , 3 , 3 , 0 , 3 , 4 , 4 , 0 , 4 , 1 index of face edges static int g_faceedges 16 0 , 1 , 2 , 2 , 3 , 4 , 4 , 5 , 6 , 6 , 7 , 0 , 5 , 3 , 1 , 7 number of faces adjacent to each edge static int g_edgenfaces 8 2 , 2 , 2 , 2 , 2 , 2 , 2 , 2 index of faces incident to a given edge static int g_edgefaces 16 0 , 3 , 0 , 4 , 0 , 1 , 1 , 4 , 1 , 2 , 2 , 4 , 2 , 3 , 3 , 4 number of faces incident to each vertex static int g_vertexnfaces 5 4 , 3 , 3 , 3 , 3 index of faces incident to each vertex static int g_vertexfaces 25 0 , 1 , 2 , 3 , 0 , 3 , 4 , 0 , 4 , 1 , 1 , 4 , 2 , 2 , 4 , 3 number of edges incident to each vertex static int g_vertexnedges 5 4 , 3 , 3 , 3 , 3 index of edges incident to each vertex static int g_vertexedges 25 0 , 2 , 4 , 6 , 1 , 0 , 7 , 2 , 1 , 3 , 4 , 3 , 5 , 6 , 5 , 7 Edge crease sharpness static float g_edgeCreases 8 0.0f , 2.5f , 0.0f , 2.5f , 0.0f , 2.5f , 0.0f , 2.5f ------------------------------------------------------------------------------ Because existing client-code may not provide an exact match for the topological queries required by Fars interface, we can provide a converter class. This can be particularly useful for instance if the client data-structure requires additional relationships to be mapped. For instance, half-edge representations do not store unique edge indices and it can be difficult to traverse edges or faces adjacent to a given vertex. Using an intermediate wrapper class allows us to leverage existing relationships information from a mesh, and generate the missing components temporarily. For a practical example, you can look at the file hbr_to_vtr.h in the same tutorial directory. This example implements a OsdHbrConverter class as a way of interfacing PRmans half-edge representation to Far. struct Converter public Sdc SchemeType GetType const return Sdc SCHEME_CATMARK Sdc Options GetOptions const Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY return options int GetNumFaces const return g_nfaces int GetNumEdges const return g_nedges int GetNumVertices const return g_nverts Face relationships int GetNumFaceVerts int face const return g_facenverts face int const GetFaceVerts int face const return g_faceverts getCompOffset g_facenverts , face int const GetFaceEdges int face const return g_faceedges getCompOffset g_facenverts , face Edge relationships int const GetEdgeVertices int edge const return g_edgeverts edge 2 int GetNumEdgeFaces int edge const return g_edgenfaces edge int const GetEdgeFaces int edge const return g_edgefaces getCompOffset g_edgenfaces , edge Vertex relationships int GetNumVertexEdges int vert const return g_vertexnedges vert int const GetVertexEdges int vert const return g_vertexedges getCompOffset g_vertexnedges , vert int GetNumVertexFaces int vert const return g_vertexnfaces vert int const GetVertexFaces int vert const return g_vertexfaces getCompOffset g_vertexnfaces , vert private int getCompOffset int const comps , int comp const int ofs 0 for int i 0 i comp i ofs comps i return ofs ------------------------------------------------------------------------------ namespace OpenSubdiv namespace OPENSUBDIV_VERSION namespace Far template bool TopologyRefinerFactory Converter resizeComponentTopology TopologyRefiner refiner , Converter const conv Faces and face-verts int nfaces conv . GetNumFaces setNumBaseFaces refiner , nfaces for int face 0 face nfaces face int nv conv . GetNumFaceVerts face setNumBaseFaceVertices refiner , face , nv Edges and edge-faces int nedges conv . GetNumEdges setNumBaseEdges refiner , nedges for int edge 0 edge nedges edge int nf conv . GetNumEdgeFaces edge setNumBaseEdgeFaces refiner , edge , nf Vertices and vert-faces and vert-edges int nverts conv . GetNumVertices setNumBaseVertices refiner , nverts for int vert 0 vert nverts vert int ne conv . GetNumVertexEdges vert , nf conv . GetNumVertexFaces vert setNumBaseVertexEdges refiner , vert , ne setNumBaseVertexFaces refiner , vert , nf return true template bool TopologyRefinerFactory Converter assignComponentTopology TopologyRefiner refiner , Converter const conv using Far IndexArray Face relations int nfaces conv . GetNumFaces for int face 0 face nfaces face IndexArray dstFaceVerts getBaseFaceVertices refiner , face IndexArray dstFaceEdges getBaseFaceEdges refiner , face int const faceverts conv . GetFaceVerts face int const faceedges conv . GetFaceEdges face for int vert 0 vert conv . GetNumFaceVerts face vert dstFaceVerts vert faceverts vert dstFaceEdges vert faceedges vert Edge relations Note if your representation is unable to provide edge relationships ex half-edges, you can comment out this section and Far will automatically generate the missing information. int nedges conv . GetNumEdges for int edge 0 edge nedges edge Edge-vertices IndexArray dstEdgeVerts getBaseEdgeVertices refiner , edge dstEdgeVerts 0 conv . GetEdgeVertices edge 0 dstEdgeVerts 1 conv . GetEdgeVertices edge 1 Edge-faces IndexArray dstEdgeFaces getBaseEdgeFaces refiner , edge for int face 0 face conv . GetNumEdgeFaces face face dstEdgeFaces face conv . GetEdgeFaces edge face Vertex relations int nverts conv . GetNumVertices for int vert 0 vert nverts vert Vert-Faces IndexArray vertFaces getBaseVertexFaces refiner , vert LocalIndexArray vertInFaceIndices getBaseVertexFaceLocalIndicesrefiner, vert for int face 0 face conv . GetNumVertexFaces vert face vertFaces face conv . GetVertexFaces vert face Vert-Edges IndexArray vertEdges getBaseVertexEdges refiner , vert LocalIndexArray vertInEdgeIndices getBaseVertexEdgeLocalIndicesrefiner, vert for int edge 0 edge conv . GetNumVertexEdges vert edge vertEdges edge conv . GetVertexEdges vert edge populateBaseLocalIndices refiner return true template bool TopologyRefinerFactory Converter assignComponentTags TopologyRefiner refiner , Converter const conv arbitrarily sharpen the 4 bottom edges of the pyramid to 2.5f for int edge 0 edge conv . GetNumEdges edge setBaseEdgeSharpness refiner , edge , g_edgeCreases edge return true ifdef _MSC_VER template void TopologyRefinerFactory Converter reportInvalidTopology TopologyError errCode , char const msg , Converter const mesh Optional topology validation error reporting This method is called whenever the factory encounters topology validation errors. By default, nothing is reported Warning msg template bool TopologyRefinerFactory Converter assignFaceVaryingTopology TopologyRefiner refiner , Converter const conv Because of the way MSVC specializes templated functions, we had to remove the default stubs in FarTopologyRefinerFactory. In this example, no face-varying data is being added, but we still need to implement a template specialization or MSVC linker fails. return true endif namespace Far namespace OPENSUBDIV_VERSION namespace OpenSubdiv ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 ------------------------------------------------------------------------------ int main int , char Converter conv Far TopologyRefiner refiner Far TopologyRefinerFactory Converter Create conv , Far TopologyRefinerFactory Converter Options conv . GetType , conv . GetOptions int maxlevel 5 Uniformly refine the topology up to maxlevel refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Allocate a buffer for vertex primvar data. The buffer length is set to be the sum of all children vertices up to the highest level of refinement. std vector Vertex vbuffer refiner - GetNumVerticesTotal Vertex verts vbuffer 0 Initialize coarse mesh positions int nCoarseVerts g_nverts for int i 0 i nCoarseVerts i verts i . SetPosition g_verts i 0 , g_verts i 1 , g_verts i 2 Interpolate vertex primvar data Far PrimvarRefiner primvarRefiner refiner Vertex src verts for int level 1 level maxlevel level Vertex dst src refiner - GetLevel level -1 . GetNumVertices primvarRefiner . Interpolate level , src , dst src dst Output OBJ of the highest level refined ----------- Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int nverts refLastLevel . GetNumVertices int nfaces refLastLevel . GetNumFaces Print vertex positions int firstOfLastVerts refiner - GetNumVerticesTotal - nverts for int vert 0 vert nverts vert float const pos verts firstOfLastVerts vert . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 Print faces for int face 0 face nfaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face all refined Catmark faces should be quads assert fverts . size 4 printf f for int vert 0 vert fverts . size vert printf d , fverts vert 1 OBJ uses 1-based arrays... printf n delete refiner return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n- \"tags\": \"\",\n- \"loc\": \"far_tutorial_3_1.html\"\n- }, {\n \"title\": \"far_tutorial_4_2.cpp\",\n- \"text\": \"far_tutorial_4_2.cpp far_tutorial_4_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_4_2far_tutorial_4_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to create and manipulate both vertex and varying FarStencilTable to interpolate 2 primvar data buffers vertex positions and vertex colors. include opensubdivfartopologyDescriptor.h include opensubdivfarstencilTable.h include opensubdivfarstencilTableFactory.h include cstdio include cstring ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _data 0 src . _data 0 _data 1 src . _data 1 _data 2 src . _data 2 void Clear void 0 _data 0 _data 1 _data 2 0.0f void AddWithWeight Vertex const src , float weight _data 0 weight src . _data 0 _data 1 weight src . _data 1 _data 2 weight src . _data 2 Public interface ------------------------------------ float const GetData const return _data private float _data 3 ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 24 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f Per-vertex RGB color data static float g_colors 24 1.0f , 0.0f , 0.5f , 0.0f , 1.0f , 0.0f , 0.0f , 0.0f , 1.0f , 1.0f , 1.0f , 1.0f , 1.0f , 1.0f , 0.0f , 0.0f , 1.0f , 1.0f , 1.0f , 0.0f , 1.0f , 0.0f , 0.0f , 0.0f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv static Far TopologyRefiner createTopologyRefiner ------------------------------------------------------------------------------ int main int , char Generate a FarTopologyRefiner see tutorial_1_1 for details. Far TopologyRefiner refiner createTopologyRefiner Uniformly refine the topology up to maxlevel. int maxlevel 4 refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel int nverts refiner - GetLevel maxlevel . GetNumVertices Use the FarStencilTable factory to create discrete stencil table Far StencilTableFactory Options options options . generateIntermediateLevels false only the highest refinement level. options . generateOffsets true Vertex primvar data Create stencils table for vertex interpolation options . interpolationMode Far StencilTableFactory INTERPOLATE_VERTEX Far StencilTable const vertexStencils Far StencilTableFactory Create refiner , options assert nverts vertexStencils - GetNumStencils Allocate vertex primvar buffer 1 stencil for each vertex std vector Vertex vertexBuffer vertexStencils - GetNumStencils Use the cube vertex positions as vertex primvar data Vertex vertexCVs reinterpret_cast Vertex g_verts Varying primvar data Create stencils table for varying interpolation options . interpolationMode Far StencilTableFactory INTERPOLATE_VARYING Far StencilTable const varyingStencils Far StencilTableFactory Create refiner , options assert nverts varyingStencils - GetNumStencils Allocate varying primvar buffer 1 stencil for each vertex std vector Vertex varyingBuffer varyingStencils - GetNumStencils Use per-vertex array of RGB colors as varying primvar data Vertex varyingCVs reinterpret_cast Vertex g_colors delete refiner Apply stencils in frame loop This section would be applied every frame after control vertices have been moved. Apply stencils on the control vertex data to update the primvar data of the refined vertices. vertexStencils - UpdateValues vertexCVs , vertexBuffer 0 varyingStencils - UpdateValues varyingCVs , varyingBuffer 0 Visualization with Maya print a MEL script that generates particles at the location of the refined vertices printf particle for int vert 0 vert int nverts vert float const pos vertexBuffer vert . GetData printf -p f f f n , pos 0 , pos 1 , pos 2 printf -c 1 n Set particle point size 20 -- very large printf addAttr -is true -ln pointSize -at long -dv 20 particleShape1 n Add per-particle color attribute rgbPP printf addAttr -ln rgbPP -dt vectorArray particleShape1 n Set per-particle color values from our varying primvar data printf setAttr particleShape1.rgbPP -type vectorArray d , nverts for int vert 0 vert nverts vert float const color varyingBuffer vert . GetData printf f f f n , color 0 , color 1 , color 2 printf n delete vertexStencils delete varyingStencils return EXIT_SUCCESS ------------------------------------------------------------------------------ static Far TopologyRefiner createTopologyRefiner Populate a topology descriptor with our raw data. typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor. Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options return refiner ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"far_tutorial_4_2.cpp far_tutorial_4_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_4_2far_tutorial_4_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to create and manipulate both vertex and varying FarStencilTable to interpolate 2 primvar data buffers vertex positions and vertex colors. include opensubdivfartopologyDescriptor.h include opensubdivfarstencilTable.h include opensubdivfarstencilTableFactory.h include cstdio include cstring ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _data 0 src . _data 0 _data 1 src . _data 1 _data 2 src . _data 2 void Clear void 0 _data 0 _data 1 _data 2 0.0f void AddWithWeight Vertex const src , float weight _data 0 weight src . _data 0 _data 1 weight src . _data 1 _data 2 weight src . _data 2 Public interface ------------------------------------ float const GetData const return _data private float _data 3 ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 24 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f Per-vertex RGB color data static float g_colors 24 1.0f , 0.0f , 0.5f , 0.0f , 1.0f , 0.0f , 0.0f , 0.0f , 1.0f , 1.0f , 1.0f , 1.0f , 1.0f , 1.0f , 0.0f , 0.0f , 1.0f , 1.0f , 1.0f , 0.0f , 1.0f , 0.0f , 0.0f , 0.0f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv static Far TopologyRefiner createTopologyRefiner ------------------------------------------------------------------------------ int main int , char Generate a FarTopologyRefiner see tutorial_1_1 for details. Far TopologyRefiner refiner createTopologyRefiner Uniformly refine the topology up to maxlevel. int maxlevel 4 refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel int nverts refiner - GetLevel maxlevel . GetNumVertices Use the FarStencilTable factory to create discrete stencil table Far StencilTableFactory Options options options . generateIntermediateLevels false only the highest refinement level. options . generateOffsets true Vertex primvar data Create stencils table for vertex interpolation options . interpolationMode Far StencilTableFactory INTERPOLATE_VERTEX Far StencilTable const vertexStencils Far StencilTableFactory Create refiner , options assert nverts vertexStencils - GetNumStencils Allocate vertex primvar buffer 1 stencil for each vertex std vector Vertex vertexBuffer vertexStencils - GetNumStencils Use the cube vertex positions as vertex primvar data Vertex vertexCVs reinterpret_cast Vertex g_verts Varying primvar data Create stencils table for varying interpolation options . interpolationMode Far StencilTableFactory INTERPOLATE_VARYING Far StencilTable const varyingStencils Far StencilTableFactory Create refiner , options assert nverts varyingStencils - GetNumStencils Allocate varying primvar buffer 1 stencil for each vertex std vector Vertex varyingBuffer varyingStencils - GetNumStencils Use per-vertex array of RGB colors as varying primvar data Vertex varyingCVs reinterpret_cast Vertex g_colors delete refiner Apply stencils in frame loop This section would be applied every frame after control vertices have been moved. Apply stencils on the control vertex data to update the primvar data of the refined vertices. vertexStencils - UpdateValues vertexCVs , vertexBuffer 0 varyingStencils - UpdateValues varyingCVs , varyingBuffer 0 Visualization with Maya print a MEL script that generates particles at the location of the refined vertices printf particle for int vert 0 vert int nverts vert float const pos vertexBuffer vert . GetData printf -p f f f n , pos 0 , pos 1 , pos 2 printf -c 1 n Set particle point size 20 -- very large printf addAttr -is true -ln pointSize -at long -dv 20 particleShape1 n Add per-particle color attribute rgbPP printf addAttr -ln rgbPP -dt vectorArray particleShape1 n Set per-particle color values from our varying primvar data printf setAttr particleShape1.rgbPP -type vectorArray d , nverts for int vert 0 vert nverts vert float const color varyingBuffer vert . GetData printf f f f n , color 0 , color 1 , color 2 printf n delete vertexStencils delete varyingStencils return EXIT_SUCCESS ------------------------------------------------------------------------------ static Far TopologyRefiner createTopologyRefiner Populate a topology descriptor with our raw data. typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor. Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options return refiner ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"far_tutorial_4_2.html\"\n }, {\n- \"title\": \"far_tutorial_2_3.cpp\",\n- \"text\": \"far_tutorial_2_3.cpp far_tutorial_2_3.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_2_3far_tutorial_2_3.cpp ------------------------------------------------------------------------------ Tutorial description NOTE The following approaches are approximations to compute smooth normals, for highest fidelity patches should be used for positions and normals, which form the true limit surface. Building on tutorial 3, this example shows how to instantiate a simple mesh, refine it uniformly, interpolate both vertex and face-varying primvar data, and finally calculate approximated smooth normals. The resulting interpolated data is output in obj format. Currently, this tutorial supports 3 methods to approximate smooth normals CrossTriangle Calculates smooth normals accumulating per vertex using 3 verts to generate 2 vectors. This approximation has trouble when working with quads which can be non-planar since it only takes into account half of each face. CrossQuad Calculates smooth normals accumulating per vertex but this time, instead of taking into account only 3 verts it creates 2 vectors crossing the quad. This approximation builds upon CrossTriangle but takes into account the 4 verts of the face. Limit Calculates the normals at the limit for each vert at the last level of subdivision. These are the true limit normals, however, in this example they are used with verts that are not at the limit. This can lead to new visual artifacts since the normals and the positions dont match. Additionally, this approach requires extra computation to calculate the limit normals. For this reason, we strongly suggest using limit positions with limit normals. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include cstdio ------------------------------------------------------------------------------ Math helpers. Returns the normalized version of the input vector inline void normalize float n float rn 1.0f sqrtf n 0 n 0 n 1 n 1 n 2 n 2 n 0 rn n 1 rn n 2 rn Returns the cross product of p v1 and p v2. void cross float const v1 , float const v2 , float vOut vOut 0 v1 1 v2 2 - v1 2 v2 1 vOut 1 v1 2 v2 0 - v1 0 v2 2 vOut 2 v1 0 v2 1 - v1 1 v2 0 ------------------------------------------------------------------------------ Face-varying implementation. struct Vertex Minimal required interface ---------------------- Vertex Clear Vertex Vertex const src position 0 src . position 0 position 1 src . position 1 position 2 src . position 2 void Clear position 0 position 1 position 2 0.0f void AddWithWeight Vertex const src , float weight position 0 weight src . position 0 position 1 weight src . position 1 position 2 weight src . position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z position 0 x position 1 y position 2 z const float GetPosition const return position float position 3 ------------------------------------------------------------------------------ Face-varying container implementation. We are using a uv texture layout as a face-varying primtiive variable attribute. Because face-varying data is specified per-face-per-vertex, we cannot use the same container that we use for vertex or varying data. We specify a new container, which only carries u,v coordinates. Similarly to our Vertex container, we add a minimaliztic interpolation interface with a Clear and AddWithWeight methods. struct FVarVertexUV Minimal required interface ---------------------- void Clear u v 0.0f void AddWithWeight FVarVertexUV const src , float weight u weight src . u v weight src . v Basic uv layout channel float u , v struct FVarVertexColor Minimal required interface ---------------------- void Clear r g b a 0.0f void AddWithWeight FVarVertexColor const src , float weight r weight src . r g weight src . g b weight src . b a weight src . a Basic color layout channel float r , g , b , a ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h vertex primitive variable data topology static float g_verts 8 3 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 face-varying primitive variable data topology for UVs static float g_uvs 14 2 0.375 , 0.00 , 0.625 , 0.00 , 0.375 , 0.25 , 0.625 , 0.25 , 0.375 , 0.50 , 0.625 , 0.50 , 0.375 , 0.75 , 0.625 , 0.75 , 0.375 , 1.00 , 0.625 , 1.00 , 0.875 , 0.00 , 0.875 , 0.25 , 0.125 , 0.00 , 0.125 , 0.25 static int g_nuvs 14 static int g_uvIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 9 , 8 , 1 , 10 , 11 , 3 , 12 , 0 , 2 , 13 face-varying primitive variable data topology for color static float g_colors 24 4 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 static int g_ncolors 24 static int g_colorIndices 24 0 , 3 , 9 , 6 , 7 , 10 , 15 , 12 , 13 , 16 , 21 , 18 , 19 , 22 , 4 , 1 , 5 , 23 , 17 , 11 , 20 , 2 , 8 , 14 using namespace OpenSubdiv Approximation methods for smooth normal computations enum NormalApproximation CrossTriangle , CrossQuad , Limit ------------------------------------------------------------------------------ int main int argc , char argv const int maxlevel 2 enum NormalApproximation normalApproximation CrossTriangle Parsing command line parameters to see if the user wants to use a specific method to calculate normals for int i 1 i argc i if strstr argv i , -limit normalApproximation Limit else if strcmp argv i , -crossquad normalApproximation CrossQuad else if strcmp argv i , -crosstriangle normalApproximation CrossTriangle else printf Parameters n printf -crosstriangle use the cross product of vectors n printf generated from 3 verts default. n printf -crossquad use the cross product of vectors n printf generated from 4 verts. n printf -limit use normals calculated from the limit. n return 0 typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY options . SetFVarLinearInterpolation Sdc Options FVAR_LINEAR_NONE Populate a topology descriptor with our raw data Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Create a face-varying channel descriptor const int numChannels 2 const int channelUV 0 const int channelColor 1 Descriptor FVarChannel channels numChannels channels channelUV . numValues g_nuvs channels channelUV . valueIndices g_uvIndices channels channelColor . numValues g_ncolors channels channelColor . valueIndices g_colorIndices Add the channel topology to the main descriptor desc . numFVarChannels numChannels desc . fvarChannels channels Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options Uniformly refine the topolgy up to maxlevel note fullTopologyInLastLevel must be true to work with face-varying data Far TopologyRefiner UniformOptions refineOptions maxlevel refineOptions . fullTopologyInLastLevel true refiner - RefineUniform refineOptions Allocate and initialize the vertex primvar data see tutorial 2 for more details. std vector Vertex vbuffer refiner - GetNumVerticesTotal Vertex verts vbuffer 0 for int i 0 i g_nverts i verts i . SetPosition g_verts i 0 , g_verts i 1 , g_verts i 2 Allocate initialize the first channel of face-varying primvars UVs std vector FVarVertexUV fvBufferUV refiner - GetNumFVarValuesTotal channelUV FVarVertexUV fvVertsUV fvBufferUV 0 for int i 0 i g_nuvs i fvVertsUV i . u g_uvs i 0 fvVertsUV i . v g_uvs i 1 Allocate interpolate the face-varying primvar data colors std vector FVarVertexColor fvBufferColor refiner - GetNumFVarValuesTotal channelColor FVarVertexColor fvVertsColor fvBufferColor 0 for int i 0 i g_ncolors i fvVertsColor i . r g_colors i 0 fvVertsColor i . g g_colors i 1 fvVertsColor i . b g_colors i 2 fvVertsColor i . a g_colors i 3 Interpolate both vertex and face-varying primvar data Far PrimvarRefiner primvarRefiner refiner Vertex srcVert verts FVarVertexUV srcFVarUV fvVertsUV FVarVertexColor srcFVarColor fvVertsColor for int level 1 level maxlevel level Vertex dstVert srcVert refiner - GetLevel level -1 . GetNumVertices FVarVertexUV dstFVarUV srcFVarUV refiner - GetLevel level -1 . GetNumFVarValues channelUV FVarVertexColor dstFVarColor srcFVarColor refiner - GetLevel level -1 . GetNumFVarValues channelColor primvarRefiner . Interpolate level , srcVert , dstVert primvarRefiner . InterpolateFaceVarying level , srcFVarUV , dstFVarUV , channelUV primvarRefiner . InterpolateFaceVarying level , srcFVarColor , dstFVarColor , channelColor srcVert dstVert srcFVarUV dstFVarUV srcFVarColor dstFVarColor Approximate normals Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int nverts refLastLevel . GetNumVertices int nfaces refLastLevel . GetNumFaces int firstOfLastVerts refiner - GetNumVerticesTotal - nverts std vector Vertex normals nverts Different ways to approximate smooth normals For details check the description at the beginning of the file if normalApproximation Limit Approximation using the normal at the limit with verts that are not at the limit For details check the description at the beginning of the file std vector Vertex fineLimitPos nverts std vector Vertex fineDu nverts std vector Vertex fineDv nverts primvarRefiner . Limit verts firstOfLastVerts , fineLimitPos , fineDu , fineDv for int vert 0 vert nverts vert float const du fineDu vert . GetPosition float const dv fineDv vert . GetPosition float norm 3 cross du , dv , norm normals vert . SetPosition norm 0 , norm 1 , norm 2 else if normalApproximation CrossQuad Approximate smooth normals by accumulating normal vectors computed as the cross product of two vectors generated by the 4 verts that form each quad For details check the description at the beginning of the file for int f 0 f nfaces f Far ConstIndexArray faceVertices refLastLevel . GetFaceVertices f We will use the first three verts to calculate a normal const float v0 verts firstOfLastVerts faceVertices 0 . GetPosition const float v1 verts firstOfLastVerts faceVertices 1 . GetPosition const float v2 verts firstOfLastVerts faceVertices 2 . GetPosition const float v3 verts firstOfLastVerts faceVertices 3 . GetPosition Calculate the cross product between the vectors formed by v1-v0 and v2-v0, and then normalize the result float normalCalculated 0.0 , 0.0 , 0.0 float a 3 v2 0 - v0 0 , v2 1 - v0 1 , v2 2 - v0 2 float b 3 v3 0 - v1 0 , v3 1 - v1 1 , v3 2 - v1 2 cross a , b , normalCalculated normalize normalCalculated Accumulate that normal on all verts that are part of that face for int vInFace 0 vInFace faceVertices . size vInFace int vertexIndex faceVertices vInFace normals vertexIndex . position 0 normalCalculated 0 normals vertexIndex . position 1 normalCalculated 1 normals vertexIndex . position 2 normalCalculated 2 else if normalApproximation CrossTriangle Approximate smooth normals by accumulating normal vectors computed as the cross product of two vectors generated by 3 verts of the quad For details check the description at the beginning of the file for int f 0 f nfaces f Far ConstIndexArray faceVertices refLastLevel . GetFaceVertices f We will use the first three verts to calculate a normal const float v0 verts firstOfLastVerts faceVertices 0 . GetPosition const float v1 verts firstOfLastVerts faceVertices 1 . GetPosition const float v2 verts firstOfLastVerts faceVertices 2 . GetPosition Calculate the cross product between the vectors formed by v1-v0 and v2-v0, and then normalize the result float normalCalculated 0.0 , 0.0 , 0.0 float a 3 v1 0 - v0 0 , v1 1 - v0 1 , v1 2 - v0 2 float b 3 v2 0 - v0 0 , v2 1 - v0 1 , v2 2 - v0 2 cross a , b , normalCalculated normalize normalCalculated Accumulate that normal on all verts that are part of that face for int vInFace 0 vInFace faceVertices . size vInFace int vertexIndex faceVertices vInFace normals vertexIndex . position 0 normalCalculated 0 normals vertexIndex . position 1 normalCalculated 1 normals vertexIndex . position 2 normalCalculated 2 Finally we just need to normalize the accumulated normals for int vert 0 vert nverts vert normalize normals vert . position 0 Output OBJ of the highest level refined ----------- Print vertex positions for int vert 0 vert nverts vert float const pos verts firstOfLastVerts vert . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 Print vertex normals for int vert 0 vert nverts vert float const pos normals vert . GetPosition printf vn f f f n , pos 0 , pos 1 , pos 2 Print uvs int nuvs refLastLevel . GetNumFVarValues channelUV int firstOfLastUvs refiner - GetNumFVarValuesTotal channelUV - nuvs for int fvvert 0 fvvert nuvs fvvert FVarVertexUV const uv fvVertsUV firstOfLastUvs fvvert printf vt f f n , uv . u , uv . v Print faces for int face 0 face nfaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face Far ConstIndexArray fuvs refLastLevel . GetFaceFVarValues face , channelUV all refined Catmark faces should be quads assert fverts . size 4 fuvs . size 4 printf f for int vert 0 vert fverts . size vert OBJ uses 1-based arrays... printf ddd , fverts vert 1 , fuvs vert 1 , fverts vert 1 printf n delete refiner return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"far_tutorial_3_1.cpp\",\n+ \"text\": \"far_tutorial_3_1.cpp far_tutorial_3_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_3_1far_tutorial_3_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to interface a high-level topology representation with Far for better efficiency. In tutorial 0, we showed how to instantiate topology from a simple face-vertex list. Here we will show how to take advantage of more complex data structures. Many client applications that manipulate geometry use advanced data structures such as half-edge, quad-edge or winged-edge in order to represent complex topological relationships beyond the usual face-vertex lists. We can take advantage of this information. Far provides an advanced interface that allows such a client application to communicate advanced component relationships directly and avoid having Far rebuilding them redundantly. include opensubdivfartopologyRefinerFactory.h include opensubdivfarprimvarRefiner.h include cstdio ------------------------------------------------------------------------------ using namespace OpenSubdiv ------------------------------------------------------------------------------ For this tutorial, we provide the complete topological representation of a simple pyramid. In our case, we store it as a simple sequence of integers, with the understanding that client-code would provide a fully implemented data-structure such as quad-edges or winged-edges. Pyramid geometry from catmark_pyramid.h - extended for this tutorial static int g_nverts 5 , g_nedges 8 , g_nfaces 5 vertex positions static float g_verts 5 3 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 2.0f , 0.0f , 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 0.0f number of vertices in each face static int g_facenverts 5 3 , 3 , 3 , 3 , 4 index of face vertices static int g_faceverts 16 0 , 1 , 2 , 0 , 2 , 3 , 0 , 3 , 4 , 0 , 4 , 1 , 4 , 3 , 2 , 1 index of edge vertices 2 per edge static int g_edgeverts 16 0 , 1 , 1 , 2 , 2 , 0 , 2 , 3 , 3 , 0 , 3 , 4 , 4 , 0 , 4 , 1 index of face edges static int g_faceedges 16 0 , 1 , 2 , 2 , 3 , 4 , 4 , 5 , 6 , 6 , 7 , 0 , 5 , 3 , 1 , 7 number of faces adjacent to each edge static int g_edgenfaces 8 2 , 2 , 2 , 2 , 2 , 2 , 2 , 2 index of faces incident to a given edge static int g_edgefaces 16 0 , 3 , 0 , 4 , 0 , 1 , 1 , 4 , 1 , 2 , 2 , 4 , 2 , 3 , 3 , 4 number of faces incident to each vertex static int g_vertexnfaces 5 4 , 3 , 3 , 3 , 3 index of faces incident to each vertex static int g_vertexfaces 25 0 , 1 , 2 , 3 , 0 , 3 , 4 , 0 , 4 , 1 , 1 , 4 , 2 , 2 , 4 , 3 number of edges incident to each vertex static int g_vertexnedges 5 4 , 3 , 3 , 3 , 3 index of edges incident to each vertex static int g_vertexedges 25 0 , 2 , 4 , 6 , 1 , 0 , 7 , 2 , 1 , 3 , 4 , 3 , 5 , 6 , 5 , 7 Edge crease sharpness static float g_edgeCreases 8 0.0f , 2.5f , 0.0f , 2.5f , 0.0f , 2.5f , 0.0f , 2.5f ------------------------------------------------------------------------------ Because existing client-code may not provide an exact match for the topological queries required by Fars interface, we can provide a converter class. This can be particularly useful for instance if the client data-structure requires additional relationships to be mapped. For instance, half-edge representations do not store unique edge indices and it can be difficult to traverse edges or faces adjacent to a given vertex. Using an intermediate wrapper class allows us to leverage existing relationships information from a mesh, and generate the missing components temporarily. For a practical example, you can look at the file hbr_to_vtr.h in the same tutorial directory. This example implements a OsdHbrConverter class as a way of interfacing PRmans half-edge representation to Far. struct Converter public Sdc SchemeType GetType const return Sdc SCHEME_CATMARK Sdc Options GetOptions const Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY return options int GetNumFaces const return g_nfaces int GetNumEdges const return g_nedges int GetNumVertices const return g_nverts Face relationships int GetNumFaceVerts int face const return g_facenverts face int const GetFaceVerts int face const return g_faceverts getCompOffset g_facenverts , face int const GetFaceEdges int face const return g_faceedges getCompOffset g_facenverts , face Edge relationships int const GetEdgeVertices int edge const return g_edgeverts edge 2 int GetNumEdgeFaces int edge const return g_edgenfaces edge int const GetEdgeFaces int edge const return g_edgefaces getCompOffset g_edgenfaces , edge Vertex relationships int GetNumVertexEdges int vert const return g_vertexnedges vert int const GetVertexEdges int vert const return g_vertexedges getCompOffset g_vertexnedges , vert int GetNumVertexFaces int vert const return g_vertexnfaces vert int const GetVertexFaces int vert const return g_vertexfaces getCompOffset g_vertexnfaces , vert private int getCompOffset int const comps , int comp const int ofs 0 for int i 0 i comp i ofs comps i return ofs ------------------------------------------------------------------------------ namespace OpenSubdiv namespace OPENSUBDIV_VERSION namespace Far template bool TopologyRefinerFactory Converter resizeComponentTopology TopologyRefiner refiner , Converter const conv Faces and face-verts int nfaces conv . GetNumFaces setNumBaseFaces refiner , nfaces for int face 0 face nfaces face int nv conv . GetNumFaceVerts face setNumBaseFaceVertices refiner , face , nv Edges and edge-faces int nedges conv . GetNumEdges setNumBaseEdges refiner , nedges for int edge 0 edge nedges edge int nf conv . GetNumEdgeFaces edge setNumBaseEdgeFaces refiner , edge , nf Vertices and vert-faces and vert-edges int nverts conv . GetNumVertices setNumBaseVertices refiner , nverts for int vert 0 vert nverts vert int ne conv . GetNumVertexEdges vert , nf conv . GetNumVertexFaces vert setNumBaseVertexEdges refiner , vert , ne setNumBaseVertexFaces refiner , vert , nf return true template bool TopologyRefinerFactory Converter assignComponentTopology TopologyRefiner refiner , Converter const conv using Far IndexArray Face relations int nfaces conv . GetNumFaces for int face 0 face nfaces face IndexArray dstFaceVerts getBaseFaceVertices refiner , face IndexArray dstFaceEdges getBaseFaceEdges refiner , face int const faceverts conv . GetFaceVerts face int const faceedges conv . GetFaceEdges face for int vert 0 vert conv . GetNumFaceVerts face vert dstFaceVerts vert faceverts vert dstFaceEdges vert faceedges vert Edge relations Note if your representation is unable to provide edge relationships ex half-edges, you can comment out this section and Far will automatically generate the missing information. int nedges conv . GetNumEdges for int edge 0 edge nedges edge Edge-vertices IndexArray dstEdgeVerts getBaseEdgeVertices refiner , edge dstEdgeVerts 0 conv . GetEdgeVertices edge 0 dstEdgeVerts 1 conv . GetEdgeVertices edge 1 Edge-faces IndexArray dstEdgeFaces getBaseEdgeFaces refiner , edge for int face 0 face conv . GetNumEdgeFaces face face dstEdgeFaces face conv . GetEdgeFaces edge face Vertex relations int nverts conv . GetNumVertices for int vert 0 vert nverts vert Vert-Faces IndexArray vertFaces getBaseVertexFaces refiner , vert LocalIndexArray vertInFaceIndices getBaseVertexFaceLocalIndicesrefiner, vert for int face 0 face conv . GetNumVertexFaces vert face vertFaces face conv . GetVertexFaces vert face Vert-Edges IndexArray vertEdges getBaseVertexEdges refiner , vert LocalIndexArray vertInEdgeIndices getBaseVertexEdgeLocalIndicesrefiner, vert for int edge 0 edge conv . GetNumVertexEdges vert edge vertEdges edge conv . GetVertexEdges vert edge populateBaseLocalIndices refiner return true template bool TopologyRefinerFactory Converter assignComponentTags TopologyRefiner refiner , Converter const conv arbitrarily sharpen the 4 bottom edges of the pyramid to 2.5f for int edge 0 edge conv . GetNumEdges edge setBaseEdgeSharpness refiner , edge , g_edgeCreases edge return true ifdef _MSC_VER template void TopologyRefinerFactory Converter reportInvalidTopology TopologyError errCode , char const msg , Converter const mesh Optional topology validation error reporting This method is called whenever the factory encounters topology validation errors. By default, nothing is reported Warning msg template bool TopologyRefinerFactory Converter assignFaceVaryingTopology TopologyRefiner refiner , Converter const conv Because of the way MSVC specializes templated functions, we had to remove the default stubs in FarTopologyRefinerFactory. In this example, no face-varying data is being added, but we still need to implement a template specialization or MSVC linker fails. return true endif namespace Far namespace OPENSUBDIV_VERSION namespace OpenSubdiv ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 ------------------------------------------------------------------------------ int main int , char Converter conv Far TopologyRefiner refiner Far TopologyRefinerFactory Converter Create conv , Far TopologyRefinerFactory Converter Options conv . GetType , conv . GetOptions int maxlevel 5 Uniformly refine the topology up to maxlevel refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Allocate a buffer for vertex primvar data. The buffer length is set to be the sum of all children vertices up to the highest level of refinement. std vector Vertex vbuffer refiner - GetNumVerticesTotal Vertex verts vbuffer 0 Initialize coarse mesh positions int nCoarseVerts g_nverts for int i 0 i nCoarseVerts i verts i . SetPosition g_verts i 0 , g_verts i 1 , g_verts i 2 Interpolate vertex primvar data Far PrimvarRefiner primvarRefiner refiner Vertex src verts for int level 1 level maxlevel level Vertex dst src refiner - GetLevel level -1 . GetNumVertices primvarRefiner . Interpolate level , src , dst src dst Output OBJ of the highest level refined ----------- Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int nverts refLastLevel . GetNumVertices int nfaces refLastLevel . GetNumFaces Print vertex positions int firstOfLastVerts refiner - GetNumVerticesTotal - nverts for int vert 0 vert nverts vert float const pos verts firstOfLastVerts vert . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 Print faces for int face 0 face nfaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face all refined Catmark faces should be quads assert fverts . size 4 printf f for int vert 0 vert fverts . size vert printf d , fverts vert 1 OBJ uses 1-based arrays... printf n delete refiner return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"far_tutorial_2_3.html\"\n+ \"loc\": \"far_tutorial_3_1.html\"\n }, {\n \"title\": \"far_tutorial_4_1.cpp\",\n- \"text\": \"far_tutorial_4_1.cpp far_tutorial_4_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_4_1far_tutorial_4_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to create and manipulate FarStencilTable. We use the factorized stencils to interpolate vertex primvar data buffers. include opensubdivfartopologyDescriptor.h include opensubdivfarstencilTable.h include opensubdivfarstencilTableFactory.h include cstdio include cstring ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z float const GetPosition const return _position private float _position 3 ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 24 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv static Far TopologyRefiner createTopologyRefiner ------------------------------------------------------------------------------ int main int , char Generate a FarTopologyRefiner see tutorial_1_1 for details. Far TopologyRefiner refiner createTopologyRefiner Uniformly refine the topology up to maxlevel. int maxlevel 3 refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Use the FarStencilTable factory to create discrete stencil table note we only want stencils for the highest refinement level. Far StencilTableFactory Options options options . generateIntermediateLevels false options . generateOffsets true Far StencilTable const stencilTable Far StencilTableFactory Create refiner , options Allocate vertex primvar buffer 1 stencil for each vertex int nstencils stencilTable - GetNumStencils std vector Vertex vertexBuffer nstencils Quick dirty re-cast of the primvar data from our cube this is where you would drive shape deformations every frame Vertex controlValues reinterpret_cast Vertex g_verts This section would be applied every frame after control vertices have been moved. Apply stencils on the control vertex data to update the primvar data of the refined vertices. stencilTable - UpdateValues controlValues , vertexBuffer 0 Visualization with Maya print a MEL script that generates particles at the location of the refined vertices printf particle for int i 0 i int vertexBuffer . size i float const pos vertexBuffer i . GetPosition printf -p f f f n , pos 0 , pos 1 , pos 2 printf -c 1 n delete refiner delete stencilTable return EXIT_SUCCESS ------------------------------------------------------------------------------ static Far TopologyRefiner createTopologyRefiner Populate a topology descriptor with our raw data. typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor. return Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"far_tutorial_4_1.cpp far_tutorial_4_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_4_1far_tutorial_4_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to create and manipulate FarStencilTable. We use the factorized stencils to interpolate vertex primvar data buffers. include opensubdivfartopologyDescriptor.h include opensubdivfarstencilTable.h include opensubdivfarstencilTableFactory.h include cstdio include cstring ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z float const GetPosition const return _position private float _position 3 ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 24 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv static Far TopologyRefiner createTopologyRefiner ------------------------------------------------------------------------------ int main int , char Generate a FarTopologyRefiner see tutorial_1_1 for details. Far TopologyRefiner refiner createTopologyRefiner Uniformly refine the topology up to maxlevel. int maxlevel 3 refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Use the FarStencilTable factory to create discrete stencil table note we only want stencils for the highest refinement level. Far StencilTableFactory Options options options . generateIntermediateLevels false options . generateOffsets true Far StencilTable const stencilTable Far StencilTableFactory Create refiner , options Allocate vertex primvar buffer 1 stencil for each vertex int nstencils stencilTable - GetNumStencils std vector Vertex vertexBuffer nstencils Quick dirty re-cast of the primvar data from our cube this is where you would drive shape deformations every frame Vertex controlValues reinterpret_cast Vertex g_verts This section would be applied every frame after control vertices have been moved. Apply stencils on the control vertex data to update the primvar data of the refined vertices. stencilTable - UpdateValues controlValues , vertexBuffer 0 Visualization with Maya print a MEL script that generates particles at the location of the refined vertices printf particle for int i 0 i int vertexBuffer . size i float const pos vertexBuffer i . GetPosition printf -p f f f n , pos 0 , pos 1 , pos 2 printf -c 1 n delete refiner delete stencilTable return EXIT_SUCCESS ------------------------------------------------------------------------------ static Far TopologyRefiner createTopologyRefiner Populate a topology descriptor with our raw data. typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor. return Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"far_tutorial_4_1.html\"\n }, {\n \"title\": \"far_tutorial_2_2.cpp\",\n- \"text\": \"far_tutorial_2_2.cpp far_tutorial_2_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_2_2far_tutorial_2_2.cpp ------------------------------------------------------------------------------ Tutorial description Building on tutorial 0, this example shows how to instantiate a simple mesh, refine it uniformly and then interpolate both vertex and face-varying primvar data. The resulting interpolated data is output as an obj file, with the face-varying data recorded in the uv texture layout. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include cstdio ------------------------------------------------------------------------------ Face-varying implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 ------------------------------------------------------------------------------ Face-varying container implementation. We are using a uv texture layout as a face-varying primitive variable attribute. Because face-varying data is specified per-face-per-vertex, we cannot use the same container that we use for vertex or varying data. We specify a new container, which only carries u,v coordinates. Similarly to our Vertex container, we add a minimalistic interpolation interface with a Clear and AddWithWeight methods. struct FVarVertexUV Minimal required interface ---------------------- void Clear u v 0.0f void AddWithWeight FVarVertexUV const src , float weight u weight src . u v weight src . v Basic uv layout channel float u , v struct FVarVertexColor Minimal required interface ---------------------- void Clear r g b a 0.0f void AddWithWeight FVarVertexColor const src , float weight r weight src . r g weight src . g b weight src . b a weight src . a Basic color layout channel float r , g , b , a ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h vertex primitive variable data topology static float g_verts 8 3 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 face-varying primitive variable data topology for UVs static float g_uvs 14 2 0.375 , 0.00 , 0.625 , 0.00 , 0.375 , 0.25 , 0.625 , 0.25 , 0.375 , 0.50 , 0.625 , 0.50 , 0.375 , 0.75 , 0.625 , 0.75 , 0.375 , 1.00 , 0.625 , 1.00 , 0.875 , 0.00 , 0.875 , 0.25 , 0.125 , 0.00 , 0.125 , 0.25 static int g_nuvs 14 static int g_uvIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 9 , 8 , 1 , 10 , 11 , 3 , 12 , 0 , 2 , 13 face-varying primitive variable data topology for color static float g_colors 24 4 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 static int g_ncolors 24 static int g_colorIndices 24 0 , 3 , 9 , 6 , 7 , 10 , 15 , 12 , 13 , 16 , 21 , 18 , 19 , 22 , 4 , 1 , 5 , 23 , 17 , 11 , 20 , 2 , 8 , 14 using namespace OpenSubdiv ------------------------------------------------------------------------------ int main int , char int maxlevel 3 typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY options . SetFVarLinearInterpolation Sdc Options FVAR_LINEAR_NONE Populate a topology descriptor with our raw data Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices int channelUV 0 int channelColor 1 Create a face-varying channel descriptor Descriptor FVarChannel channels 2 channels channelUV . numValues g_nuvs channels channelUV . valueIndices g_uvIndices channels channelColor . numValues g_ncolors channels channelColor . valueIndices g_colorIndices Add the channel topology to the main descriptor desc . numFVarChannels 2 desc . fvarChannels channels Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options Uniformly refine the topology up to maxlevel note fullTopologyInLastLevel must be true to work with face-varying data Far TopologyRefiner UniformOptions refineOptions maxlevel refineOptions . fullTopologyInLastLevel true refiner - RefineUniform refineOptions Allocate and initialize the vertex primvar data see tutorial 2 for more details. std vector Vertex vbuffer refiner - GetNumVerticesTotal Vertex verts vbuffer 0 for int i 0 i g_nverts i verts i . SetPosition g_verts i 0 , g_verts i 1 , g_verts i 2 Allocate and initialize the first channel of face-varying primvar data UVs std vector FVarVertexUV fvBufferUV refiner - GetNumFVarValuesTotal channelUV FVarVertexUV fvVertsUV fvBufferUV 0 for int i 0 i g_nuvs i fvVertsUV i . u g_uvs i 0 fvVertsUV i . v g_uvs i 1 Allocate interpolate the face-varying primvar data colors std vector FVarVertexColor fvBufferColor refiner - GetNumFVarValuesTotal channelColor FVarVertexColor fvVertsColor fvBufferColor 0 for int i 0 i g_ncolors i fvVertsColor i . r g_colors i 0 fvVertsColor i . g g_colors i 1 fvVertsColor i . b g_colors i 2 fvVertsColor i . a g_colors i 3 Interpolate both vertex and face-varying primvar data Far PrimvarRefiner primvarRefiner refiner Vertex srcVert verts FVarVertexUV srcFVarUV fvVertsUV FVarVertexColor srcFVarColor fvVertsColor for int level 1 level maxlevel level Vertex dstVert srcVert refiner - GetLevel level -1 . GetNumVertices FVarVertexUV dstFVarUV srcFVarUV refiner - GetLevel level -1 . GetNumFVarValues channelUV FVarVertexColor dstFVarColor srcFVarColor refiner - GetLevel level -1 . GetNumFVarValues channelColor primvarRefiner . Interpolate level , srcVert , dstVert primvarRefiner . InterpolateFaceVarying level , srcFVarUV , dstFVarUV , channelUV primvarRefiner . InterpolateFaceVarying level , srcFVarColor , dstFVarColor , channelColor srcVert dstVert srcFVarUV dstFVarUV srcFVarColor dstFVarColor Output OBJ of the highest level refined ----------- Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int nverts refLastLevel . GetNumVertices int nuvs refLastLevel . GetNumFVarValues channelUV int ncolors refLastLevel . GetNumFVarValues channelColor int nfaces refLastLevel . GetNumFaces Print vertex positions int firstOfLastVerts refiner - GetNumVerticesTotal - nverts for int vert 0 vert nverts vert float const pos verts firstOfLastVerts vert . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 Print uvs int firstOfLastUvs refiner - GetNumFVarValuesTotal channelUV - nuvs for int fvvert 0 fvvert nuvs fvvert FVarVertexUV const uv fvVertsUV firstOfLastUvs fvvert printf vt f f n , uv . u , uv . v Print colors int firstOfLastColors refiner - GetNumFVarValuesTotal channelColor - ncolors for int fvvert 0 fvvert ncolors fvvert FVarVertexColor const c fvVertsColor firstOfLastColors fvvert printf c f f f f n , c . r , c . g , c . b , c . a Print faces for int face 0 face nfaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face Far ConstIndexArray fuvs refLastLevel . GetFaceFVarValues face , channelUV all refined Catmark faces should be quads assert fverts . size 4 fuvs . size 4 printf f for int vert 0 vert fverts . size vert OBJ uses 1-based arrays... printf dd , fverts vert 1 , fuvs vert 1 printf n delete refiner return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"far_tutorial_2_2.cpp far_tutorial_2_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_2_2far_tutorial_2_2.cpp ------------------------------------------------------------------------------ Tutorial description Building on tutorial 0, this example shows how to instantiate a simple mesh, refine it uniformly and then interpolate both vertex and face-varying primvar data. The resulting interpolated data is output as an obj file, with the face-varying data recorded in the uv texture layout. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include cstdio ------------------------------------------------------------------------------ Face-varying implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 ------------------------------------------------------------------------------ Face-varying container implementation. We are using a uv texture layout as a face-varying primitive variable attribute. Because face-varying data is specified per-face-per-vertex, we cannot use the same container that we use for vertex or varying data. We specify a new container, which only carries u,v coordinates. Similarly to our Vertex container, we add a minimalistic interpolation interface with a Clear and AddWithWeight methods. struct FVarVertexUV Minimal required interface ---------------------- void Clear u v 0.0f void AddWithWeight FVarVertexUV const src , float weight u weight src . u v weight src . v Basic uv layout channel float u , v struct FVarVertexColor Minimal required interface ---------------------- void Clear r g b a 0.0f void AddWithWeight FVarVertexColor const src , float weight r weight src . r g weight src . g b weight src . b a weight src . a Basic color layout channel float r , g , b , a ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h vertex primitive variable data topology static float g_verts 8 3 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 face-varying primitive variable data topology for UVs static float g_uvs 14 2 0.375 , 0.00 , 0.625 , 0.00 , 0.375 , 0.25 , 0.625 , 0.25 , 0.375 , 0.50 , 0.625 , 0.50 , 0.375 , 0.75 , 0.625 , 0.75 , 0.375 , 1.00 , 0.625 , 1.00 , 0.875 , 0.00 , 0.875 , 0.25 , 0.125 , 0.00 , 0.125 , 0.25 static int g_nuvs 14 static int g_uvIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 9 , 8 , 1 , 10 , 11 , 3 , 12 , 0 , 2 , 13 face-varying primitive variable data topology for color static float g_colors 24 4 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 static int g_ncolors 24 static int g_colorIndices 24 0 , 3 , 9 , 6 , 7 , 10 , 15 , 12 , 13 , 16 , 21 , 18 , 19 , 22 , 4 , 1 , 5 , 23 , 17 , 11 , 20 , 2 , 8 , 14 using namespace OpenSubdiv ------------------------------------------------------------------------------ int main int , char int maxlevel 3 typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY options . SetFVarLinearInterpolation Sdc Options FVAR_LINEAR_NONE Populate a topology descriptor with our raw data Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices int channelUV 0 int channelColor 1 Create a face-varying channel descriptor Descriptor FVarChannel channels 2 channels channelUV . numValues g_nuvs channels channelUV . valueIndices g_uvIndices channels channelColor . numValues g_ncolors channels channelColor . valueIndices g_colorIndices Add the channel topology to the main descriptor desc . numFVarChannels 2 desc . fvarChannels channels Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options Uniformly refine the topology up to maxlevel note fullTopologyInLastLevel must be true to work with face-varying data Far TopologyRefiner UniformOptions refineOptions maxlevel refineOptions . fullTopologyInLastLevel true refiner - RefineUniform refineOptions Allocate and initialize the vertex primvar data see tutorial 2 for more details. std vector Vertex vbuffer refiner - GetNumVerticesTotal Vertex verts vbuffer 0 for int i 0 i g_nverts i verts i . SetPosition g_verts i 0 , g_verts i 1 , g_verts i 2 Allocate and initialize the first channel of face-varying primvar data UVs std vector FVarVertexUV fvBufferUV refiner - GetNumFVarValuesTotal channelUV FVarVertexUV fvVertsUV fvBufferUV 0 for int i 0 i g_nuvs i fvVertsUV i . u g_uvs i 0 fvVertsUV i . v g_uvs i 1 Allocate interpolate the face-varying primvar data colors std vector FVarVertexColor fvBufferColor refiner - GetNumFVarValuesTotal channelColor FVarVertexColor fvVertsColor fvBufferColor 0 for int i 0 i g_ncolors i fvVertsColor i . r g_colors i 0 fvVertsColor i . g g_colors i 1 fvVertsColor i . b g_colors i 2 fvVertsColor i . a g_colors i 3 Interpolate both vertex and face-varying primvar data Far PrimvarRefiner primvarRefiner refiner Vertex srcVert verts FVarVertexUV srcFVarUV fvVertsUV FVarVertexColor srcFVarColor fvVertsColor for int level 1 level maxlevel level Vertex dstVert srcVert refiner - GetLevel level -1 . GetNumVertices FVarVertexUV dstFVarUV srcFVarUV refiner - GetLevel level -1 . GetNumFVarValues channelUV FVarVertexColor dstFVarColor srcFVarColor refiner - GetLevel level -1 . GetNumFVarValues channelColor primvarRefiner . Interpolate level , srcVert , dstVert primvarRefiner . InterpolateFaceVarying level , srcFVarUV , dstFVarUV , channelUV primvarRefiner . InterpolateFaceVarying level , srcFVarColor , dstFVarColor , channelColor srcVert dstVert srcFVarUV dstFVarUV srcFVarColor dstFVarColor Output OBJ of the highest level refined ----------- Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int nverts refLastLevel . GetNumVertices int nuvs refLastLevel . GetNumFVarValues channelUV int ncolors refLastLevel . GetNumFVarValues channelColor int nfaces refLastLevel . GetNumFaces Print vertex positions int firstOfLastVerts refiner - GetNumVerticesTotal - nverts for int vert 0 vert nverts vert float const pos verts firstOfLastVerts vert . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 Print uvs int firstOfLastUvs refiner - GetNumFVarValuesTotal channelUV - nuvs for int fvvert 0 fvvert nuvs fvvert FVarVertexUV const uv fvVertsUV firstOfLastUvs fvvert printf vt f f n , uv . u , uv . v Print colors int firstOfLastColors refiner - GetNumFVarValuesTotal channelColor - ncolors for int fvvert 0 fvvert ncolors fvvert FVarVertexColor const c fvVertsColor firstOfLastColors fvvert printf c f f f f n , c . r , c . g , c . b , c . a Print faces for int face 0 face nfaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face Far ConstIndexArray fuvs refLastLevel . GetFaceFVarValues face , channelUV all refined Catmark faces should be quads assert fverts . size 4 fuvs . size 4 printf f for int vert 0 vert fverts . size vert OBJ uses 1-based arrays... printf dd , fverts vert 1 , fuvs vert 1 printf n delete refiner return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"far_tutorial_2_2.html\"\n }, {\n- \"title\": \"far_tutorial_2_1.cpp\",\n- \"text\": \"far_tutorial_2_1.cpp far_tutorial_2_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_2_1far_tutorial_2_1.cpp ------------------------------------------------------------------------------ Tutorial description Building on tutorial 0, this example shows how to instantiate a simple mesh, refine it uniformly and then interpolate additional sets of primvar data. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include cstdio ------------------------------------------------------------------------------ Vertex container implementation. We are adding a per-vertex color attribute to our primvar data. While they are separate properties and exist in separate buffers as when read from an Alembic file they are both of the form float3 and so we can use the same underlying type. While color and position may be the same, well make the color a varying primvar, e.g. it is constrained to being linearly interpolated between vertices, rather than smoothly like position and other vertex data. struct Point3 Minimal required interface ---------------------- Point3 void Clear void 0 _point 0 _point 1 _point 2 0.0f void AddWithWeight Point3 const src , float weight _point 0 weight src . _point 0 _point 1 weight src . _point 1 _point 2 weight src . _point 2 Public interface ------------------------------------ void SetPoint float x , float y , float z _point 0 x _point 1 y _point 2 z const float GetPoint const return _point private float _point 3 typedef Point3 VertexPosition typedef Point3 VertexColor ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 8 3 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f Per-vertex RGB color data static float g_colors 8 3 1.0f , 0.0f , 0.5f , 0.0f , 1.0f , 0.0f , 0.0f , 0.0f , 1.0f , 1.0f , 1.0f , 1.0f , 1.0f , 1.0f , 0.0f , 0.0f , 1.0f , 1.0f , 1.0f , 0.0f , 1.0f , 0.0f , 0.0f , 0.0f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv static Far TopologyRefiner createFarTopologyRefiner ------------------------------------------------------------------------------ int main int , char int maxlevel 5 Far TopologyRefiner refiner createFarTopologyRefiner Uniformly refine the topology up to maxlevel refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Allocate buffers for vertex primvar data. We assume we received the coarse data for the mesh in separate buffers from some other source, e.g. an Alembic file. Meanwhile, we want buffers for the lastfinest subdivision level to persist. We have no interest in the intermediate levels. Determine the sizes for our needs int nCoarseVerts g_nverts int nFineVerts refiner - GetLevel maxlevel . GetNumVertices int nTotalVerts refiner - GetNumVerticesTotal int nTempVerts nTotalVerts - nCoarseVerts - nFineVerts Allocate and initialize the primvar data for the original coarse vertices std vector VertexPosition coarsePosBuffer nCoarseVerts std vector VertexColor coarseClrBuffer nCoarseVerts for int i 0 i nCoarseVerts i coarsePosBuffer i . SetPoint g_verts i 0 , g_verts i 1 , g_verts i 2 coarseClrBuffer i . SetPoint g_colors i 0 , g_colors i 1 , g_colors i 2 Allocate intermediate and final storage to be populated std vector VertexPosition tempPosBuffer nTempVerts std vector VertexPosition finePosBuffer nFineVerts std vector VertexColor tempClrBuffer nTempVerts std vector VertexColor fineClrBuffer nFineVerts Interpolate all primvar data -- separate buffers can be populated on separate threads if desired VertexPosition srcPos coarsePosBuffer 0 VertexPosition dstPos tempPosBuffer 0 VertexColor srcClr coarseClrBuffer 0 VertexColor dstClr tempClrBuffer 0 Far PrimvarRefiner primvarRefiner refiner for int level 1 level maxlevel level primvarRefiner . Interpolate level , srcPos , dstPos primvarRefiner . InterpolateVarying level , srcClr , dstClr srcPos dstPos , dstPos refiner - GetLevel level . GetNumVertices srcClr dstClr , dstClr refiner - GetLevel level . GetNumVertices Interpolate the last level into the separate buffers for our final data primvarRefiner . Interpolate maxlevel , srcPos , finePosBuffer primvarRefiner . InterpolateVarying maxlevel , srcClr , fineClrBuffer Visualization with Maya print a MEL script that generates colored particles at the location of the refined vertices dont forget to turn shading on in the viewport to see the colors int nverts nFineVerts Output particle positions printf particle for int vert 0 vert nverts vert float const pos finePosBuffer vert . GetPoint printf -p f f f n , pos 0 , pos 1 , pos 2 printf n Set particle point size 20 -- very large printf addAttr -is true -ln pointSize -at long -dv 20 particleShape1 n Add per-particle color attribute rgbPP printf addAttr -ln rgbPP -dt vectorArray particleShape1 n Set per-particle color values from our primvar data printf setAttr particleShape1.rgbPP -type vectorArray d , nverts for int vert 0 vert nverts vert float const color fineClrBuffer vert . GetPoint printf f f f n , color 0 , color 1 , color 2 printf n delete refiner return EXIT_SUCCESS ------------------------------------------------------------------------------ Creates FarTopologyRefiner from raw geometry see tutorial_1_1 for more details static Far TopologyRefiner createFarTopologyRefiner Populate a topology descriptor with our raw data typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options return refiner ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"far_tutorial_2_3.cpp\",\n+ \"text\": \"far_tutorial_2_3.cpp far_tutorial_2_3.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_2_3far_tutorial_2_3.cpp ------------------------------------------------------------------------------ Tutorial description NOTE The following approaches are approximations to compute smooth normals, for highest fidelity patches should be used for positions and normals, which form the true limit surface. Building on tutorial 3, this example shows how to instantiate a simple mesh, refine it uniformly, interpolate both vertex and face-varying primvar data, and finally calculate approximated smooth normals. The resulting interpolated data is output in obj format. Currently, this tutorial supports 3 methods to approximate smooth normals CrossTriangle Calculates smooth normals accumulating per vertex using 3 verts to generate 2 vectors. This approximation has trouble when working with quads which can be non-planar since it only takes into account half of each face. CrossQuad Calculates smooth normals accumulating per vertex but this time, instead of taking into account only 3 verts it creates 2 vectors crossing the quad. This approximation builds upon CrossTriangle but takes into account the 4 verts of the face. Limit Calculates the normals at the limit for each vert at the last level of subdivision. These are the true limit normals, however, in this example they are used with verts that are not at the limit. This can lead to new visual artifacts since the normals and the positions dont match. Additionally, this approach requires extra computation to calculate the limit normals. For this reason, we strongly suggest using limit positions with limit normals. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include cstdio ------------------------------------------------------------------------------ Math helpers. Returns the normalized version of the input vector inline void normalize float n float rn 1.0f sqrtf n 0 n 0 n 1 n 1 n 2 n 2 n 0 rn n 1 rn n 2 rn Returns the cross product of p v1 and p v2. void cross float const v1 , float const v2 , float vOut vOut 0 v1 1 v2 2 - v1 2 v2 1 vOut 1 v1 2 v2 0 - v1 0 v2 2 vOut 2 v1 0 v2 1 - v1 1 v2 0 ------------------------------------------------------------------------------ Face-varying implementation. struct Vertex Minimal required interface ---------------------- Vertex Clear Vertex Vertex const src position 0 src . position 0 position 1 src . position 1 position 2 src . position 2 void Clear position 0 position 1 position 2 0.0f void AddWithWeight Vertex const src , float weight position 0 weight src . position 0 position 1 weight src . position 1 position 2 weight src . position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z position 0 x position 1 y position 2 z const float GetPosition const return position float position 3 ------------------------------------------------------------------------------ Face-varying container implementation. We are using a uv texture layout as a face-varying primtiive variable attribute. Because face-varying data is specified per-face-per-vertex, we cannot use the same container that we use for vertex or varying data. We specify a new container, which only carries u,v coordinates. Similarly to our Vertex container, we add a minimaliztic interpolation interface with a Clear and AddWithWeight methods. struct FVarVertexUV Minimal required interface ---------------------- void Clear u v 0.0f void AddWithWeight FVarVertexUV const src , float weight u weight src . u v weight src . v Basic uv layout channel float u , v struct FVarVertexColor Minimal required interface ---------------------- void Clear r g b a 0.0f void AddWithWeight FVarVertexColor const src , float weight r weight src . r g weight src . g b weight src . b a weight src . a Basic color layout channel float r , g , b , a ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h vertex primitive variable data topology static float g_verts 8 3 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 face-varying primitive variable data topology for UVs static float g_uvs 14 2 0.375 , 0.00 , 0.625 , 0.00 , 0.375 , 0.25 , 0.625 , 0.25 , 0.375 , 0.50 , 0.625 , 0.50 , 0.375 , 0.75 , 0.625 , 0.75 , 0.375 , 1.00 , 0.625 , 1.00 , 0.875 , 0.00 , 0.875 , 0.25 , 0.125 , 0.00 , 0.125 , 0.25 static int g_nuvs 14 static int g_uvIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 9 , 8 , 1 , 10 , 11 , 3 , 12 , 0 , 2 , 13 face-varying primitive variable data topology for color static float g_colors 24 4 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 , 1.0 static int g_ncolors 24 static int g_colorIndices 24 0 , 3 , 9 , 6 , 7 , 10 , 15 , 12 , 13 , 16 , 21 , 18 , 19 , 22 , 4 , 1 , 5 , 23 , 17 , 11 , 20 , 2 , 8 , 14 using namespace OpenSubdiv Approximation methods for smooth normal computations enum NormalApproximation CrossTriangle , CrossQuad , Limit ------------------------------------------------------------------------------ int main int argc , char argv const int maxlevel 2 enum NormalApproximation normalApproximation CrossTriangle Parsing command line parameters to see if the user wants to use a specific method to calculate normals for int i 1 i argc i if strstr argv i , -limit normalApproximation Limit else if strcmp argv i , -crossquad normalApproximation CrossQuad else if strcmp argv i , -crosstriangle normalApproximation CrossTriangle else printf Parameters n printf -crosstriangle use the cross product of vectors n printf generated from 3 verts default. n printf -crossquad use the cross product of vectors n printf generated from 4 verts. n printf -limit use normals calculated from the limit. n return 0 typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY options . SetFVarLinearInterpolation Sdc Options FVAR_LINEAR_NONE Populate a topology descriptor with our raw data Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Create a face-varying channel descriptor const int numChannels 2 const int channelUV 0 const int channelColor 1 Descriptor FVarChannel channels numChannels channels channelUV . numValues g_nuvs channels channelUV . valueIndices g_uvIndices channels channelColor . numValues g_ncolors channels channelColor . valueIndices g_colorIndices Add the channel topology to the main descriptor desc . numFVarChannels numChannels desc . fvarChannels channels Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options Uniformly refine the topolgy up to maxlevel note fullTopologyInLastLevel must be true to work with face-varying data Far TopologyRefiner UniformOptions refineOptions maxlevel refineOptions . fullTopologyInLastLevel true refiner - RefineUniform refineOptions Allocate and initialize the vertex primvar data see tutorial 2 for more details. std vector Vertex vbuffer refiner - GetNumVerticesTotal Vertex verts vbuffer 0 for int i 0 i g_nverts i verts i . SetPosition g_verts i 0 , g_verts i 1 , g_verts i 2 Allocate initialize the first channel of face-varying primvars UVs std vector FVarVertexUV fvBufferUV refiner - GetNumFVarValuesTotal channelUV FVarVertexUV fvVertsUV fvBufferUV 0 for int i 0 i g_nuvs i fvVertsUV i . u g_uvs i 0 fvVertsUV i . v g_uvs i 1 Allocate interpolate the face-varying primvar data colors std vector FVarVertexColor fvBufferColor refiner - GetNumFVarValuesTotal channelColor FVarVertexColor fvVertsColor fvBufferColor 0 for int i 0 i g_ncolors i fvVertsColor i . r g_colors i 0 fvVertsColor i . g g_colors i 1 fvVertsColor i . b g_colors i 2 fvVertsColor i . a g_colors i 3 Interpolate both vertex and face-varying primvar data Far PrimvarRefiner primvarRefiner refiner Vertex srcVert verts FVarVertexUV srcFVarUV fvVertsUV FVarVertexColor srcFVarColor fvVertsColor for int level 1 level maxlevel level Vertex dstVert srcVert refiner - GetLevel level -1 . GetNumVertices FVarVertexUV dstFVarUV srcFVarUV refiner - GetLevel level -1 . GetNumFVarValues channelUV FVarVertexColor dstFVarColor srcFVarColor refiner - GetLevel level -1 . GetNumFVarValues channelColor primvarRefiner . Interpolate level , srcVert , dstVert primvarRefiner . InterpolateFaceVarying level , srcFVarUV , dstFVarUV , channelUV primvarRefiner . InterpolateFaceVarying level , srcFVarColor , dstFVarColor , channelColor srcVert dstVert srcFVarUV dstFVarUV srcFVarColor dstFVarColor Approximate normals Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int nverts refLastLevel . GetNumVertices int nfaces refLastLevel . GetNumFaces int firstOfLastVerts refiner - GetNumVerticesTotal - nverts std vector Vertex normals nverts Different ways to approximate smooth normals For details check the description at the beginning of the file if normalApproximation Limit Approximation using the normal at the limit with verts that are not at the limit For details check the description at the beginning of the file std vector Vertex fineLimitPos nverts std vector Vertex fineDu nverts std vector Vertex fineDv nverts primvarRefiner . Limit verts firstOfLastVerts , fineLimitPos , fineDu , fineDv for int vert 0 vert nverts vert float const du fineDu vert . GetPosition float const dv fineDv vert . GetPosition float norm 3 cross du , dv , norm normals vert . SetPosition norm 0 , norm 1 , norm 2 else if normalApproximation CrossQuad Approximate smooth normals by accumulating normal vectors computed as the cross product of two vectors generated by the 4 verts that form each quad For details check the description at the beginning of the file for int f 0 f nfaces f Far ConstIndexArray faceVertices refLastLevel . GetFaceVertices f We will use the first three verts to calculate a normal const float v0 verts firstOfLastVerts faceVertices 0 . GetPosition const float v1 verts firstOfLastVerts faceVertices 1 . GetPosition const float v2 verts firstOfLastVerts faceVertices 2 . GetPosition const float v3 verts firstOfLastVerts faceVertices 3 . GetPosition Calculate the cross product between the vectors formed by v1-v0 and v2-v0, and then normalize the result float normalCalculated 0.0 , 0.0 , 0.0 float a 3 v2 0 - v0 0 , v2 1 - v0 1 , v2 2 - v0 2 float b 3 v3 0 - v1 0 , v3 1 - v1 1 , v3 2 - v1 2 cross a , b , normalCalculated normalize normalCalculated Accumulate that normal on all verts that are part of that face for int vInFace 0 vInFace faceVertices . size vInFace int vertexIndex faceVertices vInFace normals vertexIndex . position 0 normalCalculated 0 normals vertexIndex . position 1 normalCalculated 1 normals vertexIndex . position 2 normalCalculated 2 else if normalApproximation CrossTriangle Approximate smooth normals by accumulating normal vectors computed as the cross product of two vectors generated by 3 verts of the quad For details check the description at the beginning of the file for int f 0 f nfaces f Far ConstIndexArray faceVertices refLastLevel . GetFaceVertices f We will use the first three verts to calculate a normal const float v0 verts firstOfLastVerts faceVertices 0 . GetPosition const float v1 verts firstOfLastVerts faceVertices 1 . GetPosition const float v2 verts firstOfLastVerts faceVertices 2 . GetPosition Calculate the cross product between the vectors formed by v1-v0 and v2-v0, and then normalize the result float normalCalculated 0.0 , 0.0 , 0.0 float a 3 v1 0 - v0 0 , v1 1 - v0 1 , v1 2 - v0 2 float b 3 v2 0 - v0 0 , v2 1 - v0 1 , v2 2 - v0 2 cross a , b , normalCalculated normalize normalCalculated Accumulate that normal on all verts that are part of that face for int vInFace 0 vInFace faceVertices . size vInFace int vertexIndex faceVertices vInFace normals vertexIndex . position 0 normalCalculated 0 normals vertexIndex . position 1 normalCalculated 1 normals vertexIndex . position 2 normalCalculated 2 Finally we just need to normalize the accumulated normals for int vert 0 vert nverts vert normalize normals vert . position 0 Output OBJ of the highest level refined ----------- Print vertex positions for int vert 0 vert nverts vert float const pos verts firstOfLastVerts vert . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 Print vertex normals for int vert 0 vert nverts vert float const pos normals vert . GetPosition printf vn f f f n , pos 0 , pos 1 , pos 2 Print uvs int nuvs refLastLevel . GetNumFVarValues channelUV int firstOfLastUvs refiner - GetNumFVarValuesTotal channelUV - nuvs for int fvvert 0 fvvert nuvs fvvert FVarVertexUV const uv fvVertsUV firstOfLastUvs fvvert printf vt f f n , uv . u , uv . v Print faces for int face 0 face nfaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face Far ConstIndexArray fuvs refLastLevel . GetFaceFVarValues face , channelUV all refined Catmark faces should be quads assert fverts . size 4 fuvs . size 4 printf f for int vert 0 vert fverts . size vert OBJ uses 1-based arrays... printf ddd , fverts vert 1 , fuvs vert 1 , fverts vert 1 printf n delete refiner return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"far_tutorial_2_1.html\"\n+ \"loc\": \"far_tutorial_2_3.html\"\n }, {\n \"title\": \"far_tutorial_1_2.cpp\",\n- \"text\": \"far_tutorial_1_2.cpp far_tutorial_1_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_1_2far_tutorial_1_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial illustrates two different styles of defining classes for interpolating primvar data with the template methods in Far. The most common usage involves data of a fixed size, so the focus here is on an alternative supporting variable length data. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include cstdio using namespace OpenSubdiv Vertex data containers for interpolation - Coord3 is fixed to support 3 floats - Coord2 is fixed to support 2 floats - CoordBuffer can support a specified number of floats struct Coord3 Coord3 Coord3 float x , float y , float z _xyz 0 x , _xyz 1 y , _xyz 2 z void Clear _xyz 0 _xyz 1 _xyz 2 0.0f void AddWithWeight Coord3 const src , float weight _xyz 0 weight src . _xyz 0 _xyz 1 weight src . _xyz 1 _xyz 2 weight src . _xyz 2 float const Coords const return _xyz 0 private float _xyz 3 struct Coord2 Coord2 Coord2 float u , float v _uv 0 u , _uv 1 v void Clear _uv 0 _uv 1 0.0f void AddWithWeight Coord2 const src , float weight _uv 0 weight src . _uv 0 _uv 1 weight src . _uv 1 float const Coords const return _uv 0 private float _uv 2 struct CoordBuffer The head of an external buffer and stride is specified on construction CoordBuffer float data , int size _data data , _size size CoordBuffer _data 0 , _size 0 void Clear for int i 0 i _size i _data i 0.0f void AddWithWeight CoordBuffer const src , float weight assert src . _size _size for int i 0 i _size i _data i weight src . _data i float const Coords const return _data Defining to return a location elsewhere in the buffer is the key requirement to supporting interpolatible data of varying size CoordBuffer operator int index const return CoordBuffer _data index _size , _size private float _data int _size Global cube geometry from catmark_cube.h Topology static int g_nverts 8 static int g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 Primvar data static float g_verts 8 3 0.0f , 0.0f , 1.0f , 1.0f , 0.0f , 1.0f , 0.0f , 1.0f , 1.0f , 1.0f , 1.0f , 1.0f , 0.0f , 1.0f , 0.0f , 1.0f , 1.0f , 0.0f , 0.0f , 0.0f , 0.0f , 1.0f , 0.0f , 0.0f Creates FarTopologyRefiner from raw geometry above see tutorial_1_1 for more details static Far TopologyRefiner createFarTopologyRefiner typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options return refiner Overview of main - create a FarTopologyRefiner and uniformly refine it - allocate separate and combined data buffers for vertex positions and UVs - populate all refined data buffers and compare results - write the result in Obj format Disable warnings for exact floating point comparisons ifdef __INTEL_COMPILER pragma warning disable 1572 endif int main int , char Instantiate a FarTopologyRefiner from the global geometry Far TopologyRefiner refiner createFarTopologyRefiner Uniformly refine the topology up to maxlevel int maxlevel 2 refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Allocate and populate data buffers for vertex primvar data -- positions and UVs. We assign UV coordiantes by simply projectingassigning XY values. The position and UV buffers use their associated data types, while the combined buffer uses 5 floats per vertex. int numBaseVertices g_nverts int numTotalVertices refiner - GetNumVerticesTotal std vector Coord3 posData numTotalVertices std vector Coord2 uvData numTotalVertices int combinedStride 3 2 std vector float combinedData numTotalVertices combinedStride for int i 0 i numBaseVertices i posData i Coord3 g_verts i 0 , g_verts i 1 , g_verts i 2 uvData i Coord2 g_verts i 0 , g_verts i 1 float coordCombined combinedData i combinedStride coordCombined 0 g_verts i 0 coordCombined 1 g_verts i 1 coordCombined 2 g_verts i 2 coordCombined 3 g_verts i 0 coordCombined 4 g_verts i 1 Interpolate vertex primvar data Far PrimvarRefiner primvarRefiner refiner Coord3 posSrc posData 0 Coord2 uvSrc uvData 0 CoordBuffer combinedSrc combinedData 0 , combinedStride for int level 1 level maxlevel level int numLevelVerts refiner - GetLevel level -1 . GetNumVertices Coord3 posDst posSrc numLevelVerts Coord2 uvDst uvSrc numLevelVerts CoordBuffer combinedDst combinedSrc numLevelVerts primvarRefiner . Interpolate level , posSrc , posDst primvarRefiner . Interpolate level , uvSrc , uvDst primvarRefiner . Interpolate level , combinedSrc , combinedDst posSrc posDst uvSrc uvDst combinedSrc combinedDst Verify that the combined coords match the separate results for int i numBaseVertices i numTotalVertices i float const posCoords posData i . Coords float const uvCoords uvData i . Coords float const combCoords combinedData combinedStride i assert combCoords 0 posCoords 0 assert combCoords 1 posCoords 1 assert combCoords 2 posCoords 2 assert combCoords 3 uvCoords 0 assert combCoords 4 uvCoords 1 Output OBJ of the highest level refined Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int firstOfLastVerts numTotalVertices - refLastLevel . GetNumVertices Print vertex positions printf Vertices n for int vert firstOfLastVerts vert numTotalVertices vert float const pos combinedData vert combinedStride printf v f f f n , pos 0 , pos 1 , pos 2 printf UV coordinates n for int vert firstOfLastVerts vert numTotalVertices vert float const uv combinedData vert combinedStride 3 printf vt f f n , uv 0 , uv 1 Print faces int numFaces refLastLevel . GetNumFaces printf Faces n for int face 0 face numFaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face printf f for int fvert 0 fvert fverts . size fvert int objIndex 1 fverts fvert OBJ uses 1-based arrays... printf dd , objIndex , objIndex printf n delete refiner return EXIT_SUCCESS Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"far_tutorial_1_2.cpp far_tutorial_1_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_1_2far_tutorial_1_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial illustrates two different styles of defining classes for interpolating primvar data with the template methods in Far. The most common usage involves data of a fixed size, so the focus here is on an alternative supporting variable length data. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include cstdio using namespace OpenSubdiv Vertex data containers for interpolation - Coord3 is fixed to support 3 floats - Coord2 is fixed to support 2 floats - CoordBuffer can support a specified number of floats struct Coord3 Coord3 Coord3 float x , float y , float z _xyz 0 x , _xyz 1 y , _xyz 2 z void Clear _xyz 0 _xyz 1 _xyz 2 0.0f void AddWithWeight Coord3 const src , float weight _xyz 0 weight src . _xyz 0 _xyz 1 weight src . _xyz 1 _xyz 2 weight src . _xyz 2 float const Coords const return _xyz 0 private float _xyz 3 struct Coord2 Coord2 Coord2 float u , float v _uv 0 u , _uv 1 v void Clear _uv 0 _uv 1 0.0f void AddWithWeight Coord2 const src , float weight _uv 0 weight src . _uv 0 _uv 1 weight src . _uv 1 float const Coords const return _uv 0 private float _uv 2 struct CoordBuffer The head of an external buffer and stride is specified on construction CoordBuffer float data , int size _data data , _size size CoordBuffer _data 0 , _size 0 void Clear for int i 0 i _size i _data i 0.0f void AddWithWeight CoordBuffer const src , float weight assert src . _size _size for int i 0 i _size i _data i weight src . _data i float const Coords const return _data Defining to return a location elsewhere in the buffer is the key requirement to supporting interpolatible data of varying size CoordBuffer operator int index const return CoordBuffer _data index _size , _size private float _data int _size Global cube geometry from catmark_cube.h Topology static int g_nverts 8 static int g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 Primvar data static float g_verts 8 3 0.0f , 0.0f , 1.0f , 1.0f , 0.0f , 1.0f , 0.0f , 1.0f , 1.0f , 1.0f , 1.0f , 1.0f , 0.0f , 1.0f , 0.0f , 1.0f , 1.0f , 0.0f , 0.0f , 0.0f , 0.0f , 1.0f , 0.0f , 0.0f Creates FarTopologyRefiner from raw geometry above see tutorial_1_1 for more details static Far TopologyRefiner createFarTopologyRefiner typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options return refiner Overview of main - create a FarTopologyRefiner and uniformly refine it - allocate separate and combined data buffers for vertex positions and UVs - populate all refined data buffers and compare results - write the result in Obj format Disable warnings for exact floating point comparisons ifdef __INTEL_COMPILER pragma warning disable 1572 endif int main int , char Instantiate a FarTopologyRefiner from the global geometry Far TopologyRefiner refiner createFarTopologyRefiner Uniformly refine the topology up to maxlevel int maxlevel 2 refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Allocate and populate data buffers for vertex primvar data -- positions and UVs. We assign UV coordiantes by simply projectingassigning XY values. The position and UV buffers use their associated data types, while the combined buffer uses 5 floats per vertex. int numBaseVertices g_nverts int numTotalVertices refiner - GetNumVerticesTotal std vector Coord3 posData numTotalVertices std vector Coord2 uvData numTotalVertices int combinedStride 3 2 std vector float combinedData numTotalVertices combinedStride for int i 0 i numBaseVertices i posData i Coord3 g_verts i 0 , g_verts i 1 , g_verts i 2 uvData i Coord2 g_verts i 0 , g_verts i 1 float coordCombined combinedData i combinedStride coordCombined 0 g_verts i 0 coordCombined 1 g_verts i 1 coordCombined 2 g_verts i 2 coordCombined 3 g_verts i 0 coordCombined 4 g_verts i 1 Interpolate vertex primvar data Far PrimvarRefiner primvarRefiner refiner Coord3 posSrc posData 0 Coord2 uvSrc uvData 0 CoordBuffer combinedSrc combinedData 0 , combinedStride for int level 1 level maxlevel level int numLevelVerts refiner - GetLevel level -1 . GetNumVertices Coord3 posDst posSrc numLevelVerts Coord2 uvDst uvSrc numLevelVerts CoordBuffer combinedDst combinedSrc numLevelVerts primvarRefiner . Interpolate level , posSrc , posDst primvarRefiner . Interpolate level , uvSrc , uvDst primvarRefiner . Interpolate level , combinedSrc , combinedDst posSrc posDst uvSrc uvDst combinedSrc combinedDst Verify that the combined coords match the separate results for int i numBaseVertices i numTotalVertices i float const posCoords posData i . Coords float const uvCoords uvData i . Coords float const combCoords combinedData combinedStride i assert combCoords 0 posCoords 0 assert combCoords 1 posCoords 1 assert combCoords 2 posCoords 2 assert combCoords 3 uvCoords 0 assert combCoords 4 uvCoords 1 Output OBJ of the highest level refined Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int firstOfLastVerts numTotalVertices - refLastLevel . GetNumVertices Print vertex positions printf Vertices n for int vert firstOfLastVerts vert numTotalVertices vert float const pos combinedData vert combinedStride printf v f f f n , pos 0 , pos 1 , pos 2 printf UV coordinates n for int vert firstOfLastVerts vert numTotalVertices vert float const uv combinedData vert combinedStride 3 printf vt f f n , uv 0 , uv 1 Print faces int numFaces refLastLevel . GetNumFaces printf Faces n for int face 0 face numFaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face printf f for int fvert 0 fvert fverts . size fvert int objIndex 1 fverts fvert OBJ uses 1-based arrays... printf dd , objIndex , objIndex printf n delete refiner return EXIT_SUCCESS Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"far_tutorial_1_2.html\"\n }, {\n- \"title\": \"hbr_tutorial_2.cpp\",\n- \"text\": \"hbr_tutorial_2.cpp hbr_tutorial_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialshbrtutorial_2hbr_tutorial_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to subdivide uniformly a simple Hbr mesh. We are building upon previous tutorials and assuming a fully instantiated mesh we start with an HbrMesh pointer initialized from the same pyramid shape used in hbr_tutorial_0. We then apply the Refine function sequentially to all the faces in the mesh to generate several levels of uniform subdivision. The resulting data is then dumped to the terminal in Wavefront OBJ format for inspection. include opensubdivhbrmesh.h include opensubdivhbrcatmark.h include cassert include cstdio ------------------------------------------------------------------------------ For this tutorial, we have to flesh out the Vertex class further. Note that now the copy constructor, Clear and AddwithWeight methods have been implemented to interpolate our float3 position data. This vertex specialization pattern leaves client-code free to implement arbitrary vertex primvar data schemes or none at all to conserve efficiency struct Vertex Hbr minimal required interface ---------------------- Vertex Vertex int i Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 void AddVaryingWithWeight Vertex const , float Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 typedef OpenSubdiv HbrMesh Vertex Hmesh typedef OpenSubdiv HbrFace Vertex Hface typedef OpenSubdiv HbrVertex Vertex Hvertex typedef OpenSubdiv HbrHalfedge Vertex Hhalfedge Hmesh createMesh ------------------------------------------------------------------------------ int main int , char Hmesh hmesh createMesh int maxlevel 2 , 2 levels of subdivision firstface 0 , marker to the first face index of level 2 firstvertex 0 marker to the first vertex index of level 2 Refine the mesh to maxlevel for int level 0 level maxlevel level Total number of faces in the mesh, across all levels Note this function iterates over the list of faces and can be slow int nfaces hmesh - GetNumFaces if level maxlevel -1 Save our vertex marker firstvertex hmesh - GetNumVertices Iterate over the faces of the current level of subdivision for int face firstface face nfaces face Hface f hmesh - GetFace face Note hole tags would have to be dealt with here. f - Refine Save our face index marker for the next level firstface nfaces Output OBJ of the highest level refined ----------- Print vertex positions int nverts hmesh - GetNumVertices for int vert firstvertex vert nverts vert float const pos hmesh - GetVertex vert - GetData . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 Print faces for int face firstface face hmesh - GetNumFaces face Hface f hmesh - GetFace face assert f - GetNumVertices 4 printf f for int vert 0 vert 4 vert OBJ uses 1-based arrays printf d , f - GetVertex vert - GetID - firstvertex 1 printf n ------------------------------------------------------------------------------ Creates an Hbr mesh see hbr_tutorial_0 and hbr_tutorial_1 for more details Hmesh createMesh Pyramid geometry from catmark_pyramid.h static float verts 5 3 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 2.0f , 0.0f , 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 0.0f static int nverts 5 , nfaces 5 static int facenverts 5 3 , 3 , 3 , 3 , 4 static int faceverts 16 0 , 1 , 2 , 0 , 2 , 3 , 0 , 3 , 4 , 0 , 4 , 1 , 4 , 3 , 2 , 1 OpenSubdiv HbrCatmarkSubdivision Vertex catmark new OpenSubdiv HbrCatmarkSubdivision Vertex Hmesh hmesh new Hmesh catmark Populate the vertices Vertex v for int i 0 i nverts i v . SetPosition verts i 0 , verts i 1 , verts i 2 hmesh - NewVertex i , v Create the topology int fv faceverts for int i 0 i nfaces i int nv facenverts i bool valid true for int j 0 j nv j Hvertex const origin hmesh - GetVertex fv j , destination hmesh - GetVertex fv j 1 nv Hhalfedge const opposite destination - GetEdge origin Make sure that the vertices exist in the mesh if origin NULL destination NULL printf An edge was specified that connected a nonexistent vertex n valid false break Check for a degenerate edge if origin destination printf An edge was specified that connected a vertex to itself n valid false break Check that no more than 2 faces are adjacent to the edge if opposite opposite - GetOpposite printf A non-manifold edge incident to more than 2 faces was found n valid false break Check that the edge is unique and oriented properly if origin - GetEdge destination printf An edge connecting two vertices was specified more than once. Its likely that an incident face was flipped n valid false break if valid hmesh - NewFace nv , fv , 0 else printf Skipped face d n , i fv nv hmesh - SetInterpolateBoundaryMethod Hmesh k_InterpolateBoundaryEdgeOnly hmesh - Finish return hmesh ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"far_tutorial_2_1.cpp\",\n+ \"text\": \"far_tutorial_2_1.cpp far_tutorial_2_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_2_1far_tutorial_2_1.cpp ------------------------------------------------------------------------------ Tutorial description Building on tutorial 0, this example shows how to instantiate a simple mesh, refine it uniformly and then interpolate additional sets of primvar data. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include cstdio ------------------------------------------------------------------------------ Vertex container implementation. We are adding a per-vertex color attribute to our primvar data. While they are separate properties and exist in separate buffers as when read from an Alembic file they are both of the form float3 and so we can use the same underlying type. While color and position may be the same, well make the color a varying primvar, e.g. it is constrained to being linearly interpolated between vertices, rather than smoothly like position and other vertex data. struct Point3 Minimal required interface ---------------------- Point3 void Clear void 0 _point 0 _point 1 _point 2 0.0f void AddWithWeight Point3 const src , float weight _point 0 weight src . _point 0 _point 1 weight src . _point 1 _point 2 weight src . _point 2 Public interface ------------------------------------ void SetPoint float x , float y , float z _point 0 x _point 1 y _point 2 z const float GetPoint const return _point private float _point 3 typedef Point3 VertexPosition typedef Point3 VertexColor ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 8 3 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f Per-vertex RGB color data static float g_colors 8 3 1.0f , 0.0f , 0.5f , 0.0f , 1.0f , 0.0f , 0.0f , 0.0f , 1.0f , 1.0f , 1.0f , 1.0f , 1.0f , 1.0f , 0.0f , 0.0f , 1.0f , 1.0f , 1.0f , 0.0f , 1.0f , 0.0f , 0.0f , 0.0f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv static Far TopologyRefiner createFarTopologyRefiner ------------------------------------------------------------------------------ int main int , char int maxlevel 5 Far TopologyRefiner refiner createFarTopologyRefiner Uniformly refine the topology up to maxlevel refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Allocate buffers for vertex primvar data. We assume we received the coarse data for the mesh in separate buffers from some other source, e.g. an Alembic file. Meanwhile, we want buffers for the lastfinest subdivision level to persist. We have no interest in the intermediate levels. Determine the sizes for our needs int nCoarseVerts g_nverts int nFineVerts refiner - GetLevel maxlevel . GetNumVertices int nTotalVerts refiner - GetNumVerticesTotal int nTempVerts nTotalVerts - nCoarseVerts - nFineVerts Allocate and initialize the primvar data for the original coarse vertices std vector VertexPosition coarsePosBuffer nCoarseVerts std vector VertexColor coarseClrBuffer nCoarseVerts for int i 0 i nCoarseVerts i coarsePosBuffer i . SetPoint g_verts i 0 , g_verts i 1 , g_verts i 2 coarseClrBuffer i . SetPoint g_colors i 0 , g_colors i 1 , g_colors i 2 Allocate intermediate and final storage to be populated std vector VertexPosition tempPosBuffer nTempVerts std vector VertexPosition finePosBuffer nFineVerts std vector VertexColor tempClrBuffer nTempVerts std vector VertexColor fineClrBuffer nFineVerts Interpolate all primvar data -- separate buffers can be populated on separate threads if desired VertexPosition srcPos coarsePosBuffer 0 VertexPosition dstPos tempPosBuffer 0 VertexColor srcClr coarseClrBuffer 0 VertexColor dstClr tempClrBuffer 0 Far PrimvarRefiner primvarRefiner refiner for int level 1 level maxlevel level primvarRefiner . Interpolate level , srcPos , dstPos primvarRefiner . InterpolateVarying level , srcClr , dstClr srcPos dstPos , dstPos refiner - GetLevel level . GetNumVertices srcClr dstClr , dstClr refiner - GetLevel level . GetNumVertices Interpolate the last level into the separate buffers for our final data primvarRefiner . Interpolate maxlevel , srcPos , finePosBuffer primvarRefiner . InterpolateVarying maxlevel , srcClr , fineClrBuffer Visualization with Maya print a MEL script that generates colored particles at the location of the refined vertices dont forget to turn shading on in the viewport to see the colors int nverts nFineVerts Output particle positions printf particle for int vert 0 vert nverts vert float const pos finePosBuffer vert . GetPoint printf -p f f f n , pos 0 , pos 1 , pos 2 printf n Set particle point size 20 -- very large printf addAttr -is true -ln pointSize -at long -dv 20 particleShape1 n Add per-particle color attribute rgbPP printf addAttr -ln rgbPP -dt vectorArray particleShape1 n Set per-particle color values from our primvar data printf setAttr particleShape1.rgbPP -type vectorArray d , nverts for int vert 0 vert nverts vert float const color fineClrBuffer vert . GetPoint printf f f f n , color 0 , color 1 , color 2 printf n delete refiner return EXIT_SUCCESS ------------------------------------------------------------------------------ Creates FarTopologyRefiner from raw geometry see tutorial_1_1 for more details static Far TopologyRefiner createFarTopologyRefiner Populate a topology descriptor with our raw data typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options return refiner ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"hbr_tutorial_2.html\"\n+ \"loc\": \"far_tutorial_2_1.html\"\n }, {\n \"title\": \"far_tutorial_1_1.cpp\",\n- \"text\": \"far_tutorial_1_1.cpp far_tutorial_1_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_1_1far_tutorial_1_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial presents in a very succinct way the requisite steps to instantiate and refine a mesh with Far from simple topological data. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include cstdio ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 8 3 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv ------------------------------------------------------------------------------ int main int , char Populate a topology descriptor with our raw data typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options int maxlevel 2 Uniformly refine the topology up to maxlevel refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Allocate a buffer for vertex primvar data. The buffer length is set to be the sum of all children vertices up to the highest level of refinement. std vector Vertex vbuffer refiner - GetNumVerticesTotal Vertex verts vbuffer 0 Initialize coarse mesh positions int nCoarseVerts g_nverts for int i 0 i nCoarseVerts i verts i . SetPosition g_verts i 0 , g_verts i 1 , g_verts i 2 Interpolate vertex primvar data Far PrimvarRefiner primvarRefiner refiner Vertex src verts for int level 1 level maxlevel level Vertex dst src refiner - GetLevel level -1 . GetNumVertices primvarRefiner . Interpolate level , src , dst src dst Output OBJ of the highest level refined ----------- Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int nverts refLastLevel . GetNumVertices int nfaces refLastLevel . GetNumFaces Print vertex positions int firstOfLastVerts refiner - GetNumVerticesTotal - nverts for int vert 0 vert nverts vert float const pos verts firstOfLastVerts vert . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 Print faces for int face 0 face nfaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face all refined Catmark faces should be quads assert fverts . size 4 printf f for int vert 0 vert fverts . size vert printf d , fverts vert 1 OBJ uses 1-based arrays... printf n delete refiner return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2024-01-12 1327 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"far_tutorial_1_1.cpp far_tutorial_1_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialsfartutorial_1_1far_tutorial_1_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial presents in a very succinct way the requisite steps to instantiate and refine a mesh with Far from simple topological data. include opensubdivfartopologyDescriptor.h include opensubdivfarprimvarRefiner.h include cstdio ------------------------------------------------------------------------------ Vertex container implementation. struct Vertex Minimal required interface ---------------------- Vertex Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 ------------------------------------------------------------------------------ Cube geometry from catmark_cube.h static float g_verts 8 3 -0.5f , -0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , 0.5f , 0.5f , 0.5f , -0.5f , 0.5f , -0.5f , 0.5f , 0.5f , -0.5f , -0.5f , -0.5f , -0.5f , 0.5f , -0.5f , -0.5f static int g_nverts 8 , g_nfaces 6 static int g_vertsperface 6 4 , 4 , 4 , 4 , 4 , 4 static int g_vertIndices 24 0 , 1 , 3 , 2 , 2 , 3 , 5 , 4 , 4 , 5 , 7 , 6 , 6 , 7 , 1 , 0 , 1 , 7 , 5 , 3 , 6 , 0 , 2 , 4 using namespace OpenSubdiv ------------------------------------------------------------------------------ int main int , char Populate a topology descriptor with our raw data typedef Far TopologyDescriptor Descriptor Sdc SchemeType type OpenSubdiv Sdc SCHEME_CATMARK Sdc Options options options . SetVtxBoundaryInterpolation Sdc Options VTX_BOUNDARY_EDGE_ONLY Descriptor desc desc . numVertices g_nverts desc . numFaces g_nfaces desc . numVertsPerFace g_vertsperface desc . vertIndicesPerFace g_vertIndices Instantiate a FarTopologyRefiner from the descriptor Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc , Far TopologyRefinerFactory Descriptor Options type , options int maxlevel 2 Uniformly refine the topology up to maxlevel refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel Allocate a buffer for vertex primvar data. The buffer length is set to be the sum of all children vertices up to the highest level of refinement. std vector Vertex vbuffer refiner - GetNumVerticesTotal Vertex verts vbuffer 0 Initialize coarse mesh positions int nCoarseVerts g_nverts for int i 0 i nCoarseVerts i verts i . SetPosition g_verts i 0 , g_verts i 1 , g_verts i 2 Interpolate vertex primvar data Far PrimvarRefiner primvarRefiner refiner Vertex src verts for int level 1 level maxlevel level Vertex dst src refiner - GetLevel level -1 . GetNumVertices primvarRefiner . Interpolate level , src , dst src dst Output OBJ of the highest level refined ----------- Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int nverts refLastLevel . GetNumVertices int nfaces refLastLevel . GetNumFaces Print vertex positions int firstOfLastVerts refiner - GetNumVerticesTotal - nverts for int vert 0 vert nverts vert float const pos verts firstOfLastVerts vert . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 Print faces for int face 0 face nfaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face all refined Catmark faces should be quads assert fverts . size 4 printf f for int vert 0 vert fverts . size vert printf d , fverts vert 1 OBJ uses 1-based arrays... printf n delete refiner return EXIT_SUCCESS ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"far_tutorial_1_1.html\"\n }, {\n- \"title\": \"hbr_tutorial_0.cpp\",\n- \"text\": \"hbr_tutorial_0.cpp hbr_tutorial_0.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialshbrtutorial_0hbr_tutorial_0.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial presents in a very succinct way the requisite steps to instantiate an Hbr mesh from simple topological data. include opensubdivhbrmesh.h include opensubdivhbrcatmark.h include cstdio ------------------------------------------------------------------------------ Vertex container implementation. The HbrMeshT class is a templated interface that expects a vertex class to perform interpolation on arbitrary vertex data. For the template specialization of the HbrMesh interface to be met, our Vertex object to implement a minimal set of constructors and member functions. Since we are not going to subdivide the mesh, the struct presented here has been left minimalistic. The only customization added to our container was to provide storage and accessors for the position of a 3D vertex. struct Vertex Hbr minimal required interface ---------------------- Vertex Vertex int i Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 void AddWithWeight Vertex const , float void AddVaryingWithWeight Vertex const , float Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 typedef OpenSubdiv HbrMesh Vertex Hmesh typedef OpenSubdiv HbrFace Vertex Hface typedef OpenSubdiv HbrVertex Vertex Hvertex typedef OpenSubdiv HbrHalfedge Vertex Hhalfedge ------------------------------------------------------------------------------ Pyramid geometry from catmark_pyramid.h static float verts 5 3 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 2.0f , 0.0f , 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 0.0f static int nverts 5 , nfaces 5 static int facenverts 5 3 , 3 , 3 , 3 , 4 static int faceverts 16 0 , 1 , 2 , 0 , 2 , 3 , 0 , 3 , 4 , 0 , 4 , 1 , 4 , 3 , 2 , 1 ------------------------------------------------------------------------------ int main int , char Create a subdivision scheme Catmull-Clark here OpenSubdiv HbrCatmarkSubdivision Vertex catmark new OpenSubdiv HbrCatmarkSubdivision Vertex Create an empty Hbr mesh Hmesh hmesh new Hmesh catmark Populate the vertices Vertex v for int i 0 i nverts i Primitive variable data must be set here in our case we set the 3D position of the vertex. v . SetPosition verts i 0 , verts i 1 , verts i 2 Add the vertex to the mesh. hmesh - NewVertex i , v Create the topology int fv faceverts for int i 0 i nfaces i int nv facenverts i hmesh - NewFace nv , fv , 0 fv nv Set subdivision options By default vertex interpolation is set to none on boundaries, which can produce un-expected results, so we change it to edge-only. hmesh - SetInterpolateBoundaryMethod Hmesh k_InterpolateBoundaryEdgeOnly Call Finish to finalize the data structures before using the mesh. hmesh - Finish printf Created a pyramid with d faces and d vertices. n , hmesh - GetNumFaces , hmesh - GetNumVertices delete hmesh delete catmark ------------------------------------------------------------------------------ Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"hbr_tutorial_2.cpp\",\n+ \"text\": \"hbr_tutorial_2.cpp hbr_tutorial_2.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialshbrtutorial_2hbr_tutorial_2.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to subdivide uniformly a simple Hbr mesh. We are building upon previous tutorials and assuming a fully instantiated mesh we start with an HbrMesh pointer initialized from the same pyramid shape used in hbr_tutorial_0. We then apply the Refine function sequentially to all the faces in the mesh to generate several levels of uniform subdivision. The resulting data is then dumped to the terminal in Wavefront OBJ format for inspection. include opensubdivhbrmesh.h include opensubdivhbrcatmark.h include cassert include cstdio ------------------------------------------------------------------------------ For this tutorial, we have to flesh out the Vertex class further. Note that now the copy constructor, Clear and AddwithWeight methods have been implemented to interpolate our float3 position data. This vertex specialization pattern leaves client-code free to implement arbitrary vertex primvar data schemes or none at all to conserve efficiency struct Vertex Hbr minimal required interface ---------------------- Vertex Vertex int i Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 _position 0 _position 1 _position 2 0.0f void AddWithWeight Vertex const src , float weight _position 0 weight src . _position 0 _position 1 weight src . _position 1 _position 2 weight src . _position 2 void AddVaryingWithWeight Vertex const , float Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 typedef OpenSubdiv HbrMesh Vertex Hmesh typedef OpenSubdiv HbrFace Vertex Hface typedef OpenSubdiv HbrVertex Vertex Hvertex typedef OpenSubdiv HbrHalfedge Vertex Hhalfedge Hmesh createMesh ------------------------------------------------------------------------------ int main int , char Hmesh hmesh createMesh int maxlevel 2 , 2 levels of subdivision firstface 0 , marker to the first face index of level 2 firstvertex 0 marker to the first vertex index of level 2 Refine the mesh to maxlevel for int level 0 level maxlevel level Total number of faces in the mesh, across all levels Note this function iterates over the list of faces and can be slow int nfaces hmesh - GetNumFaces if level maxlevel -1 Save our vertex marker firstvertex hmesh - GetNumVertices Iterate over the faces of the current level of subdivision for int face firstface face nfaces face Hface f hmesh - GetFace face Note hole tags would have to be dealt with here. f - Refine Save our face index marker for the next level firstface nfaces Output OBJ of the highest level refined ----------- Print vertex positions int nverts hmesh - GetNumVertices for int vert firstvertex vert nverts vert float const pos hmesh - GetVertex vert - GetData . GetPosition printf v f f f n , pos 0 , pos 1 , pos 2 Print faces for int face firstface face hmesh - GetNumFaces face Hface f hmesh - GetFace face assert f - GetNumVertices 4 printf f for int vert 0 vert 4 vert OBJ uses 1-based arrays printf d , f - GetVertex vert - GetID - firstvertex 1 printf n ------------------------------------------------------------------------------ Creates an Hbr mesh see hbr_tutorial_0 and hbr_tutorial_1 for more details Hmesh createMesh Pyramid geometry from catmark_pyramid.h static float verts 5 3 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 2.0f , 0.0f , 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 0.0f static int nverts 5 , nfaces 5 static int facenverts 5 3 , 3 , 3 , 3 , 4 static int faceverts 16 0 , 1 , 2 , 0 , 2 , 3 , 0 , 3 , 4 , 0 , 4 , 1 , 4 , 3 , 2 , 1 OpenSubdiv HbrCatmarkSubdivision Vertex catmark new OpenSubdiv HbrCatmarkSubdivision Vertex Hmesh hmesh new Hmesh catmark Populate the vertices Vertex v for int i 0 i nverts i v . SetPosition verts i 0 , verts i 1 , verts i 2 hmesh - NewVertex i , v Create the topology int fv faceverts for int i 0 i nfaces i int nv facenverts i bool valid true for int j 0 j nv j Hvertex const origin hmesh - GetVertex fv j , destination hmesh - GetVertex fv j 1 nv Hhalfedge const opposite destination - GetEdge origin Make sure that the vertices exist in the mesh if origin NULL destination NULL printf An edge was specified that connected a nonexistent vertex n valid false break Check for a degenerate edge if origin destination printf An edge was specified that connected a vertex to itself n valid false break Check that no more than 2 faces are adjacent to the edge if opposite opposite - GetOpposite printf A non-manifold edge incident to more than 2 faces was found n valid false break Check that the edge is unique and oriented properly if origin - GetEdge destination printf An edge connecting two vertices was specified more than once. Its likely that an incident face was flipped n valid false break if valid hmesh - NewFace nv , fv , 0 else printf Skipped face d n , i fv nv hmesh - SetInterpolateBoundaryMethod Hmesh k_InterpolateBoundaryEdgeOnly hmesh - Finish return hmesh ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"hbr_tutorial_0.html\"\n+ \"loc\": \"hbr_tutorial_2.html\"\n+ }, {\n+ \"title\": \"Using Hbr\",\n+ \"text\": \"Using Hbr Using Hbr 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Vertex Template API Creating a Mesh Instantiating an HbrMesh Creating Vertices Creating Faces Wrapping Things Up Boundary Interpolation Rules Adding Creases Vertex Creases Edge Creases Holes Hierarchical Edits Face-varying Data Instantiating the HbrMesh Setting the Face-Varying Data Retrieving the Face-Varying Data Valence Operators Managing Singular Vertices Note As of OpenSubdiv 3.0, all Hbr dependencies have been removed from the core APIs Sdc , Vtr , Far , Osd . The legacy source code of Hbr is provided purely for regression and legacy purposes. If your code is currently depending on Hbr functionality, we recommend migrating to the newer APIs as we cannot guarantee that this code will be maintained in future releases. For more information see the 3.0 release notes Vertex Template API The Hbr API abstracts the vertex class through templating. Client-code is expected to provide a vertex class that implements the requisite interpolation functionality. Here is an example of a simple vertex class that accounts for 3D position, but does not support arbitrary variables or varying interpolation. struct Vertex Vertex Vertex int i Vertex const Vertex src _pos 0 src . _pos 0 _pos 1 src . _pos 1 _pos 2 src . _pos 2 Vertex void AddWithWeight const Vertex src , float weight , void 0 _pos 0 weight src . _pos 0 _pos 1 weight src . _pos 1 _pos 2 weight src . _pos 2 void AddVaryingWithWeight const Vertex , float , void 0 void Clear void 0 _pos 0 _pos 1 _pos 2 0.0f void SetPosition float x , float y , float z _pos 0 x _pos 1 y _pos 2 z void ApplyVertexEdit const OpenSubdiv HbrVertexEdit Vertex edit const float src edit . GetEdit switch edit . GetOperation case OpenSubdiv HbrHierarchicalEdit Vertex Set _pos 0 src 0 _pos 1 src 1 _pos 2 src 2 break case OpenSubdiv HbrHierarchicalEdit Vertex Add _pos 0 src 0 _pos 1 src 1 _pos 2 src 2 break case OpenSubdiv HbrHierarchicalEdit Vertex Subtract _pos 0 - src 0 _pos 1 - src 1 _pos 2 - src 2 break void ApplyMovingVertexEdit const OpenSubdiv HbrMovingVertexEdit Vertex custom functions data not required by Hbr ------------------------- Vertex float x , float y , float z _pos 0 x _pos 1 y _pos 2 z const float GetPos const return _pos float _pos 3 In some cases, if only topological analysis is required, the class can be left un-implemented. Far and Osd for instance store vertex data in serialized interleaved vectors. Here is the OsdVertex class for reference class Vertex public Vertex Vertex int index Vertex Vertex const src void AddWithWeight Vertex const i , float weight , void 0 void AddVaryingWithWeight const Vertex i , float weight , void 0 void Clear void 0 void ApplyVertexEdit FarVertexEdit const Creating a Mesh The following tutorial walks through the steps of instantiating a simple Hbr mesh. The code found in regressioncommonshape_utils.h can also be used as an example. While this implementation covers many of Hbr s features, it does not provide coverage for the complete Renderman specification though. Instantiating an HbrMesh First we need to instantiate a mesh object. Hbr supports 3 subdivision schemes Catmull-Clark catmark Loop Bilinear The scheme is selected by passing an specialized instance of HbrSubdivisionT , HbrCatmarkSubdivisionT in this case. The scheme can be shared across multiple mesh objects, so we only need a single instance. static OpenSubdiv HbrCatmarkSubdivision Vertex _scheme OpenSubdiv HbrMesh Vertex mesh new OpenSubdiv HbrMesh Vertex _scheme Creating Vertices Adding vertices to the mesh is accomplished using the HbrMeshNewVertex method. Because Hbr uses a dedicated vertex allocator to help alleviate the performance impact of intensive fragmented memory allocations. This optimization results in the following design pattern Vertex vtx for int i 0 i numVerts i Vertex v mesh - NewVertex i , vtx v-SetPosition We instantiate a single default vertex object named vtx on the stack. We then recover the pointer to the actual vertex created in the mesh from the NewVertex method. Once we have recovered that pointer, we can set the data for our vertex by using any of the custom accessors. Creating Faces Once all the vertices have been registered in the mesh, we can start adding the faces with HbrMeshNewFace . Assuming we had an obj style reader, we need to know the number of vertices in the face and the indices of these vertices. for int f 0 f numFaces f int nverts obj - GetNumVertices f const int faceverts obj - GetFaceVerts f mesh - NewFace nv , fv , 0 However, Hbr is not able to handle non-manifold geometry. In order to avoid tripping asserts or causing memory violations, lets rewrite the previous loop with some some prototype code to check the validity of the topology. for int f 0 f numFaces f int nv obj - GetNumVertices f const int fv obj - GetFaceVerts f triangles only for Loop subdivision if scheme kLoop and nv 3 printf Trying to create a Loop subd with non-triangle face n continue now check the half-edges connectivity for int j 0 j nv j OpenSubdiv HbrVertex T origin mesh - GetVertex fv j OpenSubdiv HbrVertex T destination mesh - GetVertex fv j 1 nv OpenSubdiv HbrHalfedge T opposite destination - GetEdge origin if origin NULL destination NULL printf An edge was specified that connected a nonexistent vertex n continue if origin destination printf An edge was specified that connected a vertex to itself n continue if opposite opposite - GetOpposite printf A non-manifold edge incident to more than 2 faces was found n continue if origin - GetEdge destination printf An edge connecting two vertices was specified more than once. Its likely that an incident face was flipped n continue mesh - NewFace nv , fv , 0 Wrapping Things Up Once we have vertices and faces set in our mesh, we still need to wrap things up by calling HbrMeshFinish mesh - Finish Finish iterates over the mesh to apply the boundary interpolation rules and checks for singular vertices. At this point, there is one final topology check remaining to validate the mesh mesh - Finish if mesh - GetNumDisconnectedVertices printf The specified subdivmesh contains disconnected surface components. n abort or iterate over the mesh to remove the offending vertices Boundary Interpolation Rules The rule-set can be selected using the following accessors Vertex and varying data mesh - SetInterpolateBoundaryMethod OpenSubdiv HbrMesh Vertex k_InterpolateBoundaryEdgeOnly Face-varying data mesh - SetFVarInterpolateBoundaryMethod OpenSubdiv HbrMesh Vertex k_InterpolateBoundaryEdgeOnly Additional information on boundary interpolation rules can be found here Warning The boundary interpolation rules must be set before the call to HbrMeshFinish , which sets the sharpness values to boundary edges and vertices based on these rules. Adding Creases Hbr supports a sharpness attribute on both edges and vertices. Sharpness is set using the SetSharpnessfloat accessors. Vertex Creases Given an index, vertices are very easy to access in the mesh. int idx vertex index float sharp the edge sharpness OpenSubdiv HbrVertex Vertex v mesh - GetVertex idx if v v - SetSharpness std max 0.0f , sharp else printf cannot find vertex for corner tag d n , idx Edge Creases Usually, edge creases are described with a vertex indices pair. Here is some sample code to locate the matching half-edge and set a crease sharpness. int v0 , v1 vertex indices float sharp the edge sharpness OpenSubdiv HbrVertex Vertex v mesh - GetVertex v0 , w mesh - GetVertex v1 OpenSubdiv HbrHalfedge Vertex e 0 if v w if e v - GetEdge w 0 e w - GetEdge v if e e - SetSharpness std max 0.0f , sharp else printf cannot find edge for crease tag d,d n , v0 , v1 Holes Hbr faces support a hole tag. int idx the face index OpenSubdiv HbrFace Vertex f mesh - GetFace idx if f f - SetHole else printf cannot find face for hole tag d n , idx Note The hole tag is hierarchical sub-faces can also be marked as holes. See Hierarchical Edits Hierarchical Edits Hbr supports the following types of hierarchical edits Type Function Corner edits Modify vertex sharpness Crease edits Modify edge sharpness FaceEdit Modify custom face data FVarEdit Modify face-varying data VertexEdit Modify vertex and varying data HoleEdit Set hole tag Modifications are one of the following 3 operations Operation Set Add Subtract Here is a simple example that creates a hierarchical vertex edit. path 655, 2, 3, 0 int faceid 655 , nsubfaces 2 , subfaces 2 2 , 3 , vertexid 0 int offset 0 , offset to the vertex or varying data numElems 3 number of elements to apply the modifier to x,y,z 3 bool isP false shortcut to identify modifications to the vertex position P OpenSubdiv HbrHierarchicalEdit Vertex Operation op OpenSubdiv HbrHierarchicalEdit T Set float values 3 1.0f , 0.5f , 0.0f edit values OpenSubdiv HbrVertexEdit T edit new OpenSubdiv HbrVertexEdit T faceid , nsubfaces , subfaces , vertexid , offset , floatwidth , isP , op , values Face-varying Data Here is a walk-through of how to store face-varying data for a u,v pair. Unlike vertex and varying data which is accessed through the templated vertex API, face-varying data is directly aggregated as vectors of float data. Instantiating the HbrMesh The HbrMesh needs to retain some knowledge about the face-varying data that it carries in order to refine it correctly. int fvarwidth 2 total width of the fvar data static int indices 1 0 , 1 offset set to 0 widths 1 2 2 floats in a u,v pair int const fvarcount fvarwidth 0 1 0 , fvarindices fvarwidth 0 indices NULL , fvarwidths fvarwidth 0 widths NULL mesh new OpenSubdiv HbrMesh T _scheme , fvarcount , fvarindices , fvarwidths , fvarwidth Setting the Face-Varying Data After the topology has been created, Hbr is ready to accept face-varying data. Here is some sample code for int i 0 , idx 0 i numFaces i OpenSubdiv HbrFace Vertex f mesh - GetFace i int nv f - GetNumVertices note this is not the fastest way OpenSubdiv HbrHalfedge Vertex e f - GetFirstEdge for int j 0 j nv j , e e - GetNext OpenSubdiv HbrFVarData Vertex fvt e - GetOrgVertex - GetFVarData f float const fvdata GetFaceVaryingData i , j if not fvt . IsInitialized if no fvar daa exists yet on the vertex fvt . SetAllData 2 , fvdata else if not fvt . CompareAll 2 , fvdata if there already is fvar data and there is a boundary add the new data OpenSubdiv HbrFVarData T nfvt e - GetOrgVertex - NewFVarData f nfvt . SetAllData 2 , fvdata Retrieving the Face-Varying Data The HbrFVarData structures are expanded during the refinement process, with every sub-face being assigned a set of interpolated face-varying data. This data can be accessed in 2 ways From a face, passing a vertex index OpenSubdivHbrFaceVertex f OpenSubdiv HbrFVarData const fv f . GetFVarData vindex const float data fv . GetData From a vertex, passing a pointer to an incident face OpenSubdivHbrFaceVertex f OpenSubdiv HbrFVarData const fv myVertex . GetFVarData f const float data fv . GetData Valence Operators When manipulating meshes, it is often necessary to iterate over neighboring faces or vertices. Rather than gather lists of pointers and return them, Hbr exposes an operator pattern that guarantees consistent mesh traversals. The following example shows how to use an operator to count the number of neighboring vertices use HbrVertexGetValence for proper valence counts OpenSubdivHbrVertexVertex v class MyOperator public OpenSubdiv HbrVertexOperator Vertex public int count MyOperator count 0 virtual void operator OpenSubdiv HbrVertex Vertex v count MyOperator op v - ApplyOperatorSurroundingVertices op Managing Singular Vertices Certain topological configurations would force vertices to share multiple half-edge cycles. Because Hbr is a half-edge representation, these singular vertices have to be duplicated as part of the HbrMeshFinish phase of the instantiation. These duplicated vertices can cause problems for client-code that tries to populate buffers of vertex or varying data. The following sample code shows how to match the vertex data to singular vertex splits Populating an OsdCpuVertexBuffer with vertex data positions,... float const vtxData inMeshFn . getRawPoints returnStatus OpenSubdiv OsdCpuVertexBuffer vertexBuffer OpenSubdiv OsdCpuVertexBuffer Create numVertexElements , numFarVerts vertexBuffer - UpdateData vtxData , 0 , numVertices Duplicate the vertex data into the split singular vertices std vector std pair int , int const splits hbrMesh - GetSplitVertices for int i 0 i int splits . size i vertexBuffer - UpdateData vtxData splits i . second numVertexElements , splits i . first , 1 Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"tags\": \"\",\n+ \"loc\": \"using_osd_hbr.html\"\n }, {\n \"title\": \"hbr_tutorial_1.cpp\",\n- \"text\": \"hbr_tutorial_1.cpp hbr_tutorial_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialshbrtutorial_1hbr_tutorial_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to safely create Hbr meshes from arbitrary topology. Because Hbr is a half-edge data structure, it cannot represent non-manifold topology. Ensuring that the geometry used is manifold is a requirement to use Hbr safely. This tutorial presents some simple tests to detect inappropriate topology. include opensubdivhbrmesh.h include opensubdivhbrcatmark.h include cstdio ------------------------------------------------------------------------------ struct Vertex Hbr minimal required interface ---------------------- Vertex Vertex int i Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 void AddWithWeight Vertex const , float void AddVaryingWithWeight Vertex const , float Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 typedef OpenSubdiv HbrMesh Vertex Hmesh typedef OpenSubdiv HbrFace Vertex Hface typedef OpenSubdiv HbrVertex Vertex Hvertex typedef OpenSubdiv HbrHalfedge Vertex Hhalfedge ------------------------------------------------------------------------------ Non-manifold geometry from catmark_fan.h o o f2 o------------o------------o f0 f1 o------------ o------------o The shared edge of a fan is adjacent to 3 faces, and therefore non-manifold. static float verts 8 3 -1.0 , 0.0 , -1.0 , -1.0 , 0.0 , 0.0 , 0.0 , 0.0 , 0.0 , 0.0 , 0.0 , -1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 0.0 , -1.0 , 0.0 , 1.0 , 0.0 , 0.0 , 1.0 , -1.0 static int nverts 8 , nfaces 3 static int facenverts 3 4 , 4 , 4 static int faceverts 12 0 , 1 , 2 , 3 , 3 , 2 , 4 , 5 , 3 , 2 , 6 , 7 ------------------------------------------------------------------------------ int main int , char OpenSubdiv HbrCatmarkSubdivision Vertex catmark new OpenSubdiv HbrCatmarkSubdivision Vertex Hmesh hmesh new Hmesh catmark Vertex v for int i 0 i nverts i v . SetPosition verts i 0 , verts i 1 , verts i 2 hmesh - NewVertex i , v Create the topology int fv faceverts for int i 0 i nfaces i int nv facenverts i bool valid true for int j 0 j nv j Hvertex const origin hmesh - GetVertex fv j , destination hmesh - GetVertex fv j 1 nv Hhalfedge const opposite destination - GetEdge origin Make sure that the vertices exist in the mesh if origin NULL destination NULL printf An edge was specified that connected a nonexistent vertex n valid false break Check for a degenerate edge if origin destination printf An edge was specified that connected a vertex to itself n valid false break Check that no more than 2 faces are adjacent to the edge if opposite opposite - GetOpposite printf A non-manifold edge incident to more than 2 faces was found n valid false break Check that the edge is unique and oriented properly if origin - GetEdge destination printf An edge connecting two vertices was specified more than once. Its likely that an incident face was flipped n valid false break if valid hmesh - NewFace nv , fv , 0 else printf Skipped face d n , i fv nv hmesh - SetInterpolateBoundaryMethod Hmesh k_InterpolateBoundaryEdgeOnly hmesh - Finish printf Created a fan with d faces and d vertices. n , hmesh - GetNumFaces , hmesh - GetNumVertices delete hmesh delete catmark ------------------------------------------------------------------------------ Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"hbr_tutorial_1.cpp hbr_tutorial_1.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialshbrtutorial_1hbr_tutorial_1.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial shows how to safely create Hbr meshes from arbitrary topology. Because Hbr is a half-edge data structure, it cannot represent non-manifold topology. Ensuring that the geometry used is manifold is a requirement to use Hbr safely. This tutorial presents some simple tests to detect inappropriate topology. include opensubdivhbrmesh.h include opensubdivhbrcatmark.h include cstdio ------------------------------------------------------------------------------ struct Vertex Hbr minimal required interface ---------------------- Vertex Vertex int i Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 void AddWithWeight Vertex const , float void AddVaryingWithWeight Vertex const , float Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 typedef OpenSubdiv HbrMesh Vertex Hmesh typedef OpenSubdiv HbrFace Vertex Hface typedef OpenSubdiv HbrVertex Vertex Hvertex typedef OpenSubdiv HbrHalfedge Vertex Hhalfedge ------------------------------------------------------------------------------ Non-manifold geometry from catmark_fan.h o o f2 o------------o------------o f0 f1 o------------ o------------o The shared edge of a fan is adjacent to 3 faces, and therefore non-manifold. static float verts 8 3 -1.0 , 0.0 , -1.0 , -1.0 , 0.0 , 0.0 , 0.0 , 0.0 , 0.0 , 0.0 , 0.0 , -1.0 , 1.0 , 0.0 , 0.0 , 1.0 , 0.0 , -1.0 , 0.0 , 1.0 , 0.0 , 0.0 , 1.0 , -1.0 static int nverts 8 , nfaces 3 static int facenverts 3 4 , 4 , 4 static int faceverts 12 0 , 1 , 2 , 3 , 3 , 2 , 4 , 5 , 3 , 2 , 6 , 7 ------------------------------------------------------------------------------ int main int , char OpenSubdiv HbrCatmarkSubdivision Vertex catmark new OpenSubdiv HbrCatmarkSubdivision Vertex Hmesh hmesh new Hmesh catmark Vertex v for int i 0 i nverts i v . SetPosition verts i 0 , verts i 1 , verts i 2 hmesh - NewVertex i , v Create the topology int fv faceverts for int i 0 i nfaces i int nv facenverts i bool valid true for int j 0 j nv j Hvertex const origin hmesh - GetVertex fv j , destination hmesh - GetVertex fv j 1 nv Hhalfedge const opposite destination - GetEdge origin Make sure that the vertices exist in the mesh if origin NULL destination NULL printf An edge was specified that connected a nonexistent vertex n valid false break Check for a degenerate edge if origin destination printf An edge was specified that connected a vertex to itself n valid false break Check that no more than 2 faces are adjacent to the edge if opposite opposite - GetOpposite printf A non-manifold edge incident to more than 2 faces was found n valid false break Check that the edge is unique and oriented properly if origin - GetEdge destination printf An edge connecting two vertices was specified more than once. Its likely that an incident face was flipped n valid false break if valid hmesh - NewFace nv , fv , 0 else printf Skipped face d n , i fv nv hmesh - SetInterpolateBoundaryMethod Hmesh k_InterpolateBoundaryEdgeOnly hmesh - Finish printf Created a fan with d faces and d vertices. n , hmesh - GetNumFaces , hmesh - GetNumVertices delete hmesh delete catmark ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"hbr_tutorial_1.html\"\n }, {\n- \"title\": \"Using Hbr\",\n- \"text\": \"Using Hbr Using Hbr 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Vertex Template API Creating a Mesh Instantiating an HbrMesh Creating Vertices Creating Faces Wrapping Things Up Boundary Interpolation Rules Adding Creases Vertex Creases Edge Creases Holes Hierarchical Edits Face-varying Data Instantiating the HbrMesh Setting the Face-Varying Data Retrieving the Face-Varying Data Valence Operators Managing Singular Vertices Note As of OpenSubdiv 3.0, all Hbr dependencies have been removed from the core APIs Sdc , Vtr , Far , Osd . The legacy source code of Hbr is provided purely for regression and legacy purposes. If your code is currently depending on Hbr functionality, we recommend migrating to the newer APIs as we cannot guarantee that this code will be maintained in future releases. For more information see the 3.0 release notes Vertex Template API The Hbr API abstracts the vertex class through templating. Client-code is expected to provide a vertex class that implements the requisite interpolation functionality. Here is an example of a simple vertex class that accounts for 3D position, but does not support arbitrary variables or varying interpolation. struct Vertex Vertex Vertex int i Vertex const Vertex src _pos 0 src . _pos 0 _pos 1 src . _pos 1 _pos 2 src . _pos 2 Vertex void AddWithWeight const Vertex src , float weight , void 0 _pos 0 weight src . _pos 0 _pos 1 weight src . _pos 1 _pos 2 weight src . _pos 2 void AddVaryingWithWeight const Vertex , float , void 0 void Clear void 0 _pos 0 _pos 1 _pos 2 0.0f void SetPosition float x , float y , float z _pos 0 x _pos 1 y _pos 2 z void ApplyVertexEdit const OpenSubdiv HbrVertexEdit Vertex edit const float src edit . GetEdit switch edit . GetOperation case OpenSubdiv HbrHierarchicalEdit Vertex Set _pos 0 src 0 _pos 1 src 1 _pos 2 src 2 break case OpenSubdiv HbrHierarchicalEdit Vertex Add _pos 0 src 0 _pos 1 src 1 _pos 2 src 2 break case OpenSubdiv HbrHierarchicalEdit Vertex Subtract _pos 0 - src 0 _pos 1 - src 1 _pos 2 - src 2 break void ApplyMovingVertexEdit const OpenSubdiv HbrMovingVertexEdit Vertex custom functions data not required by Hbr ------------------------- Vertex float x , float y , float z _pos 0 x _pos 1 y _pos 2 z const float GetPos const return _pos float _pos 3 In some cases, if only topological analysis is required, the class can be left un-implemented. Far and Osd for instance store vertex data in serialized interleaved vectors. Here is the OsdVertex class for reference class Vertex public Vertex Vertex int index Vertex Vertex const src void AddWithWeight Vertex const i , float weight , void 0 void AddVaryingWithWeight const Vertex i , float weight , void 0 void Clear void 0 void ApplyVertexEdit FarVertexEdit const Creating a Mesh The following tutorial walks through the steps of instantiating a simple Hbr mesh. The code found in regressioncommonshape_utils.h can also be used as an example. While this implementation covers many of Hbr s features, it does not provide coverage for the complete Renderman specification though. Instantiating an HbrMesh First we need to instantiate a mesh object. Hbr supports 3 subdivision schemes Catmull-Clark catmark Loop Bilinear The scheme is selected by passing an specialized instance of HbrSubdivisionT , HbrCatmarkSubdivisionT in this case. The scheme can be shared across multiple mesh objects, so we only need a single instance. static OpenSubdiv HbrCatmarkSubdivision Vertex _scheme OpenSubdiv HbrMesh Vertex mesh new OpenSubdiv HbrMesh Vertex _scheme Creating Vertices Adding vertices to the mesh is accomplished using the HbrMeshNewVertex method. Because Hbr uses a dedicated vertex allocator to help alleviate the performance impact of intensive fragmented memory allocations. This optimization results in the following design pattern Vertex vtx for int i 0 i numVerts i Vertex v mesh - NewVertex i , vtx v-SetPosition We instantiate a single default vertex object named vtx on the stack. We then recover the pointer to the actual vertex created in the mesh from the NewVertex method. Once we have recovered that pointer, we can set the data for our vertex by using any of the custom accessors. Creating Faces Once all the vertices have been registered in the mesh, we can start adding the faces with HbrMeshNewFace . Assuming we had an obj style reader, we need to know the number of vertices in the face and the indices of these vertices. for int f 0 f numFaces f int nverts obj - GetNumVertices f const int faceverts obj - GetFaceVerts f mesh - NewFace nv , fv , 0 However, Hbr is not able to handle non-manifold geometry. In order to avoid tripping asserts or causing memory violations, lets rewrite the previous loop with some some prototype code to check the validity of the topology. for int f 0 f numFaces f int nv obj - GetNumVertices f const int fv obj - GetFaceVerts f triangles only for Loop subdivision if scheme kLoop and nv 3 printf Trying to create a Loop subd with non-triangle face n continue now check the half-edges connectivity for int j 0 j nv j OpenSubdiv HbrVertex T origin mesh - GetVertex fv j OpenSubdiv HbrVertex T destination mesh - GetVertex fv j 1 nv OpenSubdiv HbrHalfedge T opposite destination - GetEdge origin if origin NULL destination NULL printf An edge was specified that connected a nonexistent vertex n continue if origin destination printf An edge was specified that connected a vertex to itself n continue if opposite opposite - GetOpposite printf A non-manifold edge incident to more than 2 faces was found n continue if origin - GetEdge destination printf An edge connecting two vertices was specified more than once. Its likely that an incident face was flipped n continue mesh - NewFace nv , fv , 0 Wrapping Things Up Once we have vertices and faces set in our mesh, we still need to wrap things up by calling HbrMeshFinish mesh - Finish Finish iterates over the mesh to apply the boundary interpolation rules and checks for singular vertices. At this point, there is one final topology check remaining to validate the mesh mesh - Finish if mesh - GetNumDisconnectedVertices printf The specified subdivmesh contains disconnected surface components. n abort or iterate over the mesh to remove the offending vertices Boundary Interpolation Rules The rule-set can be selected using the following accessors Vertex and varying data mesh - SetInterpolateBoundaryMethod OpenSubdiv HbrMesh Vertex k_InterpolateBoundaryEdgeOnly Face-varying data mesh - SetFVarInterpolateBoundaryMethod OpenSubdiv HbrMesh Vertex k_InterpolateBoundaryEdgeOnly Additional information on boundary interpolation rules can be found here Warning The boundary interpolation rules must be set before the call to HbrMeshFinish , which sets the sharpness values to boundary edges and vertices based on these rules. Adding Creases Hbr supports a sharpness attribute on both edges and vertices. Sharpness is set using the SetSharpnessfloat accessors. Vertex Creases Given an index, vertices are very easy to access in the mesh. int idx vertex index float sharp the edge sharpness OpenSubdiv HbrVertex Vertex v mesh - GetVertex idx if v v - SetSharpness std max 0.0f , sharp else printf cannot find vertex for corner tag d n , idx Edge Creases Usually, edge creases are described with a vertex indices pair. Here is some sample code to locate the matching half-edge and set a crease sharpness. int v0 , v1 vertex indices float sharp the edge sharpness OpenSubdiv HbrVertex Vertex v mesh - GetVertex v0 , w mesh - GetVertex v1 OpenSubdiv HbrHalfedge Vertex e 0 if v w if e v - GetEdge w 0 e w - GetEdge v if e e - SetSharpness std max 0.0f , sharp else printf cannot find edge for crease tag d,d n , v0 , v1 Holes Hbr faces support a hole tag. int idx the face index OpenSubdiv HbrFace Vertex f mesh - GetFace idx if f f - SetHole else printf cannot find face for hole tag d n , idx Note The hole tag is hierarchical sub-faces can also be marked as holes. See Hierarchical Edits Hierarchical Edits Hbr supports the following types of hierarchical edits Type Function Corner edits Modify vertex sharpness Crease edits Modify edge sharpness FaceEdit Modify custom face data FVarEdit Modify face-varying data VertexEdit Modify vertex and varying data HoleEdit Set hole tag Modifications are one of the following 3 operations Operation Set Add Subtract Here is a simple example that creates a hierarchical vertex edit. path 655, 2, 3, 0 int faceid 655 , nsubfaces 2 , subfaces 2 2 , 3 , vertexid 0 int offset 0 , offset to the vertex or varying data numElems 3 number of elements to apply the modifier to x,y,z 3 bool isP false shortcut to identify modifications to the vertex position P OpenSubdiv HbrHierarchicalEdit Vertex Operation op OpenSubdiv HbrHierarchicalEdit T Set float values 3 1.0f , 0.5f , 0.0f edit values OpenSubdiv HbrVertexEdit T edit new OpenSubdiv HbrVertexEdit T faceid , nsubfaces , subfaces , vertexid , offset , floatwidth , isP , op , values Face-varying Data Here is a walk-through of how to store face-varying data for a u,v pair. Unlike vertex and varying data which is accessed through the templated vertex API, face-varying data is directly aggregated as vectors of float data. Instantiating the HbrMesh The HbrMesh needs to retain some knowledge about the face-varying data that it carries in order to refine it correctly. int fvarwidth 2 total width of the fvar data static int indices 1 0 , 1 offset set to 0 widths 1 2 2 floats in a u,v pair int const fvarcount fvarwidth 0 1 0 , fvarindices fvarwidth 0 indices NULL , fvarwidths fvarwidth 0 widths NULL mesh new OpenSubdiv HbrMesh T _scheme , fvarcount , fvarindices , fvarwidths , fvarwidth Setting the Face-Varying Data After the topology has been created, Hbr is ready to accept face-varying data. Here is some sample code for int i 0 , idx 0 i numFaces i OpenSubdiv HbrFace Vertex f mesh - GetFace i int nv f - GetNumVertices note this is not the fastest way OpenSubdiv HbrHalfedge Vertex e f - GetFirstEdge for int j 0 j nv j , e e - GetNext OpenSubdiv HbrFVarData Vertex fvt e - GetOrgVertex - GetFVarData f float const fvdata GetFaceVaryingData i , j if not fvt . IsInitialized if no fvar daa exists yet on the vertex fvt . SetAllData 2 , fvdata else if not fvt . CompareAll 2 , fvdata if there already is fvar data and there is a boundary add the new data OpenSubdiv HbrFVarData T nfvt e - GetOrgVertex - NewFVarData f nfvt . SetAllData 2 , fvdata Retrieving the Face-Varying Data The HbrFVarData structures are expanded during the refinement process, with every sub-face being assigned a set of interpolated face-varying data. This data can be accessed in 2 ways From a face, passing a vertex index OpenSubdivHbrFaceVertex f OpenSubdiv HbrFVarData const fv f . GetFVarData vindex const float data fv . GetData From a vertex, passing a pointer to an incident face OpenSubdivHbrFaceVertex f OpenSubdiv HbrFVarData const fv myVertex . GetFVarData f const float data fv . GetData Valence Operators When manipulating meshes, it is often necessary to iterate over neighboring faces or vertices. Rather than gather lists of pointers and return them, Hbr exposes an operator pattern that guarantees consistent mesh traversals. The following example shows how to use an operator to count the number of neighboring vertices use HbrVertexGetValence for proper valence counts OpenSubdivHbrVertexVertex v class MyOperator public OpenSubdiv HbrVertexOperator Vertex public int count MyOperator count 0 virtual void operator OpenSubdiv HbrVertex Vertex v count MyOperator op v - ApplyOperatorSurroundingVertices op Managing Singular Vertices Certain topological configurations would force vertices to share multiple half-edge cycles. Because Hbr is a half-edge representation, these singular vertices have to be duplicated as part of the HbrMeshFinish phase of the instantiation. These duplicated vertices can cause problems for client-code that tries to populate buffers of vertex or varying data. The following sample code shows how to match the vertex data to singular vertex splits Populating an OsdCpuVertexBuffer with vertex data positions,... float const vtxData inMeshFn . getRawPoints returnStatus OpenSubdiv OsdCpuVertexBuffer vertexBuffer OpenSubdiv OsdCpuVertexBuffer Create numVertexElements , numFarVerts vertexBuffer - UpdateData vtxData , 0 , numVertices Duplicate the vertex data into the split singular vertices std vector std pair int , int const splits hbrMesh - GetSplitVertices for int i 0 i int splits . size i vertexBuffer - UpdateData vtxData splits i . second numVertexElements , splits i . first , 1 Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"hbr_tutorial_0.cpp\",\n+ \"text\": \"hbr_tutorial_0.cpp hbr_tutorial_0.cpp 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE httpsgithub.comPixarAnimationStudiosOpenSubdivblobreleasetutorialshbrtutorial_0hbr_tutorial_0.cpp ------------------------------------------------------------------------------ Tutorial description This tutorial presents in a very succinct way the requisite steps to instantiate an Hbr mesh from simple topological data. include opensubdivhbrmesh.h include opensubdivhbrcatmark.h include cstdio ------------------------------------------------------------------------------ Vertex container implementation. The HbrMeshT class is a templated interface that expects a vertex class to perform interpolation on arbitrary vertex data. For the template specialization of the HbrMesh interface to be met, our Vertex object to implement a minimal set of constructors and member functions. Since we are not going to subdivide the mesh, the struct presented here has been left minimalistic. The only customization added to our container was to provide storage and accessors for the position of a 3D vertex. struct Vertex Hbr minimal required interface ---------------------- Vertex Vertex int i Vertex Vertex const src _position 0 src . _position 0 _position 1 src . _position 1 _position 2 src . _position 2 void Clear void 0 void AddWithWeight Vertex const , float void AddVaryingWithWeight Vertex const , float Public interface ------------------------------------ void SetPosition float x , float y , float z _position 0 x _position 1 y _position 2 z const float GetPosition const return _position private float _position 3 typedef OpenSubdiv HbrMesh Vertex Hmesh typedef OpenSubdiv HbrFace Vertex Hface typedef OpenSubdiv HbrVertex Vertex Hvertex typedef OpenSubdiv HbrHalfedge Vertex Hhalfedge ------------------------------------------------------------------------------ Pyramid geometry from catmark_pyramid.h static float verts 5 3 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 2.0f , 0.0f , 0.0f , 0.0f , 2.0f , 0.0f , -2.0f , 0.0f , 0.0f static int nverts 5 , nfaces 5 static int facenverts 5 3 , 3 , 3 , 3 , 4 static int faceverts 16 0 , 1 , 2 , 0 , 2 , 3 , 0 , 3 , 4 , 0 , 4 , 1 , 4 , 3 , 2 , 1 ------------------------------------------------------------------------------ int main int , char Create a subdivision scheme Catmull-Clark here OpenSubdiv HbrCatmarkSubdivision Vertex catmark new OpenSubdiv HbrCatmarkSubdivision Vertex Create an empty Hbr mesh Hmesh hmesh new Hmesh catmark Populate the vertices Vertex v for int i 0 i nverts i Primitive variable data must be set here in our case we set the 3D position of the vertex. v . SetPosition verts i 0 , verts i 1 , verts i 2 Add the vertex to the mesh. hmesh - NewVertex i , v Create the topology int fv faceverts for int i 0 i nfaces i int nv facenverts i hmesh - NewFace nv , fv , 0 fv nv Set subdivision options By default vertex interpolation is set to none on boundaries, which can produce un-expected results, so we change it to edge-only. hmesh - SetInterpolateBoundaryMethod Hmesh k_InterpolateBoundaryEdgeOnly Call Finish to finalize the data structures before using the mesh. hmesh - Finish printf Created a pyramid with d faces and d vertices. n , hmesh - GetNumFaces , hmesh - GetNumVertices delete hmesh delete catmark ------------------------------------------------------------------------------ Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"using_osd_hbr.html\"\n+ \"loc\": \"hbr_tutorial_0.html\"\n }, {\n \"title\": \"Vtr Overview\",\n- \"text\": \"Vtr Overview Vtr Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Vectorized Topology Representation Vtr VtrLevel Topological Relationships Memory Efficiency VtrRefinement Parent-child and child-parent relationships Data flexibility Vectorized Topology Representation Vtr Vtr consists of a suite of classes that collectively provide an intermediate representation of topology that supports efficient refinement. Vtr is intended for internal use only and is currently accessed through the Far layer by the FarTopologyRefiner , which assembles these Vtr classes to meet the topological and refinement needs of the Far layer. What follows is therefore more intended to provide insite into the underlying architecture than to describe particular usage. For documentation more relevant to direct usage, proceed to the Far section previously noted. Vtr is vectorized in that its topological data is stored more as a collection of vectors of primitive elements rather than as the faces, vertices and edges that make up many other topological representations. It is essentially a structure-of-arrays SOA approach to topology in contrast to the more common array-of-structures pattern found in many other topological representations. Vtrs use of vectors allows it to be fairly efficient in its use of memory and similarly efficient to refine, but the topology is fixed once defined. Vtr classes are purely topological. They are even more independent of the representation of vertices, faces, etc. than Hbr in that they are not even parameterized by an interface to such components. So the same set of Vtr objects can eventually be used to serve more than one representation of these components. The primary requirement is that a mesh be expressable as an indexable set i.e. a vector or array of vertices, edges and faces. The index of a component uniquely identifies it and properties are retrieved by referring to it by index. Its worth qualifying the term topological here and elsewhere -- we generally refer to topology as subdivision topology rather than mesh topology. A subdivision hierarchy is impacted by the presence of semi-sharp creasing, as the subdivision rules change in response to that creasing. So subdivision topology includes the sharpness values assigned to edges and vertices that affect the semi-sharp creasing. The two primary classes in Vtr consist of VtrLevel - a class representing complete vertex topology for a level VtrRefinement - a class mapping a parent VtrLevel to a child level Others exist to represent the following selection and appropriate tagging of components for sparse refinement divergence of face-varying topology from the vertex topology mapping between face-varying topology at successive levels common low-level utilities, e.g. simple array classes VtrLevel VtrLevel is a complete topological description of a subdivision level, with the topological relations, sharpness values and component tags all stored in vectors literally stdvectors, but easily changed via typedefs. There are no classes or objects for the mesh component types i.e. faces, edges and vertices but simply an integer index to identify each. It can be viewed as a structure-of-arrays representation of the topology any property related to a particular component is stored in an array and accessible using the index identifying that component. So with no classes the for the components, its difficult to say what constitutes a vertex or a face they are each the sum of all the fields scattered amongst the many vectors included. Level represents a single level of a potential hierarchy and is capable of representing the complete base mesh. There are no members that relate data in one level to any other, either below or above. As such, any Level can be used as the base level for a new subdivision hierarchy potentially more than one. All relationships between separate levels are maintained in the VtrRefinement class. Topological Relationships Level requires the definition of and associations between a fixed set of indexable components for all three component types, i.e. an explicit edge list in addition to the expected set of vertices and faces. There are no explicit component objects in the representation, only an integer index VtrIndex identifying each component within the set and data associated with that component in the various vectors. The topology is stored as six sets of incident relations between the components two each for the two other component types incident each component type, i.e. for each face, its incident vertices and incident edges for each edge, its incident vertices and incident faces for each vertex, its incident edges and incident faces The collection of incidence relations is a vectorized variation of AIF the Adjacency and Incidence Framework. The set of these six incidence relations is not minimal only four are required, but that set excludes the most desired face-vertex relation but all six are kept and maintained to facilitate faster refinement. While the sizes of several vectors are directly proportional to the number of vertices, edges or faces to which the data is associated, the sizes of some of the vectors for these relations is more cumulative and so additional vectors of offsets is required typical of the face-vertex list commonly used as the minimal definition of mesh topology. Vectors for the sharpness values associated with crease edges and corner vertices are included and so sized according to the number of edges and vertices, along with additional tags for the components that may be helpful to refinement i.e. the type of subdivision Rule associated with each vertex. A Level is really just a container for data in a subdivision level, and so its public methods are primarily to access that data. Modification of the data is protected and only made available to classes that are intended to construct Levels currently the Far factory class that is responsible for building the base level, and the VtrRefinement class that constructs subsequent levels during refinement. Memory Efficiency One of the advantages in storing data in what is essentially a structure-of-arrays, rather than the array-of-structures more typical of topological representations, is that we can be more selective about memory usage in some cases. Particularly in the case of uniform refinement, when the data in subsequent levels is typically 4x its predecessor, we can minimize what we either generate or keep around at each level. For instance, if only a face-list is required at the finest level, we only need to generate one of the six topological relations the vertices incident each face. When we do keep Levels around in memory as is the case with the FarTopologyRefiner we do have do have the opportunity to prune what is not strictly necessary after the refinement. Just as with construction, whatever classes are privileged to construct a Level are likely those that will be privileged to prune its contents when needed. The current implementation of Level is far from optimal though -- there are opportunities for improvement. After one level of subdivision, the faces in a Level will be either all quads or tris. Having specializations for these cases and using the more general case in support of N-sided faces for the base level only is one possibility. Levels also allocate dozens of vectors in which to store all data. Since these vectors are of fixed size once created, they could be aggregated by partitioning one or a smaller number of larger block of memory into the desired pieces. The desire to make some of these improvements is part of why Vtr is not directly exposed for public use and instead exposed via Far. VtrRefinement While VtrLevel contains the topology for each subdivision level, VtrRefinement is responsible for creating a new level via refinement of an existing one, and for maintaining the relationships between the components in the parent and child levels. So a simplified view of a subdivision hierarchy with Vtr is a set of Levels with a Refinement between each successive pair. Refinement is a friend of Level and will populate a child level from a parent given a set of refinement parameters. Aside from parameters related to data or depth, there are two kinds of refinement supported uniform and sparse. The latter sparse refinement requires selection of an arbitrary set of components -- any dependent or neighboring components that are required for the limit will be automatically included. So feature-adaptive refinement is just one form of this selective sparse refinement, the criteria being the topological features of interest creases and extra-ordinary vertices. The intent is to eventually provide more flexibility to facilitate the refinement of particular regions of interest or more dynamicadaptive needs. Refinement has also been subclassed according to the type of topological split being performed, i.e. splitting all faces into quads or tris via the QuadRefinement and TriRefinement subclasses. As noted with VtrLevel , there is further room for improvement in memory andor performance here by combining more optimal specializations for both Refinement and Level -- with consideration of separating the uniform and sparse cases. Parent-child and child-parent relationships While Refinement populates a new child Level as part of its refinement operation, it also accumulates the relationships between the parent and child level and as with Level , this data is stored in vectors indexable by the components. The associations between components in the two levels was initially only uni-directional child components were associated with incident components of a parent component based on the parent components topology, so we had a parent-to-child mapping one to many. Storing the reverse child-to-parent mapping was avoided to reduce memory particularly in the case of uniform refinement as it often was not necessary, but a growing need for it, particularly in the case of sparse feature-adaptive refinement, lead to it being included. Data flexibility One of the advantages of the structure-of-arrays representation in both Level and Refinement is that we can make more dynamic choices about what type of data we choose to allocate and use based on needs. For instance, we can choose between maintaining the parent-child or child-parent mapping in Refinement , or both if needed, and we can remove one if no longer necessary. An active example of this is uniform refinement if we only require the face-vertex list at the finest subdivision level, there is no need to generate a complete topological description of that level as would be required of more traditional representations, and given that level is 4x the magnitude of its parent, the savings are considerable. Currently there is nothing specific to a subdivision scheme in the refinement other than the type of topological splitting to apply. The refinement does subdivide sharpness values for creasing, but that too is independent of scheme. Tags were added to the base level that are propagated through the refinement and these too are dependent on the scheme, but are applied externally. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"Vtr Overview Vtr Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Vectorized Topology Representation Vtr VtrLevel Topological Relationships Memory Efficiency VtrRefinement Parent-child and child-parent relationships Data flexibility Vectorized Topology Representation Vtr Vtr consists of a suite of classes that collectively provide an intermediate representation of topology that supports efficient refinement. Vtr is intended for internal use only and is currently accessed through the Far layer by the FarTopologyRefiner , which assembles these Vtr classes to meet the topological and refinement needs of the Far layer. What follows is therefore more intended to provide insite into the underlying architecture than to describe particular usage. For documentation more relevant to direct usage, proceed to the Far section previously noted. Vtr is vectorized in that its topological data is stored more as a collection of vectors of primitive elements rather than as the faces, vertices and edges that make up many other topological representations. It is essentially a structure-of-arrays SOA approach to topology in contrast to the more common array-of-structures pattern found in many other topological representations. Vtrs use of vectors allows it to be fairly efficient in its use of memory and similarly efficient to refine, but the topology is fixed once defined. Vtr classes are purely topological. They are even more independent of the representation of vertices, faces, etc. than Hbr in that they are not even parameterized by an interface to such components. So the same set of Vtr objects can eventually be used to serve more than one representation of these components. The primary requirement is that a mesh be expressable as an indexable set i.e. a vector or array of vertices, edges and faces. The index of a component uniquely identifies it and properties are retrieved by referring to it by index. Its worth qualifying the term topological here and elsewhere -- we generally refer to topology as subdivision topology rather than mesh topology. A subdivision hierarchy is impacted by the presence of semi-sharp creasing, as the subdivision rules change in response to that creasing. So subdivision topology includes the sharpness values assigned to edges and vertices that affect the semi-sharp creasing. The two primary classes in Vtr consist of VtrLevel - a class representing complete vertex topology for a level VtrRefinement - a class mapping a parent VtrLevel to a child level Others exist to represent the following selection and appropriate tagging of components for sparse refinement divergence of face-varying topology from the vertex topology mapping between face-varying topology at successive levels common low-level utilities, e.g. simple array classes VtrLevel VtrLevel is a complete topological description of a subdivision level, with the topological relations, sharpness values and component tags all stored in vectors literally stdvectors, but easily changed via typedefs. There are no classes or objects for the mesh component types i.e. faces, edges and vertices but simply an integer index to identify each. It can be viewed as a structure-of-arrays representation of the topology any property related to a particular component is stored in an array and accessible using the index identifying that component. So with no classes the for the components, its difficult to say what constitutes a vertex or a face they are each the sum of all the fields scattered amongst the many vectors included. Level represents a single level of a potential hierarchy and is capable of representing the complete base mesh. There are no members that relate data in one level to any other, either below or above. As such, any Level can be used as the base level for a new subdivision hierarchy potentially more than one. All relationships between separate levels are maintained in the VtrRefinement class. Topological Relationships Level requires the definition of and associations between a fixed set of indexable components for all three component types, i.e. an explicit edge list in addition to the expected set of vertices and faces. There are no explicit component objects in the representation, only an integer index VtrIndex identifying each component within the set and data associated with that component in the various vectors. The topology is stored as six sets of incident relations between the components two each for the two other component types incident each component type, i.e. for each face, its incident vertices and incident edges for each edge, its incident vertices and incident faces for each vertex, its incident edges and incident faces The collection of incidence relations is a vectorized variation of AIF the Adjacency and Incidence Framework. The set of these six incidence relations is not minimal only four are required, but that set excludes the most desired face-vertex relation but all six are kept and maintained to facilitate faster refinement. While the sizes of several vectors are directly proportional to the number of vertices, edges or faces to which the data is associated, the sizes of some of the vectors for these relations is more cumulative and so additional vectors of offsets is required typical of the face-vertex list commonly used as the minimal definition of mesh topology. Vectors for the sharpness values associated with crease edges and corner vertices are included and so sized according to the number of edges and vertices, along with additional tags for the components that may be helpful to refinement i.e. the type of subdivision Rule associated with each vertex. A Level is really just a container for data in a subdivision level, and so its public methods are primarily to access that data. Modification of the data is protected and only made available to classes that are intended to construct Levels currently the Far factory class that is responsible for building the base level, and the VtrRefinement class that constructs subsequent levels during refinement. Memory Efficiency One of the advantages in storing data in what is essentially a structure-of-arrays, rather than the array-of-structures more typical of topological representations, is that we can be more selective about memory usage in some cases. Particularly in the case of uniform refinement, when the data in subsequent levels is typically 4x its predecessor, we can minimize what we either generate or keep around at each level. For instance, if only a face-list is required at the finest level, we only need to generate one of the six topological relations the vertices incident each face. When we do keep Levels around in memory as is the case with the FarTopologyRefiner we do have do have the opportunity to prune what is not strictly necessary after the refinement. Just as with construction, whatever classes are privileged to construct a Level are likely those that will be privileged to prune its contents when needed. The current implementation of Level is far from optimal though -- there are opportunities for improvement. After one level of subdivision, the faces in a Level will be either all quads or tris. Having specializations for these cases and using the more general case in support of N-sided faces for the base level only is one possibility. Levels also allocate dozens of vectors in which to store all data. Since these vectors are of fixed size once created, they could be aggregated by partitioning one or a smaller number of larger block of memory into the desired pieces. The desire to make some of these improvements is part of why Vtr is not directly exposed for public use and instead exposed via Far. VtrRefinement While VtrLevel contains the topology for each subdivision level, VtrRefinement is responsible for creating a new level via refinement of an existing one, and for maintaining the relationships between the components in the parent and child levels. So a simplified view of a subdivision hierarchy with Vtr is a set of Levels with a Refinement between each successive pair. Refinement is a friend of Level and will populate a child level from a parent given a set of refinement parameters. Aside from parameters related to data or depth, there are two kinds of refinement supported uniform and sparse. The latter sparse refinement requires selection of an arbitrary set of components -- any dependent or neighboring components that are required for the limit will be automatically included. So feature-adaptive refinement is just one form of this selective sparse refinement, the criteria being the topological features of interest creases and extra-ordinary vertices. The intent is to eventually provide more flexibility to facilitate the refinement of particular regions of interest or more dynamicadaptive needs. Refinement has also been subclassed according to the type of topological split being performed, i.e. splitting all faces into quads or tris via the QuadRefinement and TriRefinement subclasses. As noted with VtrLevel , there is further room for improvement in memory andor performance here by combining more optimal specializations for both Refinement and Level -- with consideration of separating the uniform and sparse cases. Parent-child and child-parent relationships While Refinement populates a new child Level as part of its refinement operation, it also accumulates the relationships between the parent and child level and as with Level , this data is stored in vectors indexable by the components. The associations between components in the two levels was initially only uni-directional child components were associated with incident components of a parent component based on the parent components topology, so we had a parent-to-child mapping one to many. Storing the reverse child-to-parent mapping was avoided to reduce memory particularly in the case of uniform refinement as it often was not necessary, but a growing need for it, particularly in the case of sparse feature-adaptive refinement, lead to it being included. Data flexibility One of the advantages of the structure-of-arrays representation in both Level and Refinement is that we can make more dynamic choices about what type of data we choose to allocate and use based on needs. For instance, we can choose between maintaining the parent-child or child-parent mapping in Refinement , or both if needed, and we can remove one if no longer necessary. An active example of this is uniform refinement if we only require the face-vertex list at the finest subdivision level, there is no need to generate a complete topological description of that level as would be required of more traditional representations, and given that level is 4x the magnitude of its parent, the savings are considerable. Currently there is nothing specific to a subdivision scheme in the refinement other than the type of topological splitting to apply. The refinement does subdivide sharpness values for creasing, but that too is independent of scheme. Tags were added to the base level that are propagated through the refinement and these too are dependent on the scheme, but are applied externally. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"vtr_overview.html\"\n }, {\n- \"title\": \"Release Notes\",\n- \"text\": \"Release Notes Release Notes 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Release 2.6.0 Release 2.5.1 Release 2.5.0 Release 2.4.1 Release 2.4.0 Release 2.3.5 Release 2.3.4 Release 2.3.3 Release 2.3.2 Release 2.3.1 Release 2.3.0 Release 2.2.0 Release 2.1.0 Release 2.0.1 Release 2.0.0 Release 1.2.4 Release 1.2.3 Release 1.2.2 Release 1.2.1 Release 1.2.0 Release 1.1.0 Release 1.0.0 Release 2.6.0 New Features Add subdivision kernels for ARM NEON Add OsdUtilVertexSplit which creates a vertex-varying data table by duplicating vertices in a FarMesh Add basic functions to work with FV data via evaluator API Changes Added Catmark restricted vertex compute kernels that optimize for vertices with no semi-sharp creases Fix accessor omissions in osdmesh.h Add support for different subdivision schemes for OsdUtilMesh Bug Fixes Fix crashes when using rather low-end cards like Intel ones Fix a bug in the creation of an edge-vertex kernel batch Fix mismatch in declaration and usage of OsdCudaComputeRestrictedVertexA Fix a bug in the vertex order for restricted Catmark vertex-vertex kernel batches Fix a bug in FarCatmarkSubdivisionTablesFactory that prevented the CATMARK_QUAD_FACE_VERTEX kernel from being selected for subdivision level 2 or greater. Fix a bug in OsdUtilVertexSplit that occurs when getting the address of the end of a stdvector Fix error in createCLBuffer that occurs when the buffer size is zero Fix a bug in the CUDA computeRestrictedEdge kernel Fix duplicate variables with identical name Fix osdutil build errors Fix cmake diagnostic messsage Release 2.5.1 New Features Add CATMARK_QUAD_FACE_VERTEX and CATMARK_TRI_QUAD_FACE_VERTEX compute kernels optimization that takes advantage of all-quads or all-triange-and-quads meshes Bug Fixes Fix a compiler error in the GLSL Transform Feedback kernels on OS X Fix boundary interpolation in osdutil Fix bilinear stencil tangent computions Release 2.5.0 New Features Add ability to generate triangle patches for a uniformly subdivided mesh Add new example topologySharing Add interleaved buffer mode in glViewer Add GLSL compute kernel to glBatchViewer Add TBB compute kernel to glBatchViewer Add a PullDown widget to our HUD in examplescommon GUI updates cosmetic changes to GL example code Adding a programmable image shader to gl_hud Code cleanup for GLFrameBuffer in examplescommon Implement C-API accessor to evaluator topology osdutil Add command line option to CMakes options Add a CMake option to disable OpenCL Add a FindCLEW.cmake module in anticipation of using CLEW as a dependency Integrate CLEW into osd library and examples Changes Change interleaved buffer support in OsdCompute Removed OsdVertexDescriptor and replaced with OsdVertexBufferDescriptor Reorganize ComputeContext and ComputeController. Reorganize EvalStencilContext and EvalStencilController Moved transient states current vertex buffer etc to controller Reorganize EvalLimitContext and EvalLimitController Moved transient states current vertex buffer etc to controller Fix adaptive isolation of sharp corner vertices Fix incorrect FarMeshFactory logic for isolating multiple corner vertices in corner patches Change EvalLimit Gregory patch kernels to the large weights table to accomodate higher valences Fix calculation of screen space LOD tess factors for transition corner patches. Add a public constructor to OsdMesh Decrease compiler warning thresholds and fix outstanding warnings Make PTex support optional Add a NO_MAYA flag to CMakeLists to disable all Autodesk Maya dependencies in the build Document NO_MAYA command line option Bug Fixes Fix mistakenly deleted memory barrier in glsl OsdCompute kernel. Fix shape_utils genRIB function to use streams correctly. Temporary workaround for the synchronization bug of glsl compute kernel Fix Hud display for higher DPI MBP retina Fix Hud d3d11 Fix examples to use GL timer query to measure the GPU draw timing more precisely Fix glViewer stop updating during freeze. Fix file permissions on farPatchTablesFactory.h Fix some meory leaks in adaptive evaluator osdutil Fix OsdUtilAdaptiveEvaluator concurrency issue Fix OsdUtilRefiner incorrect Invalid size of patch array error reporting. Fix OsdUtilPatchPartitioner failure for triangle patches Fixes a bug that causes OsdUtilPatchPartitioner to fail to rebuild the face-varying data table correctly for triangle patches. Add missing third parameter to templated OsdDrawContext usage osdutilbatch.h Return success status from openSubdiv_finishEvaluatorDescr osdutil Remove debugging stdcout calls osdutil Build errors warnings Fix OSX Core Profile build GLFrameBuffer Fix ptexViewer build error on OSX Fix framebuffer shader compiling for OSX Reordering includes to address a compile error on OSXglew environment Fix compilation errors with CLEW enabled Fix icc build problems Fix compiler warnings in OsdClVertexBuffer Fix compilation error on windowsmsvc2013 Fix build warningserrors with VS2010 Pro Fix Windows build warning in FarPatchTablesFactory Fix doxygen generation errors Release 2.4.1 Changes Add correct OpenSubdiv namespace beginend blocks. Bug Fixes Compile osdutil with -fPIC for correct linking. Fix a bug of OsdUtilMeshBatch, the varying buffer isnt computed with CL kernels Fix FindGLFW.cmake to use the GLFW_LOCATION environment variable in Windows Fix Draw contexts do not fully initialize patch arrays Release 2.4.0 New Features Adding functionality to store uniform face-varying data across multiple levels of subdivision Add OsdUtilPatchPartitioner. It splits patcharray into subsets so that clients can draw partial surfaces for both adaptive and uniform. Changes Remove FarMesh dependency from OsdContext. Use DSA APIs for GL buffer update if available. Refactor Far API replace void- of all kernel applications with CONTEXT template parameter. It eliminates many static_casts from void- for both far and osd classes. move the big switch-cases of far default kernel launches out of Refine so that osd controllers can arbitrary mix default kernels and custom kernels. change FarKernelBatchkernelType from enum to int, clients can add custom kernel types. remove a back-pointer to farmesh from subdivision table. untemplate all subdivision table classes and template their compute methods instead. Those methods take a typed vertex storage. remove an unused argument FarMesh from the constructor of subdivision table factories. Refactor FarSubdivisionTables. Delete scheme specialized subdivision tables. The base class FarSubdivisionTables already has all tables, so we just need scheme enum to identify which scheme the subdivision tables belong to. This brings a lot of code cleanups around far factory classes. Move FarMultiMeshFactory to OsdUtil. Move table splicing functions of FarMultiMeshFactory into factories Change PxOsdUtil prefix to final OsdUtil prefix. Improve error reporting in osdutil refinement classes, and fix a build issue Bug Fixes Fix another multi mesh splicing bug of face varying data. Make CMake path variables more robust Fixing a crash on Marvericks wglew Update dxViewer example documentation Fix wrong logic in openSubdiv_setEvaluatorCoarsePositions Remove debug print from adaptive evaluators initialization Release 2.3.5 New Features Add the ability to read obj files to the dxViewer example Add screen-capture function to ptexViewer Update documention for Xcode builds Add documentation boundary interpolation rules and face-varying boundary interpolation rules Changes Refactoring FarPatchTables and FarPatchTablesFactory Move GL vertex buffer VBO buffer allocation out of allocate and into BindVBO Enable uvViewer on OS X now that Mavericks is released. Replacing un-necessary dynamic_cast with reinterpret_cast within FarDispatcher Minor code cleanup of FarMeshFactory Remove address space qualifiers from OpenCL kernel functions Fix OpenCL initialization to be slightly more robust Add OpenCL header include paths where necessary Add static specifiers for non-kernel CL funcs at program scope Add stddef.h to pythonosdosdshim.i Modify ptexViewer and uvViewer shaders to address some portability issues Bug Fixes Fix Gregory Boundary patch buffer overrun Fix black texels when the resolution of a ptex face is less than 4 Fix a splicing bug in FarMultiMeshFactory Fix a build error when using older versions of GLFW Fix build warnings optimized Fix FindTBB.cmake Fix FindMaya.cmake Fix glViewer support for GLSL compute Fix ptexViewer enable specular pass in both IBL and point lighting Fix Zlib include in ptexViewer Fix ptexViewer shader errors. Fix osdPolySmooth Maya plugin Fix UV merging in osdPolySmooth code example Add cleanup function to osdPolySmooth Maya plugin Fix Maya OsdPolySmooth node component output Fix GLSL array instantiation syntax for glStencilViewer Fix examples to run correctly on high DPI displays with GLFW 3 Release 2.3.4 New Features Adding CPUOMPTBB Context Controller pairs for CPU evaluation of smooth normals Added adaptiveEvaluator class inspired by Sergeys work in blender OsdUtil Changes Changed the HUD to ignore mouse clicks when not visible. Updates for blender development OsdUtil Add C compatible API to access the adaptiveEvaluator class from non-C OsdUtil Update license headers to apache OsdUtil CMake build improvement make osd a cmake object library remove compiling redundancies Improve stringification of shaders kernels in CMake build Bug Fixes Fixed iOS build Fixed VS2010 warningserrors. Fix OsdCpuEvalLimitKernel Fix maxvalence calculation in FarMeshFactory Fix FarStencilFactory control stencil caching Removing assert for high-valence vertices running off limit tangent pre-computed table. Fix degenerate stencil limit tangent code path. Fix unused variable build warnings gcc 4.8.2 - Fedora 19 Fix build warning from osdutiladaptiveEvaluator.cpp Release 2.3.3 Changes Modify Far remapping of singular vertices to point to their source vertex. Refactoring Ptex Mipmap and Analytic Displacement code Adding some documentation for Chaikin crease rule Misc. improvements to PxOsdUtilsMesh Adding recommended isolation output to OsdPolySmooth node Bug Fixes Adding an error check on version parsing of main CMakeLists Fix regex in FindMaya.cmake that breaks with recent versions of Maya Fix crashes induced by typeid Fixed VS2010 build warning Fix build break in hbr_regression Fix incorrect capitalization in GL ptexViewer shader.glsl Fix OSX build add stdlib.h include Release 2.3.2 New Features Adding control cage drawing to ptexViewer Adding Maya osdPolySmooth plugin into OpenSubdiv examples. Changes Removing some glGetError checks that are causing problems for Autodesk D3D11DrawRegistry returns the common shader config for all non-tess patcharrays. Updates to simple cpu osdutil classes Bug Fixes Fix Hbr Chaikin crease rule Fix Chaikin tag parsing Fix return value of allocate function for OsdCPUGLVertxBuffer Fixed GLSL shader portability. Fix FindGLFW.cmake for GLFW 3.03 on OSX Fixed compiler warnings. Fixed VS2010 build errors Fixed WIN32 build error when no DXSDK installed. Fix OSX build stdlib.h needs to be included in glPtexMipmapTexture.h Fix for crash in new meshrefiner code in OsdUtil Release 2.3.1 New Features Add DX11 version of ptex mipmap loader Add DX11 ptex viewer work in progress Add DX11 fractional partitioning, normal derivatives computation Add memory usage controls to Ptex loader Add face-varying boundary interpolation parsing to shape_utils Add simple HbrMesh and FarMesh wrapper classes to osdutil Changes Amend language of attribution file NOTICE.txt Optimize a bit of ptex mipmap lookup. Show ptex memory usage in GL and DX11 ptexViewers Improve ptex guttering Addding some video links to our collection of external resources Bug Fixes Fix edge-only face-varying interpolation Fix Far to handle disconnected vertices in an Hbr mesh Fixed ptex cache resource release sequence Fix build symbol conflict in Far Fix patch parambuffer generation in OsdD3D11DrawContext Fix a minor osdutil build warning seen with gcc 4.8.1 Fix VS2010 build errors Release 2.3.0 New Features Added Analytical displacement mapping Analytic Displacement Mapping using Hardware Tessellation Niessner and Loop TOG 2013 Added a new ptex mipmap loader Added face varying macros for loop subdivision Added the uvViewer example to see how face varying interpolation rule works Added a slider component and cleanup hud code. Changes Adding license attribution files, improved language of the code headers Install documentation into the Filesystem Hierarchy Standard location Set GLFW_OPENGL_FORWARD_COMPAT on Mac OS to make samples work on that platform Added surface normal mode mipmap to ptxViewer Bug Fixes Fix a bug of bad fvar splicing for loop surface. Fix incorrect bilinear limit tangents in FarStencilTablesFactory Fix boundary interpolation rules doc Added an error check on updating cuda buffer Fix face varying rendering on loop surface Fixed glBatchViewer build for GLFW 2.x Expand search paths for FindGLFW.cmake for Debian and other Linux architectures Fix CMake executable builds for ICC Fix bhr baseline regression, so reference files are real OBJs Fixed clKernelBundle.cpp to build on Android. Fix misc build warings Release 2.2.0 New Features Added subdivision stencil functionality Far OsdEval Bug Fixes Fix D3D11DrawContext to check for NULL pointers Fix cpuEvalLimitController crash bug Fixed search path suffixes for ICC libs Fixed invalid initialization of glslTransformFeedback kernel. Release 2.1.0 New Features Added TBB Compute back-end on Linux contribution from Sheng Fu Added support for ICC compiler still Beta Changes Added constructor to OsdMesh with a FarMesh as input Modify CMake to name and sym-link DSOs based on Linux ABI versioning spec Added command line input to DX11 viewer FarMultiMesh can splice uniform and adaptive meshes together. Bug Fixes Fix FarMultiMesh splicing Removed unnecessary cudaThreadSynchronize calls. Fix glViewer overlapping HUD menus Fix facevarying rendering in glBatchViewer Fix build of GLSL transform feedback kernels Fix Getting Started documentation Release 2.0.1 New Features New CLA files to reflect Apache 2.0 licensing Changes Move all public headers to includeopensubdiv... Adding Osd documentation based on Siggraph slides Bug Fixes Fix incorrect transition pattern 3 in GLSL HLSL shaders Fix CMake build to not link GPU-based libraries into libosdCPU Fix support for GLEW on OSX Fix GLFW Xrandr xf86vmode dependency paths for X11 based systems Fix HUD display overlaps in code examples Fix FindGLEW.cmake to be aware of multiarch on linux systems Fix some hard-coded include paths in CMake build Release 2.0.0 New Features New CMake build flags NO_LIB, NO_CUDA, NO_PYTHON Changes OpenSubdiv is now under Apache 2.0 license HbrHalfedge and HbrFVarData copy constructors are now private Documentation style matched to graphics.pixar.com new content Add an animation freeze button to ptexViewer Variable name changes for better readability across all example shader code Bug Fixes Fix incorrect patch generation for patches with 2 non-consecutive boundary edges Fix undefined gl_PrimitiveID shader build errors Fix for shader macro OSD_DISPLACEMENT_CALLBACK Fix out-of-bounds stdvector access in FarPatchTablesFactory Release 1.2.4 New Features Adding support for fractional tessellation of patches Adding a much needed API documention system based on Docutils RST markup Adding support for face-varying interpolation in GLSL APIs Adding varying data buffers to OsdMesh Adding accessors to the vertex buffers in OsdGlMesh Adding face-varying data to regression shapes Changes Cleanup of common bicubic patch shader code GLSL HLSL for portability ATI OSX drivers Bug Fixes Fix FarVertexEditTablesFactory to insert properly vertex edit batches fixes incorrect hierarchical hole in regression shape Fix FarPatchMap quadtree to not drop top-level non-quad faces Fix Gregory patches bug with incorrect max-valence Fix FarPatchTablesGetNumFaces and FarPatchTablesGetFaceVertices functions to return the correct values Fix face indexing GLSL code ptex works on non-quads again Fix face-varying data splicing in FarMultiMeshFactory Fix ptex face indexing in FarMultiMeshFactory Fix glew include to not break builds Fix Clang ICC build failures with FarPatchTables Fix build and example code to work with GFLW 3.0 Fix cmake to have ptex dynamically linked in OSX Release 1.2.3 New Features Adding Varying and Face-Varying data interpolation to EvalLimit Changes EvalLimit API refactor the EvalContext now has dedicated structs to track all the vertex, varying and face-varying data streams. Also renamed some buffers into tables to maintain code consistency EvalLimit optimization switch serial indexing to a quad-tree based search Bug Fixes Face-varying data bug fixes making sure the data is carried around appropriately Fixes for OpenCL use with the new batching APIs GLSL general shader code cleanup fixes for better portability GLSL Tranform Feedback initialization fix Critical fix for FarMultiMesh batching indexing was incorrect Fix osdutil CL implementation protect includes on systems with no OpenCL SDK installed Fix face-varying interpolation on adaptive patches FarPatchTables fix IsFeatureAdaptive to return the correct answer Fix Far factories to handle the absence of face-varying data correctly. Many GLSL shader code style fixes which should help with ATI OSX shader compiling Release 1.2.2 New Features Introducing the EvalLimit API the Eval module aims at providing support for computational tasks that are not related to drawing the surfaces. The EvalLimit sub-module provides an API that enables client code to evaluate primitive variables on the limit surface. OsdxxxComputeController minor optimization. Added early exit to Refine method to avoid unnecessary interop. Changes OsdGLDawContext minor API change. Protecting some member variables and adding const accessors OsdError minor API refactor, added Warning functions. Bug Fixes Fix Ptex bug prevent corner texel guttering code to from going into infinite loops Adding the ability for a FarMeshFactory to construct patchTables starting from firstLevel in uniform subdivision mode Consolidating the color coding of bicubic patch types through all our our code examples this is used mostly as a debugging tool Fixing some MSVC build warnings Update to the outdated README.md Release 1.2.1 New Features Added CUDA runtime error checking Release 1.2.0 Changes Major Far refactor around patchTables to introduce the draw batching API Renaming osd_util to osdutil Bug Fixes Fix GLSL transform feedback initialization bug in ptexViewer Minor bug typo fixes Release 1.1.0 New Features release initiated because of the switch to Git Flow Release 1.0.0 Oringal release Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n- \"tags\": \"\",\n- \"loc\": \"release_notes_2x.html\"\n- }, {\n \"title\": \"Tutorials\",\n- \"text\": \"Tutorials Tutorials 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Bfr Tutorials 1. Basic Evaluation and Tessellation Tutorial 1.1 Tutorial 1.2 Tutorial 1.3 Tutorial 1.4 Tutorial 1.5 2. More on Tessellation Tutorial 2.1 Tutorial 2.2 3. Additional Topics Tutorial 3.1 Tutorial 3.2 Far Tutorials 1. Basic Construction and Interpolation Tutorial 1.1 Tutorial 1.2 2. Uniform Refinement and Primvar Data Types Tutorial 2.1 Tutorial 2.2 Tutorial 2.3 3. Creating a Custom FarTopologyRefinerFactory Tutorial 3.1 4. Construction and Usage of FarStencilTables Tutorial 4.1 Tutorial 4.2 Tutorial 4.3 5. Construction and Usage of FarPatchTables Tutorial 5.1 Tutorial 5.2 Tutorial 5.3 Osd Tutorials Tutorial 0 Hbr Tutorials Tutorial 0 Tutorial 1 Tutorial 2 The tutorial source code can be found in the github.com repository or in your local repository roottutorials . Bfr Tutorials All tutorials for the Bfr interface follow a similar pattern without any command line arguments, a default mesh usually a cube is used and the results printed to standard output in Obj format. Command line arguments can be used to specify an alternate mesh for input, as well as directing the output to a specified Obj file. Some tutorials may offer additional command line options to trigger internal options relevant to the topic, e.g. those illustrating tessellation may support a -quads option to use the option to tessellate quad-bases subdivision meshes with quads. 1. Basic Evaluation and Tessellation Tutorial 1.1 This tutorial illustrates the use of BfrSurfaceFactory and BfrSurface to evaluate points on the limit of each face. The limit positions at all corners of the face are evaluated and connected to the limit position in the center of the face -- creating a simple triangular tessellation. code Tutorial 1.2 This tutorial shows the added use of BfrTessellation to identify the set of points and connectivity for a uniform tessellation. Both a Surface and Tessellation is identified for each face, with the Tessellation indicating which points are to be evaluated by the Surface. code Tessellation patterns for quad-based subdivision schemes can optionally preserve quads left or be fully triangulated center, while triangular schemes always yield triangles right Tutorial 1.3 This tutorial extends the previous tutorial on uniform Tessellation by adding face-varying Surfaces to compute corresponding UVs for each evaluated position. code Tutorial 1.4 This tutorial extends the previous tutorial on uniform tessellation of position and UV by illustrating how additional mesh data interleaved with the position and UV data is easily handled. code Tutorial 1.5 This tutorial is similar to the first tutorial showing uniform tessellation of position but makes use of limit stencils for its evaluation of points of the tessellation pattern. code 2. More on Tessellation Tutorial 2.1 This tutorial extends the use of Tessellation to illustrate the use of non-uniform tessellation rates per edge. A simple edge-length metric is used to determine the tessellation rate for each edge of a face. code The following contrasts uniform tessellation left with the resulting length-based tessellations -- preserving quads center and fully triangulated right Tutorial 2.2 This tutorial is a more complex extension of the use of Tessellation that illustrates how the separation and association of tessellation data with the boundary and interior of the face can be used. Limit points evaluated on the vertices and edges of a face the boundary of the Tessellation are computed once and shared with adjacent faces -- creating a topologically watertight tessellation of the mesh. code 3. Additional Topics Tutorial 3.1 This tutorial shows a basic example of the more advanced topic of creating a subclass of SurfaceFactory adapted to a connected mesh representation -- requiring an implementation of the SurfaceFactoryMeshAdapter interface for that mesh. A simplified version of the implementation of FarTopologyRefiner is provided. Note that the code imported here is that of the main program, not the separate header and source files of the custom subclass illustrated -- which current documentation scripts cannot import. Tutorial 3.2 This tutorial shows how to initialize and retain Surfaces for later use. The simple uniform tessellation tutorial is modified to first create and populate a simple caching structure that initializes and stores the Surfaces for all faces of the mesh. The loop for each face of the mesh then retrieves its Surface and associated patch points from the cache. code Far Tutorials 1. Basic Construction and Interpolation Tutorial 1.1 This tutorial presents the requisite steps to instantiate a mesh as a FarTopologyRefiner from simple topological data and to interpolate vertex data associated with the mesh. code Tutorial 1.2 This tutorial makes use of a different vertex data definition for use when vertex data is of arbitrary width. Uniform refinement is applied to data buffers of three types two of fixed but different sizes and the third a union of the two that is dynamically sized and constructed. code 2. Uniform Refinement and Primvar Data Types Tutorial 2.1 Building on the basic tutorial, this example shows how to instantiate a simple mesh, refine it uniformly and then interpolate both vertex and varying primvar data. code Tutorial 2.2 Building on the previous tutorial, this example shows how to instantiate a simple mesh, refine it uniformly and then interpolate both vertex and face-varying primvar data. The resulting interpolated data is output in Obj format, with the face-varying data recorded in the UV texture layout. code Tutorial 2.3 Building on previous tutorials, this example shows how to instantiate a simple mesh, refine it uniformly, interpolate both vertex and face-varying primvar data, and finally calculate approximated smooth normals. The resulting interpolated data is output in Obj format. code 3. Creating a Custom FarTopologyRefinerFactory Tutorial 3.1 Previous tutorials have instantiated topology from a simple face-vertex list via the FarTopologyDescriptor and its TopologyRefinerFactory. This tutorial shows how to more efficiently convert an existing high-level topology representation to a FarTopologyDescriptor with a custom factory class. code 4. Construction and Usage of FarStencilTables Tutorial 4.1 This tutorial shows how to create and manipulate a StencilTable. Factorized stencils are used to efficiently interpolate vertex primvar data buffers. code Tutorial 4.2 This tutorial shows how to create and manipulate StencilTables for both vertex and varying primvar data buffers vertex positions and varying colors. code Tutorial 4.3 This tutorial shows how to create and manipulate tables of cascading stencils to apply hierarchical vertex edits. code 5. Construction and Usage of FarPatchTables Tutorial 5.1 This tutorial shows how to compute points on the limit surface at arbitrary parametric locations using a FarPatchTable constructed from adaptive refinement. code Tutorial 5.2 Building on the previous tutorial, this example shows how to manage the limit surface of a potentially large mesh by creating and evaluating separate PatchTables for selected groups of faces of the mesh. code Tutorial 5.3 Building on the previous tutorials for both PatchTables and StencilTables, this example shows how to construct a LimitStencilTable to repeatedly evaluate an arbitrary collection of points on the limit surface. code Osd Tutorials Tutorial 0 This tutorial demonstrates the manipulation of Osd Evaluator and BufferDescriptor. code Hbr Tutorials Use of Hbr is no longer recommended -- these tutorials are included solely for historical reference. Tutorial 0 This tutorial presents, in a very succinct way, the requisite steps to instantiate an Hbr mesh from simple topological data. code Tutorial 1 This tutorial shows how to safely create Hbr meshes from arbitrary topology. Because Hbr is a half-edge data structure, it cannot represent non-manifold topology. Ensuring that the geometry used is manifold is a requirement to use Hbr safely. This tutorial presents some simple tests to detect inappropriate topology. code Tutorial 2 This tutorial shows how to subdivide uniformly a simple Hbr mesh. We are building upon previous tutorials and assuming a fully instantiated mesh we start with an HbrMesh pointer initialized from the same pyramid shape used in hbr_tutorial_0. We then apply the Refine function sequentially to all the faces in the mesh to generate several levels of uniform subdivision. The resulting data is then dumped to the terminal in Wavefront OBJ format for inspection. code Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"Tutorials Tutorials 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Bfr Tutorials 1. Basic Evaluation and Tessellation Tutorial 1.1 Tutorial 1.2 Tutorial 1.3 Tutorial 1.4 Tutorial 1.5 2. More on Tessellation Tutorial 2.1 Tutorial 2.2 3. Additional Topics Tutorial 3.1 Tutorial 3.2 Far Tutorials 1. Basic Construction and Interpolation Tutorial 1.1 Tutorial 1.2 2. Uniform Refinement and Primvar Data Types Tutorial 2.1 Tutorial 2.2 Tutorial 2.3 3. Creating a Custom FarTopologyRefinerFactory Tutorial 3.1 4. Construction and Usage of FarStencilTables Tutorial 4.1 Tutorial 4.2 Tutorial 4.3 5. Construction and Usage of FarPatchTables Tutorial 5.1 Tutorial 5.2 Tutorial 5.3 Osd Tutorials Tutorial 0 Hbr Tutorials Tutorial 0 Tutorial 1 Tutorial 2 The tutorial source code can be found in the github.com repository or in your local repository roottutorials . Bfr Tutorials All tutorials for the Bfr interface follow a similar pattern without any command line arguments, a default mesh usually a cube is used and the results printed to standard output in Obj format. Command line arguments can be used to specify an alternate mesh for input, as well as directing the output to a specified Obj file. Some tutorials may offer additional command line options to trigger internal options relevant to the topic, e.g. those illustrating tessellation may support a -quads option to use the option to tessellate quad-bases subdivision meshes with quads. 1. Basic Evaluation and Tessellation Tutorial 1.1 This tutorial illustrates the use of BfrSurfaceFactory and BfrSurface to evaluate points on the limit of each face. The limit positions at all corners of the face are evaluated and connected to the limit position in the center of the face -- creating a simple triangular tessellation. code Tutorial 1.2 This tutorial shows the added use of BfrTessellation to identify the set of points and connectivity for a uniform tessellation. Both a Surface and Tessellation is identified for each face, with the Tessellation indicating which points are to be evaluated by the Surface. code Tessellation patterns for quad-based subdivision schemes can optionally preserve quads left or be fully triangulated center, while triangular schemes always yield triangles right Tutorial 1.3 This tutorial extends the previous tutorial on uniform Tessellation by adding face-varying Surfaces to compute corresponding UVs for each evaluated position. code Tutorial 1.4 This tutorial extends the previous tutorial on uniform tessellation of position and UV by illustrating how additional mesh data interleaved with the position and UV data is easily handled. code Tutorial 1.5 This tutorial is similar to the first tutorial showing uniform tessellation of position but makes use of limit stencils for its evaluation of points of the tessellation pattern. code 2. More on Tessellation Tutorial 2.1 This tutorial extends the use of Tessellation to illustrate the use of non-uniform tessellation rates per edge. A simple edge-length metric is used to determine the tessellation rate for each edge of a face. code The following contrasts uniform tessellation left with the resulting length-based tessellations -- preserving quads center and fully triangulated right Tutorial 2.2 This tutorial is a more complex extension of the use of Tessellation that illustrates how the separation and association of tessellation data with the boundary and interior of the face can be used. Limit points evaluated on the vertices and edges of a face the boundary of the Tessellation are computed once and shared with adjacent faces -- creating a topologically watertight tessellation of the mesh. code 3. Additional Topics Tutorial 3.1 This tutorial shows a basic example of the more advanced topic of creating a subclass of SurfaceFactory adapted to a connected mesh representation -- requiring an implementation of the SurfaceFactoryMeshAdapter interface for that mesh. A simplified version of the implementation of FarTopologyRefiner is provided. Note that the code imported here is that of the main program, not the separate header and source files of the custom subclass illustrated -- which current documentation scripts cannot import. Tutorial 3.2 This tutorial shows how to initialize and retain Surfaces for later use. The simple uniform tessellation tutorial is modified to first create and populate a simple caching structure that initializes and stores the Surfaces for all faces of the mesh. The loop for each face of the mesh then retrieves its Surface and associated patch points from the cache. code Far Tutorials 1. Basic Construction and Interpolation Tutorial 1.1 This tutorial presents the requisite steps to instantiate a mesh as a FarTopologyRefiner from simple topological data and to interpolate vertex data associated with the mesh. code Tutorial 1.2 This tutorial makes use of a different vertex data definition for use when vertex data is of arbitrary width. Uniform refinement is applied to data buffers of three types two of fixed but different sizes and the third a union of the two that is dynamically sized and constructed. code 2. Uniform Refinement and Primvar Data Types Tutorial 2.1 Building on the basic tutorial, this example shows how to instantiate a simple mesh, refine it uniformly and then interpolate both vertex and varying primvar data. code Tutorial 2.2 Building on the previous tutorial, this example shows how to instantiate a simple mesh, refine it uniformly and then interpolate both vertex and face-varying primvar data. The resulting interpolated data is output in Obj format, with the face-varying data recorded in the UV texture layout. code Tutorial 2.3 Building on previous tutorials, this example shows how to instantiate a simple mesh, refine it uniformly, interpolate both vertex and face-varying primvar data, and finally calculate approximated smooth normals. The resulting interpolated data is output in Obj format. code 3. Creating a Custom FarTopologyRefinerFactory Tutorial 3.1 Previous tutorials have instantiated topology from a simple face-vertex list via the FarTopologyDescriptor and its TopologyRefinerFactory. This tutorial shows how to more efficiently convert an existing high-level topology representation to a FarTopologyDescriptor with a custom factory class. code 4. Construction and Usage of FarStencilTables Tutorial 4.1 This tutorial shows how to create and manipulate a StencilTable. Factorized stencils are used to efficiently interpolate vertex primvar data buffers. code Tutorial 4.2 This tutorial shows how to create and manipulate StencilTables for both vertex and varying primvar data buffers vertex positions and varying colors. code Tutorial 4.3 This tutorial shows how to create and manipulate tables of cascading stencils to apply hierarchical vertex edits. code 5. Construction and Usage of FarPatchTables Tutorial 5.1 This tutorial shows how to compute points on the limit surface at arbitrary parametric locations using a FarPatchTable constructed from adaptive refinement. code Tutorial 5.2 Building on the previous tutorial, this example shows how to manage the limit surface of a potentially large mesh by creating and evaluating separate PatchTables for selected groups of faces of the mesh. code Tutorial 5.3 Building on the previous tutorials for both PatchTables and StencilTables, this example shows how to construct a LimitStencilTable to repeatedly evaluate an arbitrary collection of points on the limit surface. code Osd Tutorials Tutorial 0 This tutorial demonstrates the manipulation of Osd Evaluator and BufferDescriptor. code Hbr Tutorials Use of Hbr is no longer recommended -- these tutorials are included solely for historical reference. Tutorial 0 This tutorial presents, in a very succinct way, the requisite steps to instantiate an Hbr mesh from simple topological data. code Tutorial 1 This tutorial shows how to safely create Hbr meshes from arbitrary topology. Because Hbr is a half-edge data structure, it cannot represent non-manifold topology. Ensuring that the geometry used is manifold is a requirement to use Hbr safely. This tutorial presents some simple tests to detect inappropriate topology. code Tutorial 2 This tutorial shows how to subdivide uniformly a simple Hbr mesh. We are building upon previous tutorials and assuming a fully instantiated mesh we start with an HbrMesh pointer initialized from the same pyramid shape used in hbr_tutorial_0. We then apply the Refine function sequentially to all the faces in the mesh to generate several levels of uniform subdivision. The resulting data is then dumped to the terminal in Wavefront OBJ format for inspection. code Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"tutorials.html\"\n }, {\n- \"title\": \"Sdc Overview\",\n- \"text\": \"Sdc Overview Sdc Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Subdivision Core Sdc Types, Traits and Options Creasing support Scheme-specific support The FACE, EDGE and VERTEX interfaces The MASK interface Subdivision Core Sdc Sdc is the lowest level layer in OpenSubdiv. Its intent is to separate the core subdivision details from any particular representation of a mesh it was previously bound to Hbr to facilitate the generation of consistent results with other mesh representations, both internal and external to OpenSubdiv. The functionality can be divided roughly into three sections types, traits and options for the supported subdivision schemes computations required to support semi-sharp creasing computations for mask weights of subdivided vertices for all schemes For most common usage, familiarity with only the first of these is necessary -- primarily the use of public types and constants for the choice of subdivision scheme and its associated options. The latter two provide the basis for a more comprehensive implementation of subdivision, which requires considerably more understanding and effort. Overall, the approach was to extract the functionality at the lowest level possible. In some cases, the implementation is not far from being simple global functions. The intent was to start at a low level and build any higher level functionality as needed. What exists now is functional for ongoing development and anticipated needs within OpenSubdiv for the near future. The intent of Sdc is to provide the building blocks for OpenSubdiv and its clients to efficiently process the specific set of supported subdivision schemes. It is not intended to be a general framework for defining customized subdivision schemes. Types, Traits and Options The most basic type is the enum SdcSchemeType that identifies the fixed set of subdivision schemes supported by OpenSubdiv Bilinear , Catmark and Loop . With this alone, we intend to avoid all dynamic casting issues related to the scheme by simply adding members to the associated subclasses for inspection. In addition to the type enum itself, a class defining a fixed set of traits associated with each scheme is provided. While these traits are available as static methods in the interface of a class supporting more functionality for each scheme to be described shortly, the SchemeTypeTraits provide queries of the traits for a variable of type SdcSchemeType -- enabling parameterization of code by the value of a trait without templates or virtual inheritance a simple internal table of traits is constructed and trivially indexed. The second contribution is the collection of all variations in one place that can be applied to the subdivision schemes, i.e. the boundary interpolation rules, creasing method, edge subdivision choices, etc. The fact that these are all declared in one place alone should help clients see the full set of variations that are possible. A simple Options struct a set of bitfields aggregates all of these variations into a single object the equivalent of an integer in this case that are passed around to other Sdc classes andor methods and are expected to be used at a higher level both within OpenSubdiv and externally. By aggregating the options and passing them around as a group, it allows us to extend the set easily in future without the need to rewire a lot of interfaces to accommodate the new choice. Clients can enable new choices at the highest level and be assured that they will propagate to the lowest level where they are relevant. Unlike other options structs used elsewhere to specify variations of a particular method, SdcOptions defines all options that affect the shape of the underlying limit surface of a subdivision mesh. Other operations at higher levels in the library may have options that approximate the shape and so create a slightly different appearance, but SdcOptions is a fundamental part of the definition of the true limit surface. Creasing support Since the computations involved in the support of semi-sharp creasing are independent of the subdivision scheme, the goal in Sdc was to encapsulate all related creasing functionality in a similarly independent manner. Computations involving sharpness values are also much less dependent on topology -- there are vertices and edges with sharpness values, but knowledge of faces or boundary edges is not required, -- so the complexity of topological neighborhoods required for more scheme-specific functionality is arguably not necessary here. Creasing computations have been provided as methods defined on a Crease class that is constructed with a set of Options. Its methods typically take sharpness values as inputs and compute a corresponding set of sharpness values as a result. For the Uniform creasing method previously known as Normal , the computations may be so trivial as to question whether such an interface is worth it, but for Chaikin or other schemes in the future that are non-trivial, the benefits should be clear. Functionality is divided between both uniform and non-uniform, so clients have some control over avoiding unnecessary overhead, e.g. non-uniform computations typically require neighboring sharpness values around a vertex, while uniform does not. Also included as part of the Crease class is the Rule enum -- this indicates if a vertex is Smooth, Crease, Dart or Corner referred to as the mask in Hbr and is a function of the sharpness values at and around a vertex. Knowing the Rule for a vertex can accelerate mask queries, and the Rule can often be inferred based on the origin of a vertex e.g. it originated from the middle of a face, was the child of a Smooth vertex, etc.. Methods are defined for the Crease class to subdivide edge and vertex sharpness values determine the Rule for a vertex based on incident sharpness values determine the transitional weight between two sets of sharpness values Being all low-level and working directly on sharpness values, it is a clients responsibility to coordinate the application of any hierarchical crease edits with their computations. Similarly, in keeping with this as a low-level interface, values are passed as primitive arrays. This follows the trend in OpenSubdiv of dealing with data of various kinds e.g. weights, component indices, now sharpness values, etc. in small contiguous sets of values. In most internal cases we can refer to a set of values or gather what will typically be a small number of values on the stack for temporary use. Scheme-specific support While the SchemeTypeTraits class provides traits for each subdivision scheme supported by OpenSubdiv i.e. Bilinear , Catmark and Loop , the Scheme class provides these more directly, Additionally, the Scheme class provides methods for computing the various sets of weights used to compute new vertices resulting from subdivision. The collection of weights used to compute a single vertex at a new subdivision level is typically referred to as a mask . The primary purpose of the Scheme class is to provide such masks in a manner both general and efficient. Each subdivision scheme has its own values for its masks, and each are provided as specializations of the template class SchemeSchemeType TYPE . The intent is to minimize the amount of code specific to each scheme. The computation of mask weights for subdivided vertices is the most significant contribution of Sdc. The use of semi-sharp creasing with each non-linear subdivision scheme complicates what are otherwise simple masks determined solely by the topology, and packaging that functionality to achieve both the generality and efficiency desired has been a challenge. Mask queries are defined in the Scheme class template, which has specializations for each of the supported subdivision schemes. Mask queries are defined in terms of interfaces for two template parameters the first defining the topological neighborhood of a vertex, and a second defining a container in which to gather the individual weights template typename FACE , typename MASK void ComputeFaceVertexMask FACE const faceNeighborhood , MASK faceVertexMask , ... const Each mask query is expected to call methods defined for the FACE , EDGE or VERTEX classes to obtain the information they require typically these methods are simple queries about the topology and associated sharpness values. Clients are free to use their own mesh representations to gather the requested information as quickly as possible, or to cache some subset as member variables for immediate inline retrieval. In general, the set of weights for a subdivided vertex is dependent on the following the topology around the parent component from which the vertex originates the type of subdivision Rule applicable to the parent component the type of subdivision Rule applicable to the new child vertex a transitional weight blending the effect between differing parent and child rules This seems fairly straight-forward, until we look at some of the dependencies involved the parent Rule requires the sharpness values at and around the parent component the child Rule requires the subdivided sharpness values at and around the new child vertex though it can sometimes be trivially inferred from the parent the transitional weight between differing rules requires all parent and child sharpness values Clearly the sharpness values are inspected multiple times and so it pays to have them available for retrieval. Computing them on an as-needed basis may be simple for uniform creasing, but a non-uniform creasing method requires traversing topological neighborhoods, and that in addition to the computation itself can be costly. The point here is that it is potentially unreasonable to expect to evaluate the mask weights completely independent of any other consideration. Expecting and encouraging the client to have subdivided sharpness values first, for use in more than one place, is therefore recommended. The complexity of the general case above is also unnecessary for most vertices. Any client using Sdc typically has more information about the nature of the vertex being subdivided and much of this can be avoided -- particularly for the smooth interior case that often dominates. More on that in the details of the Scheme classes. Given that most of the complexity has been moved into the template parameters for the mask queries, the Scheme class remains fairly simple. Like the Crease class, it is instantiated with a set of Options to avoid them cluttering the interface. It is currently little more than a few methods for the limit and refinement masks for each vertex type, plus the few fixed traits of the scheme as static methods. The mask queries have been written in a way that greatly simplifies the specializations required for each scheme. The generic implementation for both the edge-vertex and vertex-vertex masks take care of all of the creasing logic, requiring only a small set of specific masks to be assigned for each Scheme smooth and crease masks for an edge-vertex, and smooth, crease and corner masks for a vertex-vertex. Other than the Bilinear case, which will specialize the mask queries to trivialize them for linear interpolation, the specializations for each Scheme should only require defining this set of masks -- and with two of them common edge-vertex crease and vertex-vertex corner the Catmark scheme only needs to define three. The FACE, EDGE and VERTEX interfaces Mask queries require an interface to a topological neighborhood, currently labeled FACE , EDGE and VERTEX . This naming potentially implies more generality than intended, as such classes are only expected to provide the methods required of the mask queries to compute its associated weights. While all methods must be defined, some may rarely be invoked, and the client has considerable flexibility in the implementation of these they can defer some evaluations lazily until required, or be pro-active and cache information in member variables for immediate access. An approach discussed in the past has alluded to iterator classes that clients would write to traverse their meshes. The mask queries would then be parameterized in terms of a more general and generic mesh component that would make use of more general traversal iterators. The advantage here is the iterators are written once, then traversal is left to the query and only what is necessary is gathered. The disadvantages are that clients are forced to write these to do anything, getting them correct and efficient may not be trivial or possible in some cases, and that the same data e.g. subdivided sharpness may be gathered or computed multiple times for different purposes. The other extreme was to gather everything possible required at once, but that is objectionable. The approach taken here provides a reasonable compromise between the two. The mask queries ask for exactly what they want, and the provided classes are expected to deliver it as efficiently as possible. In some cases the client may already be storing it in a more accessible form and general topological iteration can be avoided. The information requested of these classes in the three mask queries is as follows For FACE the number of incident vertices For EDGE the number of incident faces the sharpness value of the parent edge the sharpness values of the two child edges the number of vertices per incident face For VERTEX the number of incident faces the number of incident edges the sharpness value of the parent vertex the sharpness values for each incident parent edge the sharpness value of the child vertex the sharpness values for each incident child edge The latter should not be surprising given the dependencies noted above. There are also a few more to consider for future use, e.g. whether the EDGE or VERTEX is manifold or not. In most cases, additional information can be provided to the mask queries i.e. pre-determined Rules, and most of the child sharpness values are not necessary. The most demanding situation is a fractional crease that decays to zero -- in which case all parent and child sharpness values in the neighborhood are required to determine the proper transitional weight. The MASK interface Methods dealing with the collections of weights defining a mask are typically parameterized by a MASK template parameter that contains the weights. The set of mask weights is currently divided into vertex-weights, edge-weights and face-weights -- consistent with previous usage in OpenSubdiv and providing some useful correlation between the full set of weights and topology. The vertex-weights refer to parent vertices incident the parent component from which a vertex originated, the edge-weights the vertices opposite incident edges of the parent, and the face-weights the center of incident parent faces. Note the latter is NOT in terms of vertices of the parent but potentially vertices in the child originating from faces of the parent. This has been done historically in OpenSubdiv but is finding less use -- particularly when it comes to providing greater support for the Loop scheme -- and is a point needing attention. So the mask queries require the following capabilities assign the number of vertex, edge andor face weights retrieve the number of vertex, edge andor face weights assign individual vertex, edge andor face weights by index retrieve individual vertex, edge andor face weights by index through a set of methods required of all MASK classes. Since the maximum number of weights is typically known based on the topology, usage within Vtr, Far or Hbr is expected to simply define buffers on the stack. Another option is to utilize pre-allocated tables, partitioned into the three sets of weights on construction of a MASK , and populated by the mask queries. A potentially useful side-effect of this is that the client can define their weights to be stored in either single or double-precision. With that possibility in mind, care was taken within the mask queries to make use of a declared type in the MASK interface MASKWeight for intermediate calculations. Having support for double-precision masks in Sdc does enable it at higher levels in OpenSubdiv if later desired, and that support is made almost trivial with MASK being generic. It is important to remember here that these masks are being defined consistent with existing usage within OpenSubdiv both Hbr and the subdivision tables generated by Far . As noted above, the face weights correspond to the centers of incident faces, i.e. vertices on the same level as the vertex for which the mask is being computed, and not relative to vertices in the parent level as with the other sets of weights. It is true that the weights can be translated into a set in terms solely of parent vertices, but in the general case i.e. Catmark subdivision with non-quads in the base mesh this requires additional topological association. In general we would need N-3 weights for the N-3 vertices between the two incident edges, where N is the number of vertices of each face typically 4 even at level 0. Perhaps such a translation method could be provided on the mask class, with an optional indication of the incident face topology for the irregular cases. The Loop scheme does not have face weights , for a vertex-vertex mask, but for an edge-vertex mask it does require weights associated with the faces incident the edge -- either the vertex opposite the edge for each triangle, or its center which has no other use for Loop. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"Subdivision Surfaces\",\n+ \"text\": \"Subdivision Surfaces Subdivision Surfaces 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Overview Piecewise Parametric Surfaces Parametric Patches Piecewise Surfaces Arbitrary Topology Regular versus Irregular Features Non-manifold Topology Subdivision versus Tessellation Subdivision Tessellation Which to Use Mesh Data and Topology Separating Data from Topology Vertex and Varying Data Face-Varying Data and Topology Schemes and Options Subdivision Schemes Boundary Interpolation Rules Face-varying Interpolation Rules Semi-Sharp Creases Other Options Chaikin Rule Triangle Subdivision Rule Overview Subdivision surfaces are a common modeling primitive that has gained popularity in animation and visual effects over the past decades. As the name suggests, subdivision surfaces are fundamentally surfaces . More specifically, subdivision surfaces are piecewise parametric surfaces defined over meshes of arbitrary topology -- both concepts that will be described in the sections that follow. Subdivision is both an operation that can be applied to a polygonal mesh to refine it, and a mathematical tool that defines the underlying smooth surface to which repeated subdivision of the mesh converges. Explicit subdivision is simple to apply some number of times to provide a smoother mesh, and that simplicity has historically lead to many tools representing the shape this way. In contrast, deriving the smooth surface that ultimately defines the shape -- its limit surface -- is considerably more complex but provides greater accuracy and flexibility. These differences have led to confusion in how some tools expose subdivision surfaces. The ultimate goal is to have all tools use subdivision surfaces as true surface primitives. The focus here is therefore less on subdivision and more on the nature of the surface that results from it. In addition to providing a consistent implementation of subdivision -- one that includes a number of widely used feature extensions -- a significant value of OpenSubdiv is that it makes the limit surface more accessible. Since its introduction, OpenSubdiv has received interest from users and developers with a wide variety of skills, interests and backgrounds. This document is intended to present subdivision surfaces from a perspective helpful in making use of OpenSubdiv. One purpose it serves is to provide a high level overview for those with less experience with the algorithms or mathematics of subdivision. The other is to provide an overview of the feature set available with OpenSubdiv, and to introduce those capabilities with the terminology used by OpenSubdiv as much of it is overloaded. Piecewise Parametric Surfaces Piecewise parametric surfaces are arguably the most widely used geometric representation in industrial design, entertainment and many other areas. Many of the objects we deal with everyday -- cars, mobile phones, laptops -- were all designed and visualized first as piecewise parametric surfaces before those designs were approved and pursued. Piecewise parametric surfaces are ultimately just collections of simpler modeling primitives referred to as patches. Patches constitute the pieces of the larger surface in much the same way as a face or polygon constitutes a piece of a polygonal mesh. Parametric Patches Patches are the building blocks of piecewise smooth surfaces, and many different kinds of patches have evolved to meet the needs of geometric modeling. Two of the more effective and common patches are illustrated below Single bicubic B-Spline patch Single bicubic Bezier patch Patches consist of a set of points or vertices that affect a rectangular piece of smooth surface triangular patches also exist. That rectangle is parameterized in its two directions, transforming a simple 2D rectangle into the 3D surface u,v 2D domain of a patch Mapping from u,v to x,y,z The points that control the shape of the surface are usually referred to as control points or control vertices, and the collection of the entire set defining a patch as the control mesh, the control hull, the control cage or simply the hull, the cage, etc. For the sake of brevity we will frequently use the term cage, which serves us more generally later. So a patch essentially consist of two entities its control points and the surface affected by them. The way the control points affect the surface is what makes the different types of patches unique. Even patches defined by the same number of points can have different behavior. Note that all 16 points of the B-Spline patch above are relatively far from the surface they define compared to the similar Bezier patch. The two patches in that example actually represent exactly the same piece of surface -- each with a set of control points having different effects on it. In mathematical terms, each control point has a basis function associated with it that affects the surface in a particular way when only that point is moved Bicubic B-Spline basis function Bicubic Bezier basis funciton It is these basis functions that often give rise to the names of the different patches. There are pros and cons to these different properties of the control points of patches, which become more apparent as we assemble patches into piecewise surfaces. Piecewise Surfaces Piecewise parametric surfaces are collections of patches. For rectangular patches, one of the simplest ways to construct a collection is to define a set of patches using a rectangular grid of control points Piecewise B-Spline surface Piecewise Bezier surface Note that we can overlap the points of adjacent B-spline patches. This overlapping means that moving one control point affects multiple patches -- but it also ensures that those patches always meet smoothly this was a design intention and not true for other patch types. Adjacent Bezier patches only share points at their boundaries and coordinating the points across those boundaries to keep the surface smooth is possible, but awkward. This makes B-splines a more favorable surface representation for interactive modeling, but Bezier patches serve many other useful purposes. A more complicated B-spline surface Part of a more complicated B-Spline surface Just as a patch consisted of a cage and a surface, the same is now true of the collection. The control cage is manipulated by a designer and the surface of each of the patches involved is displayed so they can assess its effect. Arbitrary Topology Piecewise surfaces discussed thus far have been restricted to collections of patches over regular grids of control points. There is a certain simplicity with rectangular parametric surfaces that is appealing, but a surface representation that supports arbitrary topology has many other advantages. Rectangular parametric surfaces gained widespread adoption despite their topological limitations, and their popularity continues today in some areas. Complex objects often need many such surfaces to represent them and a variety of techniques have evolved to assemble them effectively, including stitching multiple surfaces together or cutting holes into them trimming. These are complicated techniques, and while effective in some contexts e.g. industrial design they become cumbersome in others e.g. animation and visual effects. A single polygonal mesh can represent shapes with far more complexity than a single rectangular piecewise surface, but its faceted nature eventually becomes a problem. Subdivision surfaces combine the topological flexibility of polygonal meshes with the underlying smoothness of piecewise parametric surfaces. Just as rectangular piecewise parametric surfaces have a collection of control points its cage stored as a grid and an underlying surface, subdivision surfaces also have a collection of control points its cage stored as a mesh and an underlying surface often referred as its limit surface. Regular versus Irregular Features A mesh contains the vertices and faces that form the cage for the underlying surface, and the topology of that mesh can be arbitrarily complex. In areas where the faces and vertices of the mesh are connected to form rectangular grids, the limit surface becomes one of the rectangular piecewise parametric surfaces previously mentioned. These regions of the mesh are said to be regular they provide behavior familiar from the use of similar rectangular surfaces and their limit surface is relatively simple to deal with. All other areas are considered irregular they provide the desired topological flexibility and so are less familiar and less predictable in some cases and their limit surface can be much more complicated. Irregular features come in a number of forms. The most widely referred to is an extra-ordinary vertex, i.e. a vertex which, in the case of a quad subdivision scheme like Catmull-Clark, does not have four incident faces. Irregular vertex and incident faces Regular and irregular regions of the surface The presence of these irregular features makes the limit surface around them similarly irregular, i.e. it cannot be represented as simply as it can for regular regions. Its worth noting that irregular regions shrink in size and become more isolated as subdivision is applied. A face with a lot of extra-ordinary vertices around it makes for a very complicated surface, and isolating these features is a way to help deal with that complexity Two valence-5 vertices nearby Isolation subdivided once Isolation subdivided twice Its generally necessary to perform some kind of local subdivision in these areas to break these pieces of surface into smaller, more manageable pieces, and the term feature adaptive subdivision has become popular in recent years to describe this process. Whether this is done explicitly or implicitly, globally or locally, what matters most is that there is an underlying piece of limit surface for each face -- albeit a potentially complicated one at an irregular feature -- that can be evaluated in much the same way as rectangular piecewise surfaces. Patches of the regular regions Patches of the irregular region While supporting a smooth surface in these irregular areas is the main advantage of subdivision surfaces, both the complexity of the resulting surfaces and their quality are reasons to use them with care. When the topology is largely irregular, there is a higher cost associated with its surface, so minimizing irregularities is advantageous. And in some cases the surface quality, i.e. the perceived smoothness, of the irregular surfaces can lead to undesirable artefacts. An arbitrary polygonal mesh will often not make a good subdivision cage, regardless of how good that polygonal mesh appears. As with rectangular piecewise parametric surfaces, the cage should be shaped to affect the underlying surface it is intended to represent. See Modeling Tips for related recommendations. Non-manifold Topology Since the cage of a subdivision surface is stored in a mesh, and often manipulated in the same context as polygonal meshes, the topic of manifold versus non-manifold topology warrants some attention. There are many definitions or descriptions of what distinguishes a manifold mesh from one that is not. These range from concise but abstract mathematical definitions to sets of examples showing manifold and non-manifold meshes -- all have their value and an appropriate audience. The following is not a strict definition but serves well to illustrate most local topological configurations that cause a mesh to be non-manifold. Consider standing on the faces of a mesh and walking around each vertex in turn. Assuming a right-hand winding order of faces, stand on the side of the face in the positive normal direction. And when walking, step across each incident edge in a counter-clockwise direction to the next incident face. For an interior vertex start at the corner of any incident face walk around the vertex across each incident edge to the next unvisited face repeat if you arrive back where you started and any incident faces or edges were not visited, the mesh is non-manifold Similarly, for a boundary vertex start at the corner of the face containing the leading boundary edge walk around the vertex across each incident edge to the next unvisited face repeat if you arrive at another boundary edge and any incident faces or edges were not visited, the mesh is non-manifold If you can walk around all vertices this way and dont encounter any non-manifold features, the mesh is likely manifold. Obviously if a vertex has no faces, there is nothing to walk around and this test cant succeed, so it is again non-manifold. All of the faces around a vertex should also be in the same orientation, otherwise two adjacent faces have normals in opposite directions and the mesh will be considered non-manifold, so we should really include that constraint when stepping to the next face to be more strict. Consider walking around the indicated vertices of the following non-manifold meshes Edges with 2 incident faces Faces sharing a vertex but no edges As mentioned earlier, many tools do not support non-manifold meshes, and in some contexts, e.g. 3D printing, they should be strictly avoided. Sometimes a manifold mesh may be desired and enforced as an end result, but the mesh may temporarily become non-manifold due to a particular sequence of modeling operations. Rather than supporting or advocating the use of non-manifold meshes, OpenSubdiv strives to be robust in the presence of non-manifold features to simplify the usage of its clients -- sparing them the need for topological analysis to determine when OpenSubdiv can or cannot be used. Although subdivision rules are not as well standardized in areas where the mesh is not manifold, OpenSubdiv provides simple rules and a reasonable limit surface in most cases. Surface around edges with 2 incident faces Surface for faces sharing a vertex but no edges As with the case of regular versus irregular features, since every face has a corresponding piece of surface associated with it -- whether locally manifold or not -- the term arbitrary topology can be said to include non-manifold topology. Subdivision versus Tessellation The preceding sections illustrate subdivision surfaces as piecewise parametric surfaces of arbitrary topology. As piecewise parametric surfaces, they consist of a cage and the underlying surface defined by that cage. Two techniques used to display subdivision surfaces are subdivision and tessellation. Both have their legitimate uses, but there is an important distinction between them subdivision operates on a cage and produces a refined cage tessellation operates on a surface and produces a discretization of that surface The existence and relative simplicity of the subdivision algorithm makes it easy to apply repeatedly to approximate the shape of the surface, but with the result being a refined cage, that approximation is not always very accurate. When compared to a cage refined to a different level, or a tessellation that uses points evaluated directly on the limit surface, the discrepancies can be confusing. Subdivision Subdivision is the process that gives subdivision surfaces their name, but it is not unique to them. Being piecewise parametric surfaces, lets first look at subdivision in the context of the simpler parametric patches that comprise them. Subdivision is a special case of refinement , which is key to the success of some of the most widely used types of parametric patches and their aggregate surfaces. A surface can be refined when an algorithm exists such that more control points can be introduced while keeping the shape of the surface exactly the same . For interactive and design purposes, this allows a designer to introduce more resolution for finer control without introducing undesired side effects in the shape. For more analytical purposes, it allows the surface to be broken into pieces, often adaptively, while being faithful to the original shape. One reason why both B-spline and Bezier patches are so widely used is that both of them can be refined. Uniform subdivision -- the process of splitting each of the patches in one or both of its directions -- is a special case of refinement that both of these patch types support B-Spline surface and its cage Cage subdivided 1x Cage subdivided 2x In the cases illustrated above for B-Splines, the uniformly refined cages produce the same limit surface as the original granted in more pieces. So it is fair to say that both uniform B-splines and Bezier surfaces are subdivision surfaces. The limit surface remains the same with the many more control points roughly 4x with each iteration of subdivision, and those points are closer to but not on the surface. It may be tempting to use these new control points to represent the surface, but using the same number of points evaluated at corresponding uniformly spaced parametric locations on the surface is usually simpler and more effective. Note also that points of the cage typically do not have any normal vectors associated with them, though we can evaluate normals explicitly for arbitrary locations on the surface just as we do for position. So if displaying a cage as a shaded surface, normal vectors at each of the control points must be contrived. Both the positions and normals of the points on the finer cage are therefore both approximations. For more general subdivision surfaces, the same is true. Subdivision will refine a mesh of arbitrary topology, but the resulting points will not lie on the limit surface and any normal vectors contrived from and associated with these points will only be approximations to those of the limit surface. Tessellation There is little need to use subdivision to approximate a parametric surface when it can be computed directly, i.e. it can be tessellated. We can evaluate at arbitrary locations on the surface and connect the resulting points to form a tessellation -- a discretization of the limit surface -- that is far more flexible than the results achieved from uniform subdivision Uniform 3x3 tessellation of B-spline surface Curvature-adaptive tessellation of B-spline surface For a simple parametric surface, the direct evaluation of the limit surface is also simple, but for more complicated subdivision surfaces of arbitrary topology, this is less the case. The lack of a clear understanding of the relationship between the limit surface and the cage has historically lead to many applications avoiding tessellation. Its worth mentioning that subdivision can be used to generate a tessellation even when the limit surface is not available for direct evaluation. The recursive nature of subdivision does give rise to formulae that allow a point on the limit surface to be computed that corresponds to each point of the cage. This process is often referred to as snapping or pushing the points of the cage onto the limit surface. Subdivided 1x and snapped to limit surface Subdivided 2x and snapped to limit surface Since the end result is a connected set of points on the limit surface, this forms a tessellation of the limit surface, and we consider it a separate process to subdivision though it does make use of it. The fact that such a tessellation might have been achieved using subdivision is indistinguishable from the final result -- the same tessellation might just as easily have been generated by evaluating limit patches of the cage uniformly 2x, 4x, 8x, etc. along each edge. Which to Use Subdivision is undeniably useful in creating finer cages to manipulate the surface, but tessellation is preferred for displaying the surface when the patches are available for direct evaluation. There was a time when global refinement was pursued in limited circles as a way of rapidly evaluating parametric surfaces along isoparametric lines, but patch evaluation, i.e. tessellation, generally prevails. Considerable confusion has arisen due the way the two techniques have been employed and presented when displaying the shape in end-user applications. One can argue that if an application displays a representation of the surface that is satisfactory for its purposes, then it is not necessary to burden the user with additional terminology and choices. But when two representations of the same surface differ considerably between two applications, the lack of any explanation or control leads to confusion. As long as applications make different choices on how to display the surface, we seek a balance between simplicity and control. Since subdivided points do not lie on the limit surface, it is important to make it clear to users when subdivision is being used instead of tessellation. This is particularly true in applications where the cage and the surface are displayed in the same style as there is no visual cue for users to make that distinction. Mesh Data and Topology The ability of subdivision surfaces to support arbitrary topology leads to the use of meshes to store both the topology of the cage and the data values associated with its control points, i.e. its vertices. The shape of a mesh, or the subdivision surface that results from it, is a combination of the topology of the mesh and the position data associated with its vertices. When dealing with meshes there are advantages to separating the topology from the data, and this is even more important when dealing with subdivision surfaces. The shape referred to above is not just the shape of the mesh the cage in this case but could be the shape of a refined cage or the limit surface. By observing the roles that both the data and topology play in operations such as subdivision and evaluation, significant advantages can be gained by managing data, topology and the associated computations accordingly. While the main purpose of subdivision surfaces is to use position data associated with the vertices to define a smooth, continuous limit surface, there are many cases where non-positional data is associated with a mesh. That data may often be interpolated smoothly like position, but often it is preferred to interpolate it linearly or even make it discontinuous along edges of the mesh. Texture coordinates and color are common examples here. Other than position, which is assigned to and associated with vertices, there are no constraints on how arbitrary data can or should be associated or interpolated. Texture coordinates, for example, can be assigned to create a completely smooth limit surface like the position, linearly interpolated across faces, or even made discontinuous between them. There are, however, consequences to consider -- both in terms of data management and performance -- which are described below as the terminology and techniques used to achieve each are defined. Separating Data from Topology While the topology of meshes used to store subdivision surfaces is arbitrarily complex and variable, the topology of the parametric patches that make up its limit surface are simple and fixed. Bicubic B-Spline and Bezier patches are both defined by a simple 4x4 grid of control points and a set of basis functions for each point that collectively form the resulting surface. For such a patch, the position at a given parametric location is the result of the combination of position data associated with its control points and the weights of the corresponding basis functions weights being the values of basis functions evaluated at a parametric location. The topology and the basis functions remain the same, so we can make use of the weights independent of the data. If the positions of the control points change, we can simply recombine the new position data with the weights that we just used and apply the same combination. The fixed topology of a parametric patch and two shapes resulting from two sets of positions. Similarly, for a piecewise surface, the position at a given parametric location is the result of the single patch containing that parametric location evaluated at the given position. The control points involved are the subset of control points associated with that particular patch. If the topology of the surface is fixed, so too is the topology of the collection of patches that comprise that surface. If the positions of those control points change, we can recombine the new position data with the same weights for the subset of points associated with the patch. More complex but fixed topology of a surface and two shapes resulting from two sets of positions. This holds for a piecewise surface of arbitrary topology. Regardless of how complex the topology, as long as it remains fixed i.e. relationships between vertices, edges and faces does not change or anything other settings affecting subdivision rules, the same techniques apply. This is just one example of the value of separating computations involving topology from those involving the data. Both subdivision and evaluation can be factored into steps involving topology computing the weights and combining the data separately. Three shapes resulting from three sets of positions for a mesh of fixed topology. When the topology is fixed, enormous savings are possible by pre-computing information associated with the topology and organizing the data associated with the control points in a way that can be efficiently combined with it. This is key to understanding some of the techniques used to process subdivision surfaces. For a mesh of arbitrary topology, the control points of the underlying surface are the vertices, and position data associated with them is most familiar. But there is nothing that requires that the control points of a patch have to represent position -- the same techniques apply regardless of the type of data involved. Vertex and Varying Data The most typical and fundamental operation is to evaluate a position on the surface, i.e. evaluate the underlying patches of the limit surface using the x,y,z positions at the vertices of the mesh. Given a parametric u,v location on one such patch, the data-independent evaluation method first computes the weights and then combines the x,y,z vertex positions resulting in an x,y,z position at that location. But the weights and their combination can be applied to any data at the vertices, e.g. color, texture coordinates or anything else. Data associated with the vertices that is interpolated this way, including position, is said to be vertex data or to have vertex interpolation. Specifying other data as vertex data will result in it being smoothly interpolated in exactly the same way using exactly the same weights as the position. So to capture a simple 2D projection of the surface for texture coordinates, 2D values matching the x,y of the positions would be used. If linear interpolation of data associated with vertices is desired instead, the data is said to be varying data or to have varying interpolation. Here the non-linear evaluation of the patches defining the smooth limit surface is ignored and weights for simple linear interpolation are used. This is a common choice for texture coordinates as evaluation of texture without the need of bicubic patches is computationally cheaper. The linear interpolation will not capture the smoothness required of a true projection between the vertices, but both vertex and varying interpolation have their uses. Projected texture smoothly interpolated from vertex data Projected texture linearly interpolated from varying data Since both vertex and varying data is associated with vertices a unique value assigned to each, the resulting surface will be continuous -- piecewise smooth in the case of vertex data and piecewise linear in the case of varying. Face-Varying Data and Topology In order to support discontinuities in data on the surface, unlike vertex and varying data, there must be multiple values associated with vertices, edges andor faces, in order for a discontinuity to exist. Discontinuities are made possible by assigning values to the corners of faces, similar to the way in which vertices are assigned to the corners of faces when defining the topology of the mesh. Recalling the assignment of vertices to faces Vertex indices are assigned to all corners of each face as part of mesh construction and are often referred to as the face-vertices of an individual face or the mesh. All face-vertices that share the same vertex index will be connected by that vertex and share the same vertex data associated with it. By assigning a different set of indices to the face-vertices -- indices not referring to the vertices but some set of data to be associated with the corners of each face -- corners that share the same vertex no longer need to share the same data value and the data can be made discontinuous between faces This method of associating data values with the face-vertices of the mesh is said to be assigning face-varying data for face-varying interpolation. An interpolated value will vary continuously within a face i.e. the patch of the limit surface associated with the face but not necessarily across the edges or vertices shared with adjacent faces. Disjoint face-varying UV regions applied to the limit surface The combination of associating data values not with the vertices the control points but the face corners, and the resulting data-dependent discontinuities that result, make this a considerably more complicated approach than vertex or varying. The added complexity of the data alone is reason to only use it when necessary, i.e. when discontinuities are desired and present. Part of the complexity of dealing with face-varying data and interpolation is the way in which the interpolation behavior can be defined. Where the data is continuous, the interpolation can be specified to be as smooth as the underlying limit surface of vertex data or simply linear as achieved with varying data. Where the data is discontinuous -- across interior edges and around vertices -- the discontinuities create boundaries for the data, and partition the underlying surface into disjoint regions. The interpolation along these boundaries can also be specified as smooth or linear in a number of ways many of which have a historical basis. A more complete description of the different linear interpolation options with face-varying data and interpolation is given later. These options make it possible to treat the data as either vertex or varying, but with the added presence of discontinuities. An essential point to remember with face-varying interpolation is that each set of data is free to have its own discontinuities -- this leads to each data set having both unique topology and size. The topology specified for a collection of face-varying data is referred to as a channel and is unique to face-varying interpolation. Unlike vertex and varying interpolation, which both associate a data value with a vertex, the number of values in a face-varying channel is not fixed by the number of vertices or faces. The number of indices assigned to the face-corners will be the same for all channels, but the number of unique values referred to by these indices may not. We can take advantage of the common mesh topology in areas where the data is continuous, but we lose some of those advantages around the discontinuities. This results in the higher complexity and cost of a face-varying channel compared to vertex or varying data. If the topology for a channel is fixed, though, similar techniques can be applied to factor computation related to the topology so that changes to the data can be processed efficiently. Schemes and Options While previous sections have described subdivision surfaces in more general terms, this section describes a number of common variations often referred to as extensions to the subdivision algorithms and the ways that they are represented in OpenSubdiv. The number and nature of the extensions here significantly complicate what are otherwise fairly simple subdivision algorithms. Historically applications have supported either a subset or have had varying implementations of the same feature. OpenSubdiv strives to provide a consistent and efficient implementation of this feature set. Given the varying presentations of some of these features elsewhere, the naming chosen by OpenSubdiv is emphasized here. Subdivision Schemes OpenSubdiv provides two well known subdivision surface types -- Catmull-Clark often referred to more tersely as Catmark and Loop subdivision. Catmull-Clark is more widely used and suited to quad-dominant meshes, while Loop is preferred for and requires purely triangulated meshes. The many examples from previous sections have illustrated the more popular Catmull-Clark scheme. For an example of Loop Boundary Interpolation Rules Boundary interpolation rules control how subdivision and the limit surface behave for faces adjacent to boundary edges and vertices. The following choices are available via the enumeration SdcOptionsVtxBoundaryInterpolation Mode Behavior VTX_BOUNDARY_NONE No boundary edge interpolation is applied by default boundary faces are tagged as holes so that the boundary vertices continue to support the adjacent interior faces, but no surface corresponding to the boundary faces is generated boundary faces can be selectively interpolated by sharpening all boundary edges incident the vertices of the face VTX_BOUNDARY_EDGE_ONLY A sequence of boundary vertices defines a smooth curve to which the limit surface along boundary faces extends VTX_BOUNDARY_EDGE_AND_CORNER Similar to edge-only but the smooth curve resulting on the boundary is made to interpolate corner vertices vertices with exactly one incident face On a grid example In practice, it is rare to use no boundary interpolation at all -- this feature has its uses in allowing separate meshes to be seamlessly joined together by replicating the vertices along boundaries, but these uses are limited. Given the global nature of the setting, it is usually preferable to explicitly make the boundary faces holes in the areas where surfaces from separate meshes are joined, rather than sharpening edges to interpolate the desired boundaries everywhere else. The remaining edge only and edge and corner choices are then solely distinguished by whether or not the surface at corner vertices is smooth or sharp. Face-varying Interpolation Rules Face-varying interpolation rules control how face-varying data is interpolated both in the interior of face-varying regions smooth or linear and at the boundaries where it is discontinuous constrained to be linear or pinned in a number of ways. Where the topology is continuous and the interpolation chosen to be smooth, the behavior of face-varying interpolation will match that of the vertex interpolation. Choices for face-varying interpolation are most commonly available in the context of UVs for texture coordinates and a number of names for such choices have evolved in different applications over the years. The choices offered by OpenSubdiv cover a wide range of popular applications. The feature is named face-varying linear interpolation -- rather than boundary interpolation commonly used -- to emphasize that it can be applied to the entire surface not just boundaries and that the effects are to make the surface behave more linearly in various ways. The following choices are available for the SdcOptionsFVarLinearInterpolation enum -- the ordering here applying progressively more linear constraints Mode Behavior FVAR_LINEAR_NONE smooth everywhere the mesh is smooth FVAR_LINEAR_CORNERS_ONLY linearly interpolate sharpen or pin corners only FVAR_LINEAR_CORNERS_PLUS1 CORNERS_ONLY sharpening of junctions of 3 or more regions FVAR_LINEAR_CORNERS_PLUS2 CORNERS_PLUS1 sharpening of darts and concave corners FVAR_LINEAR_BOUNDARIES linear interpolation along all boundary edges and corners FVAR_LINEAR_ALL linear interpolation everywhere boundaries and interior These rules cannot make the interpolation of the face-varying data smoother than that of the vertices. The presence of sharp features of the mesh created by sharpness values, boundary interpolation rules, or the subdivision scheme itself e.g. Bilinear take precedence. All face-varying interpolation modes illustrated in UV space using a simple 4x4 grid of quads segmented into three UV regions their control point locations implied by interpolation in the FVAR_LINEAR_ALL case For those familiar, this shape and its assigned UV sets are available for inspection in the catmark_fvar_bound1 shape of OpenSubdivs example and regression shapes. Semi-Sharp Creases Just as some types of parametric surfaces support additional shaping controls to affect creasing along the boundaries between surface elements, OpenSubdiv provides additional sharpness values or weights associated with edges and vertices to achieve similar results over arbitrary topology. Setting sharpness values to a maximum value 10 in this case -- a number chosen for historical reasons effectively modifies the subdivision rules so that the boundaries between the piecewise smooth surfaces are infinitely sharp or discontinuous. But since real world surfaces never really have infinitely sharp edges, especially when viewed sufficiently close, it is often preferable to set the sharpness lower than this value, making the crease semi-sharp. A constant weight value assigned to a sequence of edges connected edges therefore enables the creation of features akin to fillets and blends without adding extra rows of vertices though that technique still has its merits Sharpness values range from 0-10, with a value of 0 or less having no effect on the surface and a value of 10 or more making the feature completely sharp. It should be noted that infinitely sharp creases are really tangent discontinuities in the surface, implying that the geometric normals are also discontinuous there. Therefore, displacing along the normal will likely tear apart the surface along the crease. If you really want to displace a surface at a crease, it may be better to make the crease semi-sharp. Other Options While the preceding options represent features available in a wide-variety of tools and modeling formats, a few others exist whose recognition and adoption is more limited. In some cases, they offer improvements to undesirable behavior of the subdivision algorithms, but their effects are less than ideal. Given both their limited effectiveness and lack of recognition, these options should be used with caution. Chaikin Rule The Chaikin Rule is a variation of the semi-sharp creasing method that attempts to improve the appearance of creases along a sequence of connected edges when the sharpness values differ. This choice modifies the subdivision of sharpness values using Chaikins curve subdivision algorithm to consider all sharpness values of edges around a common vertex when determining the sharpness of child edges. The creasing method can be set using the values defined in the enumeration SdcOptionsCreasingMethod Mode Behavior CREASE_UNIFORM Apply regular semi-sharp crease rules CREASE_CHAIKIN Apply Chaikin semi-sharp crease rules Example of contiguous semi-sharp creases interpolation Triangle Subdivision Rule The triangle subdivision rule is a rule added to the Catmull-Clark scheme that modifies the behavior at triangular faces to improve the undesirable surface artefacts that often result in such areas. Mode Behavior TRI_SUB_CATMARK Default Catmark scheme weights TRI_SUB_SMOOTH Smooth triangle weights Cylinder example This rule was empirically determined to make triangles subdivide more smoothly. However, this rule breaks the nice property that two separate meshes can be joined seamlessly by overlapping their boundaries i.e. when there are triangles at either boundary, it is impossible to join the meshes seamlessly Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"sdc_overview.html\"\n+ \"loc\": \"subdivision_surfaces.html\"\n+ }, {\n+ \"title\": \"Release Notes\",\n+ \"text\": \"Release Notes Release Notes 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Release 2.6.0 Release 2.5.1 Release 2.5.0 Release 2.4.1 Release 2.4.0 Release 2.3.5 Release 2.3.4 Release 2.3.3 Release 2.3.2 Release 2.3.1 Release 2.3.0 Release 2.2.0 Release 2.1.0 Release 2.0.1 Release 2.0.0 Release 1.2.4 Release 1.2.3 Release 1.2.2 Release 1.2.1 Release 1.2.0 Release 1.1.0 Release 1.0.0 Release 2.6.0 New Features Add subdivision kernels for ARM NEON Add OsdUtilVertexSplit which creates a vertex-varying data table by duplicating vertices in a FarMesh Add basic functions to work with FV data via evaluator API Changes Added Catmark restricted vertex compute kernels that optimize for vertices with no semi-sharp creases Fix accessor omissions in osdmesh.h Add support for different subdivision schemes for OsdUtilMesh Bug Fixes Fix crashes when using rather low-end cards like Intel ones Fix a bug in the creation of an edge-vertex kernel batch Fix mismatch in declaration and usage of OsdCudaComputeRestrictedVertexA Fix a bug in the vertex order for restricted Catmark vertex-vertex kernel batches Fix a bug in FarCatmarkSubdivisionTablesFactory that prevented the CATMARK_QUAD_FACE_VERTEX kernel from being selected for subdivision level 2 or greater. Fix a bug in OsdUtilVertexSplit that occurs when getting the address of the end of a stdvector Fix error in createCLBuffer that occurs when the buffer size is zero Fix a bug in the CUDA computeRestrictedEdge kernel Fix duplicate variables with identical name Fix osdutil build errors Fix cmake diagnostic messsage Release 2.5.1 New Features Add CATMARK_QUAD_FACE_VERTEX and CATMARK_TRI_QUAD_FACE_VERTEX compute kernels optimization that takes advantage of all-quads or all-triange-and-quads meshes Bug Fixes Fix a compiler error in the GLSL Transform Feedback kernels on OS X Fix boundary interpolation in osdutil Fix bilinear stencil tangent computions Release 2.5.0 New Features Add ability to generate triangle patches for a uniformly subdivided mesh Add new example topologySharing Add interleaved buffer mode in glViewer Add GLSL compute kernel to glBatchViewer Add TBB compute kernel to glBatchViewer Add a PullDown widget to our HUD in examplescommon GUI updates cosmetic changes to GL example code Adding a programmable image shader to gl_hud Code cleanup for GLFrameBuffer in examplescommon Implement C-API accessor to evaluator topology osdutil Add command line option to CMakes options Add a CMake option to disable OpenCL Add a FindCLEW.cmake module in anticipation of using CLEW as a dependency Integrate CLEW into osd library and examples Changes Change interleaved buffer support in OsdCompute Removed OsdVertexDescriptor and replaced with OsdVertexBufferDescriptor Reorganize ComputeContext and ComputeController. Reorganize EvalStencilContext and EvalStencilController Moved transient states current vertex buffer etc to controller Reorganize EvalLimitContext and EvalLimitController Moved transient states current vertex buffer etc to controller Fix adaptive isolation of sharp corner vertices Fix incorrect FarMeshFactory logic for isolating multiple corner vertices in corner patches Change EvalLimit Gregory patch kernels to the large weights table to accomodate higher valences Fix calculation of screen space LOD tess factors for transition corner patches. Add a public constructor to OsdMesh Decrease compiler warning thresholds and fix outstanding warnings Make PTex support optional Add a NO_MAYA flag to CMakeLists to disable all Autodesk Maya dependencies in the build Document NO_MAYA command line option Bug Fixes Fix mistakenly deleted memory barrier in glsl OsdCompute kernel. Fix shape_utils genRIB function to use streams correctly. Temporary workaround for the synchronization bug of glsl compute kernel Fix Hud display for higher DPI MBP retina Fix Hud d3d11 Fix examples to use GL timer query to measure the GPU draw timing more precisely Fix glViewer stop updating during freeze. Fix file permissions on farPatchTablesFactory.h Fix some meory leaks in adaptive evaluator osdutil Fix OsdUtilAdaptiveEvaluator concurrency issue Fix OsdUtilRefiner incorrect Invalid size of patch array error reporting. Fix OsdUtilPatchPartitioner failure for triangle patches Fixes a bug that causes OsdUtilPatchPartitioner to fail to rebuild the face-varying data table correctly for triangle patches. Add missing third parameter to templated OsdDrawContext usage osdutilbatch.h Return success status from openSubdiv_finishEvaluatorDescr osdutil Remove debugging stdcout calls osdutil Build errors warnings Fix OSX Core Profile build GLFrameBuffer Fix ptexViewer build error on OSX Fix framebuffer shader compiling for OSX Reordering includes to address a compile error on OSXglew environment Fix compilation errors with CLEW enabled Fix icc build problems Fix compiler warnings in OsdClVertexBuffer Fix compilation error on windowsmsvc2013 Fix build warningserrors with VS2010 Pro Fix Windows build warning in FarPatchTablesFactory Fix doxygen generation errors Release 2.4.1 Changes Add correct OpenSubdiv namespace beginend blocks. Bug Fixes Compile osdutil with -fPIC for correct linking. Fix a bug of OsdUtilMeshBatch, the varying buffer isnt computed with CL kernels Fix FindGLFW.cmake to use the GLFW_LOCATION environment variable in Windows Fix Draw contexts do not fully initialize patch arrays Release 2.4.0 New Features Adding functionality to store uniform face-varying data across multiple levels of subdivision Add OsdUtilPatchPartitioner. It splits patcharray into subsets so that clients can draw partial surfaces for both adaptive and uniform. Changes Remove FarMesh dependency from OsdContext. Use DSA APIs for GL buffer update if available. Refactor Far API replace void- of all kernel applications with CONTEXT template parameter. It eliminates many static_casts from void- for both far and osd classes. move the big switch-cases of far default kernel launches out of Refine so that osd controllers can arbitrary mix default kernels and custom kernels. change FarKernelBatchkernelType from enum to int, clients can add custom kernel types. remove a back-pointer to farmesh from subdivision table. untemplate all subdivision table classes and template their compute methods instead. Those methods take a typed vertex storage. remove an unused argument FarMesh from the constructor of subdivision table factories. Refactor FarSubdivisionTables. Delete scheme specialized subdivision tables. The base class FarSubdivisionTables already has all tables, so we just need scheme enum to identify which scheme the subdivision tables belong to. This brings a lot of code cleanups around far factory classes. Move FarMultiMeshFactory to OsdUtil. Move table splicing functions of FarMultiMeshFactory into factories Change PxOsdUtil prefix to final OsdUtil prefix. Improve error reporting in osdutil refinement classes, and fix a build issue Bug Fixes Fix another multi mesh splicing bug of face varying data. Make CMake path variables more robust Fixing a crash on Marvericks wglew Update dxViewer example documentation Fix wrong logic in openSubdiv_setEvaluatorCoarsePositions Remove debug print from adaptive evaluators initialization Release 2.3.5 New Features Add the ability to read obj files to the dxViewer example Add screen-capture function to ptexViewer Update documention for Xcode builds Add documentation boundary interpolation rules and face-varying boundary interpolation rules Changes Refactoring FarPatchTables and FarPatchTablesFactory Move GL vertex buffer VBO buffer allocation out of allocate and into BindVBO Enable uvViewer on OS X now that Mavericks is released. Replacing un-necessary dynamic_cast with reinterpret_cast within FarDispatcher Minor code cleanup of FarMeshFactory Remove address space qualifiers from OpenCL kernel functions Fix OpenCL initialization to be slightly more robust Add OpenCL header include paths where necessary Add static specifiers for non-kernel CL funcs at program scope Add stddef.h to pythonosdosdshim.i Modify ptexViewer and uvViewer shaders to address some portability issues Bug Fixes Fix Gregory Boundary patch buffer overrun Fix black texels when the resolution of a ptex face is less than 4 Fix a splicing bug in FarMultiMeshFactory Fix a build error when using older versions of GLFW Fix build warnings optimized Fix FindTBB.cmake Fix FindMaya.cmake Fix glViewer support for GLSL compute Fix ptexViewer enable specular pass in both IBL and point lighting Fix Zlib include in ptexViewer Fix ptexViewer shader errors. Fix osdPolySmooth Maya plugin Fix UV merging in osdPolySmooth code example Add cleanup function to osdPolySmooth Maya plugin Fix Maya OsdPolySmooth node component output Fix GLSL array instantiation syntax for glStencilViewer Fix examples to run correctly on high DPI displays with GLFW 3 Release 2.3.4 New Features Adding CPUOMPTBB Context Controller pairs for CPU evaluation of smooth normals Added adaptiveEvaluator class inspired by Sergeys work in blender OsdUtil Changes Changed the HUD to ignore mouse clicks when not visible. Updates for blender development OsdUtil Add C compatible API to access the adaptiveEvaluator class from non-C OsdUtil Update license headers to apache OsdUtil CMake build improvement make osd a cmake object library remove compiling redundancies Improve stringification of shaders kernels in CMake build Bug Fixes Fixed iOS build Fixed VS2010 warningserrors. Fix OsdCpuEvalLimitKernel Fix maxvalence calculation in FarMeshFactory Fix FarStencilFactory control stencil caching Removing assert for high-valence vertices running off limit tangent pre-computed table. Fix degenerate stencil limit tangent code path. Fix unused variable build warnings gcc 4.8.2 - Fedora 19 Fix build warning from osdutiladaptiveEvaluator.cpp Release 2.3.3 Changes Modify Far remapping of singular vertices to point to their source vertex. Refactoring Ptex Mipmap and Analytic Displacement code Adding some documentation for Chaikin crease rule Misc. improvements to PxOsdUtilsMesh Adding recommended isolation output to OsdPolySmooth node Bug Fixes Adding an error check on version parsing of main CMakeLists Fix regex in FindMaya.cmake that breaks with recent versions of Maya Fix crashes induced by typeid Fixed VS2010 build warning Fix build break in hbr_regression Fix incorrect capitalization in GL ptexViewer shader.glsl Fix OSX build add stdlib.h include Release 2.3.2 New Features Adding control cage drawing to ptexViewer Adding Maya osdPolySmooth plugin into OpenSubdiv examples. Changes Removing some glGetError checks that are causing problems for Autodesk D3D11DrawRegistry returns the common shader config for all non-tess patcharrays. Updates to simple cpu osdutil classes Bug Fixes Fix Hbr Chaikin crease rule Fix Chaikin tag parsing Fix return value of allocate function for OsdCPUGLVertxBuffer Fixed GLSL shader portability. Fix FindGLFW.cmake for GLFW 3.03 on OSX Fixed compiler warnings. Fixed VS2010 build errors Fixed WIN32 build error when no DXSDK installed. Fix OSX build stdlib.h needs to be included in glPtexMipmapTexture.h Fix for crash in new meshrefiner code in OsdUtil Release 2.3.1 New Features Add DX11 version of ptex mipmap loader Add DX11 ptex viewer work in progress Add DX11 fractional partitioning, normal derivatives computation Add memory usage controls to Ptex loader Add face-varying boundary interpolation parsing to shape_utils Add simple HbrMesh and FarMesh wrapper classes to osdutil Changes Amend language of attribution file NOTICE.txt Optimize a bit of ptex mipmap lookup. Show ptex memory usage in GL and DX11 ptexViewers Improve ptex guttering Addding some video links to our collection of external resources Bug Fixes Fix edge-only face-varying interpolation Fix Far to handle disconnected vertices in an Hbr mesh Fixed ptex cache resource release sequence Fix build symbol conflict in Far Fix patch parambuffer generation in OsdD3D11DrawContext Fix a minor osdutil build warning seen with gcc 4.8.1 Fix VS2010 build errors Release 2.3.0 New Features Added Analytical displacement mapping Analytic Displacement Mapping using Hardware Tessellation Niessner and Loop TOG 2013 Added a new ptex mipmap loader Added face varying macros for loop subdivision Added the uvViewer example to see how face varying interpolation rule works Added a slider component and cleanup hud code. Changes Adding license attribution files, improved language of the code headers Install documentation into the Filesystem Hierarchy Standard location Set GLFW_OPENGL_FORWARD_COMPAT on Mac OS to make samples work on that platform Added surface normal mode mipmap to ptxViewer Bug Fixes Fix a bug of bad fvar splicing for loop surface. Fix incorrect bilinear limit tangents in FarStencilTablesFactory Fix boundary interpolation rules doc Added an error check on updating cuda buffer Fix face varying rendering on loop surface Fixed glBatchViewer build for GLFW 2.x Expand search paths for FindGLFW.cmake for Debian and other Linux architectures Fix CMake executable builds for ICC Fix bhr baseline regression, so reference files are real OBJs Fixed clKernelBundle.cpp to build on Android. Fix misc build warings Release 2.2.0 New Features Added subdivision stencil functionality Far OsdEval Bug Fixes Fix D3D11DrawContext to check for NULL pointers Fix cpuEvalLimitController crash bug Fixed search path suffixes for ICC libs Fixed invalid initialization of glslTransformFeedback kernel. Release 2.1.0 New Features Added TBB Compute back-end on Linux contribution from Sheng Fu Added support for ICC compiler still Beta Changes Added constructor to OsdMesh with a FarMesh as input Modify CMake to name and sym-link DSOs based on Linux ABI versioning spec Added command line input to DX11 viewer FarMultiMesh can splice uniform and adaptive meshes together. Bug Fixes Fix FarMultiMesh splicing Removed unnecessary cudaThreadSynchronize calls. Fix glViewer overlapping HUD menus Fix facevarying rendering in glBatchViewer Fix build of GLSL transform feedback kernels Fix Getting Started documentation Release 2.0.1 New Features New CLA files to reflect Apache 2.0 licensing Changes Move all public headers to includeopensubdiv... Adding Osd documentation based on Siggraph slides Bug Fixes Fix incorrect transition pattern 3 in GLSL HLSL shaders Fix CMake build to not link GPU-based libraries into libosdCPU Fix support for GLEW on OSX Fix GLFW Xrandr xf86vmode dependency paths for X11 based systems Fix HUD display overlaps in code examples Fix FindGLEW.cmake to be aware of multiarch on linux systems Fix some hard-coded include paths in CMake build Release 2.0.0 New Features New CMake build flags NO_LIB, NO_CUDA, NO_PYTHON Changes OpenSubdiv is now under Apache 2.0 license HbrHalfedge and HbrFVarData copy constructors are now private Documentation style matched to graphics.pixar.com new content Add an animation freeze button to ptexViewer Variable name changes for better readability across all example shader code Bug Fixes Fix incorrect patch generation for patches with 2 non-consecutive boundary edges Fix undefined gl_PrimitiveID shader build errors Fix for shader macro OSD_DISPLACEMENT_CALLBACK Fix out-of-bounds stdvector access in FarPatchTablesFactory Release 1.2.4 New Features Adding support for fractional tessellation of patches Adding a much needed API documention system based on Docutils RST markup Adding support for face-varying interpolation in GLSL APIs Adding varying data buffers to OsdMesh Adding accessors to the vertex buffers in OsdGlMesh Adding face-varying data to regression shapes Changes Cleanup of common bicubic patch shader code GLSL HLSL for portability ATI OSX drivers Bug Fixes Fix FarVertexEditTablesFactory to insert properly vertex edit batches fixes incorrect hierarchical hole in regression shape Fix FarPatchMap quadtree to not drop top-level non-quad faces Fix Gregory patches bug with incorrect max-valence Fix FarPatchTablesGetNumFaces and FarPatchTablesGetFaceVertices functions to return the correct values Fix face indexing GLSL code ptex works on non-quads again Fix face-varying data splicing in FarMultiMeshFactory Fix ptex face indexing in FarMultiMeshFactory Fix glew include to not break builds Fix Clang ICC build failures with FarPatchTables Fix build and example code to work with GFLW 3.0 Fix cmake to have ptex dynamically linked in OSX Release 1.2.3 New Features Adding Varying and Face-Varying data interpolation to EvalLimit Changes EvalLimit API refactor the EvalContext now has dedicated structs to track all the vertex, varying and face-varying data streams. Also renamed some buffers into tables to maintain code consistency EvalLimit optimization switch serial indexing to a quad-tree based search Bug Fixes Face-varying data bug fixes making sure the data is carried around appropriately Fixes for OpenCL use with the new batching APIs GLSL general shader code cleanup fixes for better portability GLSL Tranform Feedback initialization fix Critical fix for FarMultiMesh batching indexing was incorrect Fix osdutil CL implementation protect includes on systems with no OpenCL SDK installed Fix face-varying interpolation on adaptive patches FarPatchTables fix IsFeatureAdaptive to return the correct answer Fix Far factories to handle the absence of face-varying data correctly. Many GLSL shader code style fixes which should help with ATI OSX shader compiling Release 1.2.2 New Features Introducing the EvalLimit API the Eval module aims at providing support for computational tasks that are not related to drawing the surfaces. The EvalLimit sub-module provides an API that enables client code to evaluate primitive variables on the limit surface. OsdxxxComputeController minor optimization. Added early exit to Refine method to avoid unnecessary interop. Changes OsdGLDawContext minor API change. Protecting some member variables and adding const accessors OsdError minor API refactor, added Warning functions. Bug Fixes Fix Ptex bug prevent corner texel guttering code to from going into infinite loops Adding the ability for a FarMeshFactory to construct patchTables starting from firstLevel in uniform subdivision mode Consolidating the color coding of bicubic patch types through all our our code examples this is used mostly as a debugging tool Fixing some MSVC build warnings Update to the outdated README.md Release 1.2.1 New Features Added CUDA runtime error checking Release 1.2.0 Changes Major Far refactor around patchTables to introduce the draw batching API Renaming osd_util to osdutil Bug Fixes Fix GLSL transform feedback initialization bug in ptexViewer Minor bug typo fixes Release 1.1.0 New Features release initiated because of the switch to Git Flow Release 1.0.0 Oringal release Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"tags\": \"\",\n+ \"loc\": \"release_notes_2x.html\"\n }, {\n \"title\": \"3.0 - 3.5 Release Notes\",\n- \"text\": \"3.0 - 3.5 Release Notes 3.0 - 3.5 Release Notes 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Release 3.5 Release 3.5.0 - Sep 2022 Release 3.4 Release 3.4.4 - Feb 2021 Release 3.4.3 - Apr 2020 Release 3.4.0 - Jun 2019 Release 3.3 Release 3.3.3 - Jul 2018 Release 3.3.2 - Jun 2018 Release 3.3.1 - Feb 1018 Release 3.3.0 - Aug 2017 Release 3.2 Release 3.2.0 - Feb 2017 Release 3.1 Release 3.1.1 - Jan 2017 Release 3.1.0 - Oct 2016 Release 3.0 Release 3.0.5 - Mar 2016 Release 3.0.4 - Feb 2016 Release 3.0.3 - Oct 2015 Release 3.0.2 - Aug 2015 Release 3.0.1 - Aug 2015 Release 3.0.0 - Jun 2015 Release 3.0.0 RC2 Release 3.0.0 RC1 Previous 2.x Release Notes Release 3.5 Release 3.5.0 - Sep 2022 Release 3.5.0 is a significant release with new features, several configuration improvements, and a few other improvements and bug fixes. For more information on the following, see Release 3.5 Deprecation Announcements Hbr is deprecated and will be removed from subsequent releases New Features Simplified Surface Evaluation Bfr Tessellation Patterns Bfr Changes Suppression of GCC compiler warnings GitHub 1253, 1254, 1270 Additional methods for FarTopologyLevel GitHub 1227, 1255 Improved mixed partial derivative at Gregory patch corners GitHub 1252 Minor improvements to Far tutorials GitHub 1226, 1241 Added CMake config GitHub 1242 Updated CMake minimum version to 3.12 GitHub 1237, 1261 Updated documentation build scripts for Python 3 1265, 1266 Updated stringify build tool for improved cross compilation support GitHub 1267 Added NO_MACOS_FRAMEWORKS build option GitHub 1238 Updated Azure pipelines agents for Unbuntu and macOS GitHub 1247, 1256 Removed obsolete AppVeyor and Travis CI scripts GitHub 1259 Bug Fixes Cache active program for OsdGLComputeEvaluator GitHub 1244 Fixed member initialization warnings in OsdD3D11ComputeEvaluator GitHub 1239 Fixed GLSL shader source to remove storage qualifiers from struct members GitHub 1271 Fixed use of CMake variables for Apple builds GitHub 1235 Fixed build errors when using OpenGL without GLFW GitHub 1257 Fixed links to embedded videos GitHub 1231 Release 3.4 Release 3.4.4 - Feb 2021 Release 3.4.4 is a minor release including bug fixes and configuration improvements Changes The master branch on GitHub has been renamed release GitHub 1218 1219 The CMake configuration has been updated to allow use as a sub-project GitHub 1206 Removed obsolete references to hbr from examplesfarViewer GitHub 1217 Bug Fixes Fixed bug with sparse PatchTables and irregular face-varying seams GitHub 1203 Fixed loss of precision when using double precision stencil tables GitHub 1207 Fixed reset of FarTopologyRefinerGetMaxLevel after call to Unrefine GitHub 1208 Fixed linking with -ldl on unix systems GitHub 1196 Fixed naming and installation of macOS frameworks GitHub 1194 1201 Fixed GL version and extension processing and dynamic loading on macOS GitHub 1216 Fixed FindDocutils.cmake to be more robust GitHub 1213 1220 Fixed errors using build_scriptsbuild_osd.py with Python3 GitHub 1206 Release 3.4.3 - Apr 2020 Release 3.4.3 is a minor release including bug fixes and configuration improvements Changes GLEW is no longer required by default GitHub 1183 1184 Removed false Ptex link dependency from libosdCPU GitHub 1174 Removed false GLFW link dependency from DX11 and Metal examples GitHub 1178 Removed link dependency on unused TBB libraries GitHub 1064 Added option to disable building of dynamic shared libraries GitHub 1169 Added new tutorial for FarLimitStencilTable GitHub 1176 Updated use of EXT_direct_state_access to ARB_direct_state_access GitHub 1184 Fixed C strict aliasing warnings GitHub 1182 Fixed MSVC warnings in example code GitHub 1158 1172 Fixed compatibility with Visual Studio 2019 GitHub 1173 1189 Fixed CMake CMP0054 warnings GitHub 1180 Added prefix to OpenSubdiv CMake macros GitHub 1157 Moved utilities in examplescommon to regressioncommon GitHub 1167 Minor fixes to Far tutorials GitHub 1175 1177 Switched to Azure Pipelines for continuous integration testing instead of Travis-CI and AppVeyor GitHub 1168 1190 Bug Fixes Fixed selective boundary interpolation for case SdcOptionsVTX_BOUNDARY_NONE GitHub 1170 1171 Fixed static library linking to address missing symbols GitHub 1192 Additional fixes for dynamic and static linking GitHub 1193 Release 3.4.0 - Jun 2019 Release 3.4.0 is a significant release with several new features, bug fixes, and general code and configuration improvements. For more information on the following, please see Release 3.4 New Features Triangular Patches for Loop subdivision Improvements to Introductory Documentation Sparse Patch Tables and Adaptive Refinement Full Support for Double Precision in Far Changes Added new build script GitHub 1068 Added support for newer DirectX SDKs GitHub 1066 Patch arrays extended to support combined regular and irregular types GitHub 995 FarPatchTables and adaptive refinement supported for Bilinear scheme GitHub 1035 New FarPatchTableFactory method to determine adaptive refinement options GitHub 1047 New FarPatchTableFactory options to align primvar buffers of uniform tables GitHub 986 FarStencilTableUpdateValues overloaded to support separate base buffer GitHub 1011 FarLimitStencilTableFactory updated to create face-varying tables GitHub 1012 Regular patches on boundaries no longer require additional isolation GitHub 1025 Inclusion of OpenSubdiv header files in source code now consistent GitHub 767 Re-organization of and additions to Far tutorials GitHub 1083 examples now use common command-line conventions and parsing GitHub 1056 Bug Fixes Fixed FarPrimvarRefiner internal limitFVar prototype GitHub 979 Fixed FarStencilTable append when base StencilTable empty GitHub 982 Patches around non-manifold vertices now free of cracks GitHub 1013 Release 3.3 Release 3.3.3 - Jul 2018 Release 3.3.3 is bug-fix release addressing regressions from release 3.3.2 Bug Fixes Fixed a regression in PatchTable construction with varying patches GitHub 976 Fixed a regression in PatchTable construction for face-varying patches GitHub 972 Fixed a bug in the initialization of FarSourcePatch GitHub 971 Release 3.3.2 - Jun 2018 Release 3.3.2 is a minor release with potentially significant performance improvements to the patch pre-processing stages Changes Improved performance of PatchTable construction GitHub 966 The resulting improved accuracy will produce slight numerical differences in computations involving patches, e.g. StencilTable and PatchTable evaluation Bug Fixes FarPatchTableFactory now supports PatchTable construction with ENDCAP_BILINEAR_BASIS specified Release 3.3.1 - Feb 1018 Release 3.3.1 is a minor bug-fix release Bug Fixes Fixed GLSLHLSLMetal patch shader code to resolve degenerate normals GitHub 947 Fixed problems with face-varying patches in uniform PatchTables GitHub 946 Fixed integer overflow bugs for large meshes in PatchTable factories GitHub 957 Fixed computation of PatchParam for triangle refinement GitHub 962 Changes Added build options NO_GLFW and NO_GLFW_X11 Added additional shapes with infinitely sharp creases to the Metal and DX11 example viewers Disabled GL tests during CI runs on Linux Improved stability of examplesglImaging in CI runs by testing GL version Release 3.3.0 - Aug 2017 Release 3.3.0 is significant release adding an Osd implementation for Apples Metal API New Features Added an Osd implementation for Apples Metal API Added the mtlViewer example Changes Fixed several instances of local variable shadowing that could cause build warnings Updated continuous-integration build scripts and added testing on macOS Release 3.2 Release 3.2.0 - Feb 2017 Release 3.2.0 is a minor release containing API additions and bug fixes New Features Extended FarStencilTableFactory to support face-varying Extended Osd Evaluator classes to support evaluation of 1st and 2nd derivatives Added an option to disable generation of legacy sharp corner patches Changes Corrected numerous spelling errors in doxygen comments Updated glFVarViewer with improved error detection and command line parsing Added option to build using MSVC with static CRT Bug Fixes Fixed a double delete of GL program in OsdGLComputeEvaluator Release 3.1 Release 3.1.1 - Jan 2017 Release 3.1.1 is a minor bug-fix release. Bug Fixes Fixed a bug with non-manifold face-varying topology causing a crash during patch table creation Fixed GLEW compilation and linking with dynamic GLEW libraries on Windows Fixed GLFW linking with GLFW 3.2 on X11 platforms Release 3.1.0 - Oct 2016 Release 3.1.0 is a significant release with several new features, bug fixes, and general code and configuration improvements. For more information on the following, please see Release 3.1 New Features Bicubic Face-Varying Patches Varying and Face-Varying Evaluation Second Order Derivative Evaluation Separate Levels of Feature Isolation Sharp Patches for Infinitely Sharp Features Changes Enabled the use of CMakes folder feature Removed the use of iso646 alternative keywords and, or, not, etc. to improve portability Added numerical valued preprocessor directives OPENSUBDIV_VERSION_MAJOR, etc. to opensubdivversion.h Improved documentation for FarPatchParam and added Unnormalize to complement Normalize Added additional topology queries to FarTopologyLevel Updated glFVarViewer and glEvalLimit viewer to make use of bicubic face-varying patches Updated glViewer and dxViewer to add a toggle for InfSharpPatch Updated dxPtexViewer for improved feature parity with glPtexViewer Improved far_regression to exercise shapes independent of Hbr compatibility Added support for Appveyor continuous integration testing Removed cmakeFindIlmBase Removed mayaPolySmooth example Bug Fixes Fixed Ptex version parsing and compatibility issues Fixed compatibility issues with VS2015 Fixed bug interpolating face-varying data with Bilinear scheme Fixed bug with refinement using Chaikin creasing Fixed bugs with HUD sliders in the example viewers Release 3.0 Release 3.0.5 - Mar 2016 Release 3.0.5 is a minor stability release with performance and correctness bug fixes. Bug Fixes The previous release reduced transient memory use during PatchTable construction, but increased the amount of memory consumed by the resulting PatchTable itself, this regression has been fixed. The example Ptex texture sampling code has been fixed to prevent sampling beyond the texels for a face when multisample rasterization is enabled. Release 3.0.4 - Feb 2016 Release 3.0.4 is a minor stability release which includes important performance and bug fixes. New Features Added accessor methods to FarLimitStencilTable to retrieve limit stencil data including derivative weights Added support for OpenCL event control to OsdCLVertexBuffer and OsdCLEvaluator Changes Major reduction in memory use during FarPatchTable construction for topologies with large numbers of extraordinary features Improved performance for GL and D3D11 tessellation control hull shader execution when drawing BSpline patches with the single crease patch optimization enabled Bug Fixes Restored support for drawing with fractional tessellation Fixed far_tutorial_6 to refine primvar data only up to the number of levels produced by topological refinement Fixed build warnings and errors reported by Visual Studio 2015 Release 3.0.3 - Oct 2015 Release 3.0.3 is a minor stability release which includes important performance and bug fixes. New Features Smooth normal generation tutorial, far_tutorial_8 Changes Major performance improvement in PatchTable construction Improved patch approximations for non-manifold features Bug Fixes Fixed double delete in GLSL Compute controller Fixed buffer layout for GLSL Compute kernel Fixed GL buffer leak in OsdGLPatchTable Fixed out-of-bounds data access for TBB and OMP stencil evaluation Fixed WIN32_LEAN_AND_MEAN typo Fixed Loop-related shader issues glFVarViewer Release 3.0.2 - Aug 2015 Release 3.0.2 is a minor release for a specific fix. Bug Fixes Fixed drawing of single crease patches Release 3.0.1 - Aug 2015 Release 3.0.1 is a minor release focused on stability and correctness. Changes Added a references section to the documentation, please see References Removed references to AddVaryingWithWeight from examples and tutorials Added more regression test shapes Addressed general compiler warnings e.g. signed vs unsigned comparisons Addressed compiler warnings in the core libraries reported by GCCs -Wshadow Eased GCC version restriction, earlier requirement for version 4.8 or newer is no longer needed Replaced topology initialization assertions with errors Improved compatibility with ICC Improved descriptive content and formatting of Far error messages Improved build when configured to include no GPU specific code Bug Fixes Fixed handling of unconnected vertices to avoid out of bounds data access Fixed non-zero starting offsets for TbbEvalStencils and OmpEvalStencils Fixed FarStencilTableFactoryOptionsfactorizeIntermediateLevels Fixed FarPatchTablesFactoryOptionsgenerateAllLevels Fixed the behavior of VTX_BOUNDARY_NONE for meshes with bilinear scheme Fixed some template method specializations which produced duplicate definitions Disabled depth buffering when drawing the UI in the example viewers Disabled the fractional tessellation spacing option in example viewers since this mode is currently not supported Release 3.0.0 - Jun 2015 Release 3.0.0 is a major release with many significant improvements and changes. For more information on the following, please see Release 3.0 New Features Faster subdivision using less memory Support for non-manifold topology Face-Varying data specified topologically Elimination of fixed valence tables Single-crease patch for semi-sharp edges Additional irregular patch approximations Introduction of Stencil Tables Faster, simpler GPU kernels Unified adaptive shaders Updated coding style with namespaces More documentation and tutorials Bug Fixes Smooth Face-Varying interpolation around creases Release 3.0.0 RC2 New Features Documentation updates far_tutorial_3 updates for the multiple face-varying channels maya example plugin interpolates a UV channel and a vertex color channel Bug Fixes Fixed a LimitStencilTableFactory bug, which returns an invalid table PatchParam encoding changed to support refinement levels up to 10 Added Xinerama link dependency Fixed MSVC 32bit build problem Fixed minor cmake issues Fixed glViewerfarViewer stability bugs Release 3.0.0 RC1 Changes FarTopologyRefiner was split into several classes to clarify and focus the API. Interpolation of Vertex and Varying primvars in a single pass is no longer supported. The Osd layer was largely refactored. Previous 2.x Release Notes Previous releases Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"3.0 - 3.5 Release Notes 3.0 - 3.5 Release Notes 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Release 3.5 Release 3.5.0 - Sep 2022 Release 3.4 Release 3.4.4 - Feb 2021 Release 3.4.3 - Apr 2020 Release 3.4.0 - Jun 2019 Release 3.3 Release 3.3.3 - Jul 2018 Release 3.3.2 - Jun 2018 Release 3.3.1 - Feb 1018 Release 3.3.0 - Aug 2017 Release 3.2 Release 3.2.0 - Feb 2017 Release 3.1 Release 3.1.1 - Jan 2017 Release 3.1.0 - Oct 2016 Release 3.0 Release 3.0.5 - Mar 2016 Release 3.0.4 - Feb 2016 Release 3.0.3 - Oct 2015 Release 3.0.2 - Aug 2015 Release 3.0.1 - Aug 2015 Release 3.0.0 - Jun 2015 Release 3.0.0 RC2 Release 3.0.0 RC1 Previous 2.x Release Notes Release 3.5 Release 3.5.0 - Sep 2022 Release 3.5.0 is a significant release with new features, several configuration improvements, and a few other improvements and bug fixes. For more information on the following, see Release 3.5 Deprecation Announcements Hbr is deprecated and will be removed from subsequent releases New Features Simplified Surface Evaluation Bfr Tessellation Patterns Bfr Changes Suppression of GCC compiler warnings GitHub 1253, 1254, 1270 Additional methods for FarTopologyLevel GitHub 1227, 1255 Improved mixed partial derivative at Gregory patch corners GitHub 1252 Minor improvements to Far tutorials GitHub 1226, 1241 Added CMake config GitHub 1242 Updated CMake minimum version to 3.12 GitHub 1237, 1261 Updated documentation build scripts for Python 3 1265, 1266 Updated stringify build tool for improved cross compilation support GitHub 1267 Added NO_MACOS_FRAMEWORKS build option GitHub 1238 Updated Azure pipelines agents for Unbuntu and macOS GitHub 1247, 1256 Removed obsolete AppVeyor and Travis CI scripts GitHub 1259 Bug Fixes Cache active program for OsdGLComputeEvaluator GitHub 1244 Fixed member initialization warnings in OsdD3D11ComputeEvaluator GitHub 1239 Fixed GLSL shader source to remove storage qualifiers from struct members GitHub 1271 Fixed use of CMake variables for Apple builds GitHub 1235 Fixed build errors when using OpenGL without GLFW GitHub 1257 Fixed links to embedded videos GitHub 1231 Release 3.4 Release 3.4.4 - Feb 2021 Release 3.4.4 is a minor release including bug fixes and configuration improvements Changes The master branch on GitHub has been renamed release GitHub 1218 1219 The CMake configuration has been updated to allow use as a sub-project GitHub 1206 Removed obsolete references to hbr from examplesfarViewer GitHub 1217 Bug Fixes Fixed bug with sparse PatchTables and irregular face-varying seams GitHub 1203 Fixed loss of precision when using double precision stencil tables GitHub 1207 Fixed reset of FarTopologyRefinerGetMaxLevel after call to Unrefine GitHub 1208 Fixed linking with -ldl on unix systems GitHub 1196 Fixed naming and installation of macOS frameworks GitHub 1194 1201 Fixed GL version and extension processing and dynamic loading on macOS GitHub 1216 Fixed FindDocutils.cmake to be more robust GitHub 1213 1220 Fixed errors using build_scriptsbuild_osd.py with Python3 GitHub 1206 Release 3.4.3 - Apr 2020 Release 3.4.3 is a minor release including bug fixes and configuration improvements Changes GLEW is no longer required by default GitHub 1183 1184 Removed false Ptex link dependency from libosdCPU GitHub 1174 Removed false GLFW link dependency from DX11 and Metal examples GitHub 1178 Removed link dependency on unused TBB libraries GitHub 1064 Added option to disable building of dynamic shared libraries GitHub 1169 Added new tutorial for FarLimitStencilTable GitHub 1176 Updated use of EXT_direct_state_access to ARB_direct_state_access GitHub 1184 Fixed C strict aliasing warnings GitHub 1182 Fixed MSVC warnings in example code GitHub 1158 1172 Fixed compatibility with Visual Studio 2019 GitHub 1173 1189 Fixed CMake CMP0054 warnings GitHub 1180 Added prefix to OpenSubdiv CMake macros GitHub 1157 Moved utilities in examplescommon to regressioncommon GitHub 1167 Minor fixes to Far tutorials GitHub 1175 1177 Switched to Azure Pipelines for continuous integration testing instead of Travis-CI and AppVeyor GitHub 1168 1190 Bug Fixes Fixed selective boundary interpolation for case SdcOptionsVTX_BOUNDARY_NONE GitHub 1170 1171 Fixed static library linking to address missing symbols GitHub 1192 Additional fixes for dynamic and static linking GitHub 1193 Release 3.4.0 - Jun 2019 Release 3.4.0 is a significant release with several new features, bug fixes, and general code and configuration improvements. For more information on the following, please see Release 3.4 New Features Triangular Patches for Loop subdivision Improvements to Introductory Documentation Sparse Patch Tables and Adaptive Refinement Full Support for Double Precision in Far Changes Added new build script GitHub 1068 Added support for newer DirectX SDKs GitHub 1066 Patch arrays extended to support combined regular and irregular types GitHub 995 FarPatchTables and adaptive refinement supported for Bilinear scheme GitHub 1035 New FarPatchTableFactory method to determine adaptive refinement options GitHub 1047 New FarPatchTableFactory options to align primvar buffers of uniform tables GitHub 986 FarStencilTableUpdateValues overloaded to support separate base buffer GitHub 1011 FarLimitStencilTableFactory updated to create face-varying tables GitHub 1012 Regular patches on boundaries no longer require additional isolation GitHub 1025 Inclusion of OpenSubdiv header files in source code now consistent GitHub 767 Re-organization of and additions to Far tutorials GitHub 1083 examples now use common command-line conventions and parsing GitHub 1056 Bug Fixes Fixed FarPrimvarRefiner internal limitFVar prototype GitHub 979 Fixed FarStencilTable append when base StencilTable empty GitHub 982 Patches around non-manifold vertices now free of cracks GitHub 1013 Release 3.3 Release 3.3.3 - Jul 2018 Release 3.3.3 is bug-fix release addressing regressions from release 3.3.2 Bug Fixes Fixed a regression in PatchTable construction with varying patches GitHub 976 Fixed a regression in PatchTable construction for face-varying patches GitHub 972 Fixed a bug in the initialization of FarSourcePatch GitHub 971 Release 3.3.2 - Jun 2018 Release 3.3.2 is a minor release with potentially significant performance improvements to the patch pre-processing stages Changes Improved performance of PatchTable construction GitHub 966 The resulting improved accuracy will produce slight numerical differences in computations involving patches, e.g. StencilTable and PatchTable evaluation Bug Fixes FarPatchTableFactory now supports PatchTable construction with ENDCAP_BILINEAR_BASIS specified Release 3.3.1 - Feb 1018 Release 3.3.1 is a minor bug-fix release Bug Fixes Fixed GLSLHLSLMetal patch shader code to resolve degenerate normals GitHub 947 Fixed problems with face-varying patches in uniform PatchTables GitHub 946 Fixed integer overflow bugs for large meshes in PatchTable factories GitHub 957 Fixed computation of PatchParam for triangle refinement GitHub 962 Changes Added build options NO_GLFW and NO_GLFW_X11 Added additional shapes with infinitely sharp creases to the Metal and DX11 example viewers Disabled GL tests during CI runs on Linux Improved stability of examplesglImaging in CI runs by testing GL version Release 3.3.0 - Aug 2017 Release 3.3.0 is significant release adding an Osd implementation for Apples Metal API New Features Added an Osd implementation for Apples Metal API Added the mtlViewer example Changes Fixed several instances of local variable shadowing that could cause build warnings Updated continuous-integration build scripts and added testing on macOS Release 3.2 Release 3.2.0 - Feb 2017 Release 3.2.0 is a minor release containing API additions and bug fixes New Features Extended FarStencilTableFactory to support face-varying Extended Osd Evaluator classes to support evaluation of 1st and 2nd derivatives Added an option to disable generation of legacy sharp corner patches Changes Corrected numerous spelling errors in doxygen comments Updated glFVarViewer with improved error detection and command line parsing Added option to build using MSVC with static CRT Bug Fixes Fixed a double delete of GL program in OsdGLComputeEvaluator Release 3.1 Release 3.1.1 - Jan 2017 Release 3.1.1 is a minor bug-fix release. Bug Fixes Fixed a bug with non-manifold face-varying topology causing a crash during patch table creation Fixed GLEW compilation and linking with dynamic GLEW libraries on Windows Fixed GLFW linking with GLFW 3.2 on X11 platforms Release 3.1.0 - Oct 2016 Release 3.1.0 is a significant release with several new features, bug fixes, and general code and configuration improvements. For more information on the following, please see Release 3.1 New Features Bicubic Face-Varying Patches Varying and Face-Varying Evaluation Second Order Derivative Evaluation Separate Levels of Feature Isolation Sharp Patches for Infinitely Sharp Features Changes Enabled the use of CMakes folder feature Removed the use of iso646 alternative keywords and, or, not, etc. to improve portability Added numerical valued preprocessor directives OPENSUBDIV_VERSION_MAJOR, etc. to opensubdivversion.h Improved documentation for FarPatchParam and added Unnormalize to complement Normalize Added additional topology queries to FarTopologyLevel Updated glFVarViewer and glEvalLimit viewer to make use of bicubic face-varying patches Updated glViewer and dxViewer to add a toggle for InfSharpPatch Updated dxPtexViewer for improved feature parity with glPtexViewer Improved far_regression to exercise shapes independent of Hbr compatibility Added support for Appveyor continuous integration testing Removed cmakeFindIlmBase Removed mayaPolySmooth example Bug Fixes Fixed Ptex version parsing and compatibility issues Fixed compatibility issues with VS2015 Fixed bug interpolating face-varying data with Bilinear scheme Fixed bug with refinement using Chaikin creasing Fixed bugs with HUD sliders in the example viewers Release 3.0 Release 3.0.5 - Mar 2016 Release 3.0.5 is a minor stability release with performance and correctness bug fixes. Bug Fixes The previous release reduced transient memory use during PatchTable construction, but increased the amount of memory consumed by the resulting PatchTable itself, this regression has been fixed. The example Ptex texture sampling code has been fixed to prevent sampling beyond the texels for a face when multisample rasterization is enabled. Release 3.0.4 - Feb 2016 Release 3.0.4 is a minor stability release which includes important performance and bug fixes. New Features Added accessor methods to FarLimitStencilTable to retrieve limit stencil data including derivative weights Added support for OpenCL event control to OsdCLVertexBuffer and OsdCLEvaluator Changes Major reduction in memory use during FarPatchTable construction for topologies with large numbers of extraordinary features Improved performance for GL and D3D11 tessellation control hull shader execution when drawing BSpline patches with the single crease patch optimization enabled Bug Fixes Restored support for drawing with fractional tessellation Fixed far_tutorial_6 to refine primvar data only up to the number of levels produced by topological refinement Fixed build warnings and errors reported by Visual Studio 2015 Release 3.0.3 - Oct 2015 Release 3.0.3 is a minor stability release which includes important performance and bug fixes. New Features Smooth normal generation tutorial, far_tutorial_8 Changes Major performance improvement in PatchTable construction Improved patch approximations for non-manifold features Bug Fixes Fixed double delete in GLSL Compute controller Fixed buffer layout for GLSL Compute kernel Fixed GL buffer leak in OsdGLPatchTable Fixed out-of-bounds data access for TBB and OMP stencil evaluation Fixed WIN32_LEAN_AND_MEAN typo Fixed Loop-related shader issues glFVarViewer Release 3.0.2 - Aug 2015 Release 3.0.2 is a minor release for a specific fix. Bug Fixes Fixed drawing of single crease patches Release 3.0.1 - Aug 2015 Release 3.0.1 is a minor release focused on stability and correctness. Changes Added a references section to the documentation, please see References Removed references to AddVaryingWithWeight from examples and tutorials Added more regression test shapes Addressed general compiler warnings e.g. signed vs unsigned comparisons Addressed compiler warnings in the core libraries reported by GCCs -Wshadow Eased GCC version restriction, earlier requirement for version 4.8 or newer is no longer needed Replaced topology initialization assertions with errors Improved compatibility with ICC Improved descriptive content and formatting of Far error messages Improved build when configured to include no GPU specific code Bug Fixes Fixed handling of unconnected vertices to avoid out of bounds data access Fixed non-zero starting offsets for TbbEvalStencils and OmpEvalStencils Fixed FarStencilTableFactoryOptionsfactorizeIntermediateLevels Fixed FarPatchTablesFactoryOptionsgenerateAllLevels Fixed the behavior of VTX_BOUNDARY_NONE for meshes with bilinear scheme Fixed some template method specializations which produced duplicate definitions Disabled depth buffering when drawing the UI in the example viewers Disabled the fractional tessellation spacing option in example viewers since this mode is currently not supported Release 3.0.0 - Jun 2015 Release 3.0.0 is a major release with many significant improvements and changes. For more information on the following, please see Release 3.0 New Features Faster subdivision using less memory Support for non-manifold topology Face-Varying data specified topologically Elimination of fixed valence tables Single-crease patch for semi-sharp edges Additional irregular patch approximations Introduction of Stencil Tables Faster, simpler GPU kernels Unified adaptive shaders Updated coding style with namespaces More documentation and tutorials Bug Fixes Smooth Face-Varying interpolation around creases Release 3.0.0 RC2 New Features Documentation updates far_tutorial_3 updates for the multiple face-varying channels maya example plugin interpolates a UV channel and a vertex color channel Bug Fixes Fixed a LimitStencilTableFactory bug, which returns an invalid table PatchParam encoding changed to support refinement levels up to 10 Added Xinerama link dependency Fixed MSVC 32bit build problem Fixed minor cmake issues Fixed glViewerfarViewer stability bugs Release 3.0.0 RC1 Changes FarTopologyRefiner was split into several classes to clarify and focus the API. Interpolation of Vertex and Varying primvars in a single pass is no longer supported. The Osd layer was largely refactored. Previous 2.x Release Notes Previous releases Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"release_notes.html\"\n }, {\n- \"title\": \"Subdivision Surfaces\",\n- \"text\": \"Subdivision Surfaces Subdivision Surfaces 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Overview Piecewise Parametric Surfaces Parametric Patches Piecewise Surfaces Arbitrary Topology Regular versus Irregular Features Non-manifold Topology Subdivision versus Tessellation Subdivision Tessellation Which to Use Mesh Data and Topology Separating Data from Topology Vertex and Varying Data Face-Varying Data and Topology Schemes and Options Subdivision Schemes Boundary Interpolation Rules Face-varying Interpolation Rules Semi-Sharp Creases Other Options Chaikin Rule Triangle Subdivision Rule Overview Subdivision surfaces are a common modeling primitive that has gained popularity in animation and visual effects over the past decades. As the name suggests, subdivision surfaces are fundamentally surfaces . More specifically, subdivision surfaces are piecewise parametric surfaces defined over meshes of arbitrary topology -- both concepts that will be described in the sections that follow. Subdivision is both an operation that can be applied to a polygonal mesh to refine it, and a mathematical tool that defines the underlying smooth surface to which repeated subdivision of the mesh converges. Explicit subdivision is simple to apply some number of times to provide a smoother mesh, and that simplicity has historically lead to many tools representing the shape this way. In contrast, deriving the smooth surface that ultimately defines the shape -- its limit surface -- is considerably more complex but provides greater accuracy and flexibility. These differences have led to confusion in how some tools expose subdivision surfaces. The ultimate goal is to have all tools use subdivision surfaces as true surface primitives. The focus here is therefore less on subdivision and more on the nature of the surface that results from it. In addition to providing a consistent implementation of subdivision -- one that includes a number of widely used feature extensions -- a significant value of OpenSubdiv is that it makes the limit surface more accessible. Since its introduction, OpenSubdiv has received interest from users and developers with a wide variety of skills, interests and backgrounds. This document is intended to present subdivision surfaces from a perspective helpful in making use of OpenSubdiv. One purpose it serves is to provide a high level overview for those with less experience with the algorithms or mathematics of subdivision. The other is to provide an overview of the feature set available with OpenSubdiv, and to introduce those capabilities with the terminology used by OpenSubdiv as much of it is overloaded. Piecewise Parametric Surfaces Piecewise parametric surfaces are arguably the most widely used geometric representation in industrial design, entertainment and many other areas. Many of the objects we deal with everyday -- cars, mobile phones, laptops -- were all designed and visualized first as piecewise parametric surfaces before those designs were approved and pursued. Piecewise parametric surfaces are ultimately just collections of simpler modeling primitives referred to as patches. Patches constitute the pieces of the larger surface in much the same way as a face or polygon constitutes a piece of a polygonal mesh. Parametric Patches Patches are the building blocks of piecewise smooth surfaces, and many different kinds of patches have evolved to meet the needs of geometric modeling. Two of the more effective and common patches are illustrated below Single bicubic B-Spline patch Single bicubic Bezier patch Patches consist of a set of points or vertices that affect a rectangular piece of smooth surface triangular patches also exist. That rectangle is parameterized in its two directions, transforming a simple 2D rectangle into the 3D surface u,v 2D domain of a patch Mapping from u,v to x,y,z The points that control the shape of the surface are usually referred to as control points or control vertices, and the collection of the entire set defining a patch as the control mesh, the control hull, the control cage or simply the hull, the cage, etc. For the sake of brevity we will frequently use the term cage, which serves us more generally later. So a patch essentially consist of two entities its control points and the surface affected by them. The way the control points affect the surface is what makes the different types of patches unique. Even patches defined by the same number of points can have different behavior. Note that all 16 points of the B-Spline patch above are relatively far from the surface they define compared to the similar Bezier patch. The two patches in that example actually represent exactly the same piece of surface -- each with a set of control points having different effects on it. In mathematical terms, each control point has a basis function associated with it that affects the surface in a particular way when only that point is moved Bicubic B-Spline basis function Bicubic Bezier basis funciton It is these basis functions that often give rise to the names of the different patches. There are pros and cons to these different properties of the control points of patches, which become more apparent as we assemble patches into piecewise surfaces. Piecewise Surfaces Piecewise parametric surfaces are collections of patches. For rectangular patches, one of the simplest ways to construct a collection is to define a set of patches using a rectangular grid of control points Piecewise B-Spline surface Piecewise Bezier surface Note that we can overlap the points of adjacent B-spline patches. This overlapping means that moving one control point affects multiple patches -- but it also ensures that those patches always meet smoothly this was a design intention and not true for other patch types. Adjacent Bezier patches only share points at their boundaries and coordinating the points across those boundaries to keep the surface smooth is possible, but awkward. This makes B-splines a more favorable surface representation for interactive modeling, but Bezier patches serve many other useful purposes. A more complicated B-spline surface Part of a more complicated B-Spline surface Just as a patch consisted of a cage and a surface, the same is now true of the collection. The control cage is manipulated by a designer and the surface of each of the patches involved is displayed so they can assess its effect. Arbitrary Topology Piecewise surfaces discussed thus far have been restricted to collections of patches over regular grids of control points. There is a certain simplicity with rectangular parametric surfaces that is appealing, but a surface representation that supports arbitrary topology has many other advantages. Rectangular parametric surfaces gained widespread adoption despite their topological limitations, and their popularity continues today in some areas. Complex objects often need many such surfaces to represent them and a variety of techniques have evolved to assemble them effectively, including stitching multiple surfaces together or cutting holes into them trimming. These are complicated techniques, and while effective in some contexts e.g. industrial design they become cumbersome in others e.g. animation and visual effects. A single polygonal mesh can represent shapes with far more complexity than a single rectangular piecewise surface, but its faceted nature eventually becomes a problem. Subdivision surfaces combine the topological flexibility of polygonal meshes with the underlying smoothness of piecewise parametric surfaces. Just as rectangular piecewise parametric surfaces have a collection of control points its cage stored as a grid and an underlying surface, subdivision surfaces also have a collection of control points its cage stored as a mesh and an underlying surface often referred as its limit surface. Regular versus Irregular Features A mesh contains the vertices and faces that form the cage for the underlying surface, and the topology of that mesh can be arbitrarily complex. In areas where the faces and vertices of the mesh are connected to form rectangular grids, the limit surface becomes one of the rectangular piecewise parametric surfaces previously mentioned. These regions of the mesh are said to be regular they provide behavior familiar from the use of similar rectangular surfaces and their limit surface is relatively simple to deal with. All other areas are considered irregular they provide the desired topological flexibility and so are less familiar and less predictable in some cases and their limit surface can be much more complicated. Irregular features come in a number of forms. The most widely referred to is an extra-ordinary vertex, i.e. a vertex which, in the case of a quad subdivision scheme like Catmull-Clark, does not have four incident faces. Irregular vertex and incident faces Regular and irregular regions of the surface The presence of these irregular features makes the limit surface around them similarly irregular, i.e. it cannot be represented as simply as it can for regular regions. Its worth noting that irregular regions shrink in size and become more isolated as subdivision is applied. A face with a lot of extra-ordinary vertices around it makes for a very complicated surface, and isolating these features is a way to help deal with that complexity Two valence-5 vertices nearby Isolation subdivided once Isolation subdivided twice Its generally necessary to perform some kind of local subdivision in these areas to break these pieces of surface into smaller, more manageable pieces, and the term feature adaptive subdivision has become popular in recent years to describe this process. Whether this is done explicitly or implicitly, globally or locally, what matters most is that there is an underlying piece of limit surface for each face -- albeit a potentially complicated one at an irregular feature -- that can be evaluated in much the same way as rectangular piecewise surfaces. Patches of the regular regions Patches of the irregular region While supporting a smooth surface in these irregular areas is the main advantage of subdivision surfaces, both the complexity of the resulting surfaces and their quality are reasons to use them with care. When the topology is largely irregular, there is a higher cost associated with its surface, so minimizing irregularities is advantageous. And in some cases the surface quality, i.e. the perceived smoothness, of the irregular surfaces can lead to undesirable artefacts. An arbitrary polygonal mesh will often not make a good subdivision cage, regardless of how good that polygonal mesh appears. As with rectangular piecewise parametric surfaces, the cage should be shaped to affect the underlying surface it is intended to represent. See Modeling Tips for related recommendations. Non-manifold Topology Since the cage of a subdivision surface is stored in a mesh, and often manipulated in the same context as polygonal meshes, the topic of manifold versus non-manifold topology warrants some attention. There are many definitions or descriptions of what distinguishes a manifold mesh from one that is not. These range from concise but abstract mathematical definitions to sets of examples showing manifold and non-manifold meshes -- all have their value and an appropriate audience. The following is not a strict definition but serves well to illustrate most local topological configurations that cause a mesh to be non-manifold. Consider standing on the faces of a mesh and walking around each vertex in turn. Assuming a right-hand winding order of faces, stand on the side of the face in the positive normal direction. And when walking, step across each incident edge in a counter-clockwise direction to the next incident face. For an interior vertex start at the corner of any incident face walk around the vertex across each incident edge to the next unvisited face repeat if you arrive back where you started and any incident faces or edges were not visited, the mesh is non-manifold Similarly, for a boundary vertex start at the corner of the face containing the leading boundary edge walk around the vertex across each incident edge to the next unvisited face repeat if you arrive at another boundary edge and any incident faces or edges were not visited, the mesh is non-manifold If you can walk around all vertices this way and dont encounter any non-manifold features, the mesh is likely manifold. Obviously if a vertex has no faces, there is nothing to walk around and this test cant succeed, so it is again non-manifold. All of the faces around a vertex should also be in the same orientation, otherwise two adjacent faces have normals in opposite directions and the mesh will be considered non-manifold, so we should really include that constraint when stepping to the next face to be more strict. Consider walking around the indicated vertices of the following non-manifold meshes Edges with 2 incident faces Faces sharing a vertex but no edges As mentioned earlier, many tools do not support non-manifold meshes, and in some contexts, e.g. 3D printing, they should be strictly avoided. Sometimes a manifold mesh may be desired and enforced as an end result, but the mesh may temporarily become non-manifold due to a particular sequence of modeling operations. Rather than supporting or advocating the use of non-manifold meshes, OpenSubdiv strives to be robust in the presence of non-manifold features to simplify the usage of its clients -- sparing them the need for topological analysis to determine when OpenSubdiv can or cannot be used. Although subdivision rules are not as well standardized in areas where the mesh is not manifold, OpenSubdiv provides simple rules and a reasonable limit surface in most cases. Surface around edges with 2 incident faces Surface for faces sharing a vertex but no edges As with the case of regular versus irregular features, since every face has a corresponding piece of surface associated with it -- whether locally manifold or not -- the term arbitrary topology can be said to include non-manifold topology. Subdivision versus Tessellation The preceding sections illustrate subdivision surfaces as piecewise parametric surfaces of arbitrary topology. As piecewise parametric surfaces, they consist of a cage and the underlying surface defined by that cage. Two techniques used to display subdivision surfaces are subdivision and tessellation. Both have their legitimate uses, but there is an important distinction between them subdivision operates on a cage and produces a refined cage tessellation operates on a surface and produces a discretization of that surface The existence and relative simplicity of the subdivision algorithm makes it easy to apply repeatedly to approximate the shape of the surface, but with the result being a refined cage, that approximation is not always very accurate. When compared to a cage refined to a different level, or a tessellation that uses points evaluated directly on the limit surface, the discrepancies can be confusing. Subdivision Subdivision is the process that gives subdivision surfaces their name, but it is not unique to them. Being piecewise parametric surfaces, lets first look at subdivision in the context of the simpler parametric patches that comprise them. Subdivision is a special case of refinement , which is key to the success of some of the most widely used types of parametric patches and their aggregate surfaces. A surface can be refined when an algorithm exists such that more control points can be introduced while keeping the shape of the surface exactly the same . For interactive and design purposes, this allows a designer to introduce more resolution for finer control without introducing undesired side effects in the shape. For more analytical purposes, it allows the surface to be broken into pieces, often adaptively, while being faithful to the original shape. One reason why both B-spline and Bezier patches are so widely used is that both of them can be refined. Uniform subdivision -- the process of splitting each of the patches in one or both of its directions -- is a special case of refinement that both of these patch types support B-Spline surface and its cage Cage subdivided 1x Cage subdivided 2x In the cases illustrated above for B-Splines, the uniformly refined cages produce the same limit surface as the original granted in more pieces. So it is fair to say that both uniform B-splines and Bezier surfaces are subdivision surfaces. The limit surface remains the same with the many more control points roughly 4x with each iteration of subdivision, and those points are closer to but not on the surface. It may be tempting to use these new control points to represent the surface, but using the same number of points evaluated at corresponding uniformly spaced parametric locations on the surface is usually simpler and more effective. Note also that points of the cage typically do not have any normal vectors associated with them, though we can evaluate normals explicitly for arbitrary locations on the surface just as we do for position. So if displaying a cage as a shaded surface, normal vectors at each of the control points must be contrived. Both the positions and normals of the points on the finer cage are therefore both approximations. For more general subdivision surfaces, the same is true. Subdivision will refine a mesh of arbitrary topology, but the resulting points will not lie on the limit surface and any normal vectors contrived from and associated with these points will only be approximations to those of the limit surface. Tessellation There is little need to use subdivision to approximate a parametric surface when it can be computed directly, i.e. it can be tessellated. We can evaluate at arbitrary locations on the surface and connect the resulting points to form a tessellation -- a discretization of the limit surface -- that is far more flexible than the results achieved from uniform subdivision Uniform 3x3 tessellation of B-spline surface Curvature-adaptive tessellation of B-spline surface For a simple parametric surface, the direct evaluation of the limit surface is also simple, but for more complicated subdivision surfaces of arbitrary topology, this is less the case. The lack of a clear understanding of the relationship between the limit surface and the cage has historically lead to many applications avoiding tessellation. Its worth mentioning that subdivision can be used to generate a tessellation even when the limit surface is not available for direct evaluation. The recursive nature of subdivision does give rise to formulae that allow a point on the limit surface to be computed that corresponds to each point of the cage. This process is often referred to as snapping or pushing the points of the cage onto the limit surface. Subdivided 1x and snapped to limit surface Subdivided 2x and snapped to limit surface Since the end result is a connected set of points on the limit surface, this forms a tessellation of the limit surface, and we consider it a separate process to subdivision though it does make use of it. The fact that such a tessellation might have been achieved using subdivision is indistinguishable from the final result -- the same tessellation might just as easily have been generated by evaluating limit patches of the cage uniformly 2x, 4x, 8x, etc. along each edge. Which to Use Subdivision is undeniably useful in creating finer cages to manipulate the surface, but tessellation is preferred for displaying the surface when the patches are available for direct evaluation. There was a time when global refinement was pursued in limited circles as a way of rapidly evaluating parametric surfaces along isoparametric lines, but patch evaluation, i.e. tessellation, generally prevails. Considerable confusion has arisen due the way the two techniques have been employed and presented when displaying the shape in end-user applications. One can argue that if an application displays a representation of the surface that is satisfactory for its purposes, then it is not necessary to burden the user with additional terminology and choices. But when two representations of the same surface differ considerably between two applications, the lack of any explanation or control leads to confusion. As long as applications make different choices on how to display the surface, we seek a balance between simplicity and control. Since subdivided points do not lie on the limit surface, it is important to make it clear to users when subdivision is being used instead of tessellation. This is particularly true in applications where the cage and the surface are displayed in the same style as there is no visual cue for users to make that distinction. Mesh Data and Topology The ability of subdivision surfaces to support arbitrary topology leads to the use of meshes to store both the topology of the cage and the data values associated with its control points, i.e. its vertices. The shape of a mesh, or the subdivision surface that results from it, is a combination of the topology of the mesh and the position data associated with its vertices. When dealing with meshes there are advantages to separating the topology from the data, and this is even more important when dealing with subdivision surfaces. The shape referred to above is not just the shape of the mesh the cage in this case but could be the shape of a refined cage or the limit surface. By observing the roles that both the data and topology play in operations such as subdivision and evaluation, significant advantages can be gained by managing data, topology and the associated computations accordingly. While the main purpose of subdivision surfaces is to use position data associated with the vertices to define a smooth, continuous limit surface, there are many cases where non-positional data is associated with a mesh. That data may often be interpolated smoothly like position, but often it is preferred to interpolate it linearly or even make it discontinuous along edges of the mesh. Texture coordinates and color are common examples here. Other than position, which is assigned to and associated with vertices, there are no constraints on how arbitrary data can or should be associated or interpolated. Texture coordinates, for example, can be assigned to create a completely smooth limit surface like the position, linearly interpolated across faces, or even made discontinuous between them. There are, however, consequences to consider -- both in terms of data management and performance -- which are described below as the terminology and techniques used to achieve each are defined. Separating Data from Topology While the topology of meshes used to store subdivision surfaces is arbitrarily complex and variable, the topology of the parametric patches that make up its limit surface are simple and fixed. Bicubic B-Spline and Bezier patches are both defined by a simple 4x4 grid of control points and a set of basis functions for each point that collectively form the resulting surface. For such a patch, the position at a given parametric location is the result of the combination of position data associated with its control points and the weights of the corresponding basis functions weights being the values of basis functions evaluated at a parametric location. The topology and the basis functions remain the same, so we can make use of the weights independent of the data. If the positions of the control points change, we can simply recombine the new position data with the weights that we just used and apply the same combination. The fixed topology of a parametric patch and two shapes resulting from two sets of positions. Similarly, for a piecewise surface, the position at a given parametric location is the result of the single patch containing that parametric location evaluated at the given position. The control points involved are the subset of control points associated with that particular patch. If the topology of the surface is fixed, so too is the topology of the collection of patches that comprise that surface. If the positions of those control points change, we can recombine the new position data with the same weights for the subset of points associated with the patch. More complex but fixed topology of a surface and two shapes resulting from two sets of positions. This holds for a piecewise surface of arbitrary topology. Regardless of how complex the topology, as long as it remains fixed i.e. relationships between vertices, edges and faces does not change or anything other settings affecting subdivision rules, the same techniques apply. This is just one example of the value of separating computations involving topology from those involving the data. Both subdivision and evaluation can be factored into steps involving topology computing the weights and combining the data separately. Three shapes resulting from three sets of positions for a mesh of fixed topology. When the topology is fixed, enormous savings are possible by pre-computing information associated with the topology and organizing the data associated with the control points in a way that can be efficiently combined with it. This is key to understanding some of the techniques used to process subdivision surfaces. For a mesh of arbitrary topology, the control points of the underlying surface are the vertices, and position data associated with them is most familiar. But there is nothing that requires that the control points of a patch have to represent position -- the same techniques apply regardless of the type of data involved. Vertex and Varying Data The most typical and fundamental operation is to evaluate a position on the surface, i.e. evaluate the underlying patches of the limit surface using the x,y,z positions at the vertices of the mesh. Given a parametric u,v location on one such patch, the data-independent evaluation method first computes the weights and then combines the x,y,z vertex positions resulting in an x,y,z position at that location. But the weights and their combination can be applied to any data at the vertices, e.g. color, texture coordinates or anything else. Data associated with the vertices that is interpolated this way, including position, is said to be vertex data or to have vertex interpolation. Specifying other data as vertex data will result in it being smoothly interpolated in exactly the same way using exactly the same weights as the position. So to capture a simple 2D projection of the surface for texture coordinates, 2D values matching the x,y of the positions would be used. If linear interpolation of data associated with vertices is desired instead, the data is said to be varying data or to have varying interpolation. Here the non-linear evaluation of the patches defining the smooth limit surface is ignored and weights for simple linear interpolation are used. This is a common choice for texture coordinates as evaluation of texture without the need of bicubic patches is computationally cheaper. The linear interpolation will not capture the smoothness required of a true projection between the vertices, but both vertex and varying interpolation have their uses. Projected texture smoothly interpolated from vertex data Projected texture linearly interpolated from varying data Since both vertex and varying data is associated with vertices a unique value assigned to each, the resulting surface will be continuous -- piecewise smooth in the case of vertex data and piecewise linear in the case of varying. Face-Varying Data and Topology In order to support discontinuities in data on the surface, unlike vertex and varying data, there must be multiple values associated with vertices, edges andor faces, in order for a discontinuity to exist. Discontinuities are made possible by assigning values to the corners of faces, similar to the way in which vertices are assigned to the corners of faces when defining the topology of the mesh. Recalling the assignment of vertices to faces Vertex indices are assigned to all corners of each face as part of mesh construction and are often referred to as the face-vertices of an individual face or the mesh. All face-vertices that share the same vertex index will be connected by that vertex and share the same vertex data associated with it. By assigning a different set of indices to the face-vertices -- indices not referring to the vertices but some set of data to be associated with the corners of each face -- corners that share the same vertex no longer need to share the same data value and the data can be made discontinuous between faces This method of associating data values with the face-vertices of the mesh is said to be assigning face-varying data for face-varying interpolation. An interpolated value will vary continuously within a face i.e. the patch of the limit surface associated with the face but not necessarily across the edges or vertices shared with adjacent faces. Disjoint face-varying UV regions applied to the limit surface The combination of associating data values not with the vertices the control points but the face corners, and the resulting data-dependent discontinuities that result, make this a considerably more complicated approach than vertex or varying. The added complexity of the data alone is reason to only use it when necessary, i.e. when discontinuities are desired and present. Part of the complexity of dealing with face-varying data and interpolation is the way in which the interpolation behavior can be defined. Where the data is continuous, the interpolation can be specified to be as smooth as the underlying limit surface of vertex data or simply linear as achieved with varying data. Where the data is discontinuous -- across interior edges and around vertices -- the discontinuities create boundaries for the data, and partition the underlying surface into disjoint regions. The interpolation along these boundaries can also be specified as smooth or linear in a number of ways many of which have a historical basis. A more complete description of the different linear interpolation options with face-varying data and interpolation is given later. These options make it possible to treat the data as either vertex or varying, but with the added presence of discontinuities. An essential point to remember with face-varying interpolation is that each set of data is free to have its own discontinuities -- this leads to each data set having both unique topology and size. The topology specified for a collection of face-varying data is referred to as a channel and is unique to face-varying interpolation. Unlike vertex and varying interpolation, which both associate a data value with a vertex, the number of values in a face-varying channel is not fixed by the number of vertices or faces. The number of indices assigned to the face-corners will be the same for all channels, but the number of unique values referred to by these indices may not. We can take advantage of the common mesh topology in areas where the data is continuous, but we lose some of those advantages around the discontinuities. This results in the higher complexity and cost of a face-varying channel compared to vertex or varying data. If the topology for a channel is fixed, though, similar techniques can be applied to factor computation related to the topology so that changes to the data can be processed efficiently. Schemes and Options While previous sections have described subdivision surfaces in more general terms, this section describes a number of common variations often referred to as extensions to the subdivision algorithms and the ways that they are represented in OpenSubdiv. The number and nature of the extensions here significantly complicate what are otherwise fairly simple subdivision algorithms. Historically applications have supported either a subset or have had varying implementations of the same feature. OpenSubdiv strives to provide a consistent and efficient implementation of this feature set. Given the varying presentations of some of these features elsewhere, the naming chosen by OpenSubdiv is emphasized here. Subdivision Schemes OpenSubdiv provides two well known subdivision surface types -- Catmull-Clark often referred to more tersely as Catmark and Loop subdivision. Catmull-Clark is more widely used and suited to quad-dominant meshes, while Loop is preferred for and requires purely triangulated meshes. The many examples from previous sections have illustrated the more popular Catmull-Clark scheme. For an example of Loop Boundary Interpolation Rules Boundary interpolation rules control how subdivision and the limit surface behave for faces adjacent to boundary edges and vertices. The following choices are available via the enumeration SdcOptionsVtxBoundaryInterpolation Mode Behavior VTX_BOUNDARY_NONE No boundary edge interpolation is applied by default boundary faces are tagged as holes so that the boundary vertices continue to support the adjacent interior faces, but no surface corresponding to the boundary faces is generated boundary faces can be selectively interpolated by sharpening all boundary edges incident the vertices of the face VTX_BOUNDARY_EDGE_ONLY A sequence of boundary vertices defines a smooth curve to which the limit surface along boundary faces extends VTX_BOUNDARY_EDGE_AND_CORNER Similar to edge-only but the smooth curve resulting on the boundary is made to interpolate corner vertices vertices with exactly one incident face On a grid example In practice, it is rare to use no boundary interpolation at all -- this feature has its uses in allowing separate meshes to be seamlessly joined together by replicating the vertices along boundaries, but these uses are limited. Given the global nature of the setting, it is usually preferable to explicitly make the boundary faces holes in the areas where surfaces from separate meshes are joined, rather than sharpening edges to interpolate the desired boundaries everywhere else. The remaining edge only and edge and corner choices are then solely distinguished by whether or not the surface at corner vertices is smooth or sharp. Face-varying Interpolation Rules Face-varying interpolation rules control how face-varying data is interpolated both in the interior of face-varying regions smooth or linear and at the boundaries where it is discontinuous constrained to be linear or pinned in a number of ways. Where the topology is continuous and the interpolation chosen to be smooth, the behavior of face-varying interpolation will match that of the vertex interpolation. Choices for face-varying interpolation are most commonly available in the context of UVs for texture coordinates and a number of names for such choices have evolved in different applications over the years. The choices offered by OpenSubdiv cover a wide range of popular applications. The feature is named face-varying linear interpolation -- rather than boundary interpolation commonly used -- to emphasize that it can be applied to the entire surface not just boundaries and that the effects are to make the surface behave more linearly in various ways. The following choices are available for the SdcOptionsFVarLinearInterpolation enum -- the ordering here applying progressively more linear constraints Mode Behavior FVAR_LINEAR_NONE smooth everywhere the mesh is smooth FVAR_LINEAR_CORNERS_ONLY linearly interpolate sharpen or pin corners only FVAR_LINEAR_CORNERS_PLUS1 CORNERS_ONLY sharpening of junctions of 3 or more regions FVAR_LINEAR_CORNERS_PLUS2 CORNERS_PLUS1 sharpening of darts and concave corners FVAR_LINEAR_BOUNDARIES linear interpolation along all boundary edges and corners FVAR_LINEAR_ALL linear interpolation everywhere boundaries and interior These rules cannot make the interpolation of the face-varying data smoother than that of the vertices. The presence of sharp features of the mesh created by sharpness values, boundary interpolation rules, or the subdivision scheme itself e.g. Bilinear take precedence. All face-varying interpolation modes illustrated in UV space using a simple 4x4 grid of quads segmented into three UV regions their control point locations implied by interpolation in the FVAR_LINEAR_ALL case For those familiar, this shape and its assigned UV sets are available for inspection in the catmark_fvar_bound1 shape of OpenSubdivs example and regression shapes. Semi-Sharp Creases Just as some types of parametric surfaces support additional shaping controls to affect creasing along the boundaries between surface elements, OpenSubdiv provides additional sharpness values or weights associated with edges and vertices to achieve similar results over arbitrary topology. Setting sharpness values to a maximum value 10 in this case -- a number chosen for historical reasons effectively modifies the subdivision rules so that the boundaries between the piecewise smooth surfaces are infinitely sharp or discontinuous. But since real world surfaces never really have infinitely sharp edges, especially when viewed sufficiently close, it is often preferable to set the sharpness lower than this value, making the crease semi-sharp. A constant weight value assigned to a sequence of edges connected edges therefore enables the creation of features akin to fillets and blends without adding extra rows of vertices though that technique still has its merits Sharpness values range from 0-10, with a value of 0 or less having no effect on the surface and a value of 10 or more making the feature completely sharp. It should be noted that infinitely sharp creases are really tangent discontinuities in the surface, implying that the geometric normals are also discontinuous there. Therefore, displacing along the normal will likely tear apart the surface along the crease. If you really want to displace a surface at a crease, it may be better to make the crease semi-sharp. Other Options While the preceding options represent features available in a wide-variety of tools and modeling formats, a few others exist whose recognition and adoption is more limited. In some cases, they offer improvements to undesirable behavior of the subdivision algorithms, but their effects are less than ideal. Given both their limited effectiveness and lack of recognition, these options should be used with caution. Chaikin Rule The Chaikin Rule is a variation of the semi-sharp creasing method that attempts to improve the appearance of creases along a sequence of connected edges when the sharpness values differ. This choice modifies the subdivision of sharpness values using Chaikins curve subdivision algorithm to consider all sharpness values of edges around a common vertex when determining the sharpness of child edges. The creasing method can be set using the values defined in the enumeration SdcOptionsCreasingMethod Mode Behavior CREASE_UNIFORM Apply regular semi-sharp crease rules CREASE_CHAIKIN Apply Chaikin semi-sharp crease rules Example of contiguous semi-sharp creases interpolation Triangle Subdivision Rule The triangle subdivision rule is a rule added to the Catmull-Clark scheme that modifies the behavior at triangular faces to improve the undesirable surface artefacts that often result in such areas. Mode Behavior TRI_SUB_CATMARK Default Catmark scheme weights TRI_SUB_SMOOTH Smooth triangle weights Cylinder example This rule was empirically determined to make triangles subdivide more smoothly. However, this rule breaks the nice property that two separate meshes can be joined seamlessly by overlapping their boundaries i.e. when there are triangles at either boundary, it is impossible to join the meshes seamlessly Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"Sdc Overview\",\n+ \"text\": \"Sdc Overview Sdc Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Subdivision Core Sdc Types, Traits and Options Creasing support Scheme-specific support The FACE, EDGE and VERTEX interfaces The MASK interface Subdivision Core Sdc Sdc is the lowest level layer in OpenSubdiv. Its intent is to separate the core subdivision details from any particular representation of a mesh it was previously bound to Hbr to facilitate the generation of consistent results with other mesh representations, both internal and external to OpenSubdiv. The functionality can be divided roughly into three sections types, traits and options for the supported subdivision schemes computations required to support semi-sharp creasing computations for mask weights of subdivided vertices for all schemes For most common usage, familiarity with only the first of these is necessary -- primarily the use of public types and constants for the choice of subdivision scheme and its associated options. The latter two provide the basis for a more comprehensive implementation of subdivision, which requires considerably more understanding and effort. Overall, the approach was to extract the functionality at the lowest level possible. In some cases, the implementation is not far from being simple global functions. The intent was to start at a low level and build any higher level functionality as needed. What exists now is functional for ongoing development and anticipated needs within OpenSubdiv for the near future. The intent of Sdc is to provide the building blocks for OpenSubdiv and its clients to efficiently process the specific set of supported subdivision schemes. It is not intended to be a general framework for defining customized subdivision schemes. Types, Traits and Options The most basic type is the enum SdcSchemeType that identifies the fixed set of subdivision schemes supported by OpenSubdiv Bilinear , Catmark and Loop . With this alone, we intend to avoid all dynamic casting issues related to the scheme by simply adding members to the associated subclasses for inspection. In addition to the type enum itself, a class defining a fixed set of traits associated with each scheme is provided. While these traits are available as static methods in the interface of a class supporting more functionality for each scheme to be described shortly, the SchemeTypeTraits provide queries of the traits for a variable of type SdcSchemeType -- enabling parameterization of code by the value of a trait without templates or virtual inheritance a simple internal table of traits is constructed and trivially indexed. The second contribution is the collection of all variations in one place that can be applied to the subdivision schemes, i.e. the boundary interpolation rules, creasing method, edge subdivision choices, etc. The fact that these are all declared in one place alone should help clients see the full set of variations that are possible. A simple Options struct a set of bitfields aggregates all of these variations into a single object the equivalent of an integer in this case that are passed around to other Sdc classes andor methods and are expected to be used at a higher level both within OpenSubdiv and externally. By aggregating the options and passing them around as a group, it allows us to extend the set easily in future without the need to rewire a lot of interfaces to accommodate the new choice. Clients can enable new choices at the highest level and be assured that they will propagate to the lowest level where they are relevant. Unlike other options structs used elsewhere to specify variations of a particular method, SdcOptions defines all options that affect the shape of the underlying limit surface of a subdivision mesh. Other operations at higher levels in the library may have options that approximate the shape and so create a slightly different appearance, but SdcOptions is a fundamental part of the definition of the true limit surface. Creasing support Since the computations involved in the support of semi-sharp creasing are independent of the subdivision scheme, the goal in Sdc was to encapsulate all related creasing functionality in a similarly independent manner. Computations involving sharpness values are also much less dependent on topology -- there are vertices and edges with sharpness values, but knowledge of faces or boundary edges is not required, -- so the complexity of topological neighborhoods required for more scheme-specific functionality is arguably not necessary here. Creasing computations have been provided as methods defined on a Crease class that is constructed with a set of Options. Its methods typically take sharpness values as inputs and compute a corresponding set of sharpness values as a result. For the Uniform creasing method previously known as Normal , the computations may be so trivial as to question whether such an interface is worth it, but for Chaikin or other schemes in the future that are non-trivial, the benefits should be clear. Functionality is divided between both uniform and non-uniform, so clients have some control over avoiding unnecessary overhead, e.g. non-uniform computations typically require neighboring sharpness values around a vertex, while uniform does not. Also included as part of the Crease class is the Rule enum -- this indicates if a vertex is Smooth, Crease, Dart or Corner referred to as the mask in Hbr and is a function of the sharpness values at and around a vertex. Knowing the Rule for a vertex can accelerate mask queries, and the Rule can often be inferred based on the origin of a vertex e.g. it originated from the middle of a face, was the child of a Smooth vertex, etc.. Methods are defined for the Crease class to subdivide edge and vertex sharpness values determine the Rule for a vertex based on incident sharpness values determine the transitional weight between two sets of sharpness values Being all low-level and working directly on sharpness values, it is a clients responsibility to coordinate the application of any hierarchical crease edits with their computations. Similarly, in keeping with this as a low-level interface, values are passed as primitive arrays. This follows the trend in OpenSubdiv of dealing with data of various kinds e.g. weights, component indices, now sharpness values, etc. in small contiguous sets of values. In most internal cases we can refer to a set of values or gather what will typically be a small number of values on the stack for temporary use. Scheme-specific support While the SchemeTypeTraits class provides traits for each subdivision scheme supported by OpenSubdiv i.e. Bilinear , Catmark and Loop , the Scheme class provides these more directly, Additionally, the Scheme class provides methods for computing the various sets of weights used to compute new vertices resulting from subdivision. The collection of weights used to compute a single vertex at a new subdivision level is typically referred to as a mask . The primary purpose of the Scheme class is to provide such masks in a manner both general and efficient. Each subdivision scheme has its own values for its masks, and each are provided as specializations of the template class SchemeSchemeType TYPE . The intent is to minimize the amount of code specific to each scheme. The computation of mask weights for subdivided vertices is the most significant contribution of Sdc. The use of semi-sharp creasing with each non-linear subdivision scheme complicates what are otherwise simple masks determined solely by the topology, and packaging that functionality to achieve both the generality and efficiency desired has been a challenge. Mask queries are defined in the Scheme class template, which has specializations for each of the supported subdivision schemes. Mask queries are defined in terms of interfaces for two template parameters the first defining the topological neighborhood of a vertex, and a second defining a container in which to gather the individual weights template typename FACE , typename MASK void ComputeFaceVertexMask FACE const faceNeighborhood , MASK faceVertexMask , ... const Each mask query is expected to call methods defined for the FACE , EDGE or VERTEX classes to obtain the information they require typically these methods are simple queries about the topology and associated sharpness values. Clients are free to use their own mesh representations to gather the requested information as quickly as possible, or to cache some subset as member variables for immediate inline retrieval. In general, the set of weights for a subdivided vertex is dependent on the following the topology around the parent component from which the vertex originates the type of subdivision Rule applicable to the parent component the type of subdivision Rule applicable to the new child vertex a transitional weight blending the effect between differing parent and child rules This seems fairly straight-forward, until we look at some of the dependencies involved the parent Rule requires the sharpness values at and around the parent component the child Rule requires the subdivided sharpness values at and around the new child vertex though it can sometimes be trivially inferred from the parent the transitional weight between differing rules requires all parent and child sharpness values Clearly the sharpness values are inspected multiple times and so it pays to have them available for retrieval. Computing them on an as-needed basis may be simple for uniform creasing, but a non-uniform creasing method requires traversing topological neighborhoods, and that in addition to the computation itself can be costly. The point here is that it is potentially unreasonable to expect to evaluate the mask weights completely independent of any other consideration. Expecting and encouraging the client to have subdivided sharpness values first, for use in more than one place, is therefore recommended. The complexity of the general case above is also unnecessary for most vertices. Any client using Sdc typically has more information about the nature of the vertex being subdivided and much of this can be avoided -- particularly for the smooth interior case that often dominates. More on that in the details of the Scheme classes. Given that most of the complexity has been moved into the template parameters for the mask queries, the Scheme class remains fairly simple. Like the Crease class, it is instantiated with a set of Options to avoid them cluttering the interface. It is currently little more than a few methods for the limit and refinement masks for each vertex type, plus the few fixed traits of the scheme as static methods. The mask queries have been written in a way that greatly simplifies the specializations required for each scheme. The generic implementation for both the edge-vertex and vertex-vertex masks take care of all of the creasing logic, requiring only a small set of specific masks to be assigned for each Scheme smooth and crease masks for an edge-vertex, and smooth, crease and corner masks for a vertex-vertex. Other than the Bilinear case, which will specialize the mask queries to trivialize them for linear interpolation, the specializations for each Scheme should only require defining this set of masks -- and with two of them common edge-vertex crease and vertex-vertex corner the Catmark scheme only needs to define three. The FACE, EDGE and VERTEX interfaces Mask queries require an interface to a topological neighborhood, currently labeled FACE , EDGE and VERTEX . This naming potentially implies more generality than intended, as such classes are only expected to provide the methods required of the mask queries to compute its associated weights. While all methods must be defined, some may rarely be invoked, and the client has considerable flexibility in the implementation of these they can defer some evaluations lazily until required, or be pro-active and cache information in member variables for immediate access. An approach discussed in the past has alluded to iterator classes that clients would write to traverse their meshes. The mask queries would then be parameterized in terms of a more general and generic mesh component that would make use of more general traversal iterators. The advantage here is the iterators are written once, then traversal is left to the query and only what is necessary is gathered. The disadvantages are that clients are forced to write these to do anything, getting them correct and efficient may not be trivial or possible in some cases, and that the same data e.g. subdivided sharpness may be gathered or computed multiple times for different purposes. The other extreme was to gather everything possible required at once, but that is objectionable. The approach taken here provides a reasonable compromise between the two. The mask queries ask for exactly what they want, and the provided classes are expected to deliver it as efficiently as possible. In some cases the client may already be storing it in a more accessible form and general topological iteration can be avoided. The information requested of these classes in the three mask queries is as follows For FACE the number of incident vertices For EDGE the number of incident faces the sharpness value of the parent edge the sharpness values of the two child edges the number of vertices per incident face For VERTEX the number of incident faces the number of incident edges the sharpness value of the parent vertex the sharpness values for each incident parent edge the sharpness value of the child vertex the sharpness values for each incident child edge The latter should not be surprising given the dependencies noted above. There are also a few more to consider for future use, e.g. whether the EDGE or VERTEX is manifold or not. In most cases, additional information can be provided to the mask queries i.e. pre-determined Rules, and most of the child sharpness values are not necessary. The most demanding situation is a fractional crease that decays to zero -- in which case all parent and child sharpness values in the neighborhood are required to determine the proper transitional weight. The MASK interface Methods dealing with the collections of weights defining a mask are typically parameterized by a MASK template parameter that contains the weights. The set of mask weights is currently divided into vertex-weights, edge-weights and face-weights -- consistent with previous usage in OpenSubdiv and providing some useful correlation between the full set of weights and topology. The vertex-weights refer to parent vertices incident the parent component from which a vertex originated, the edge-weights the vertices opposite incident edges of the parent, and the face-weights the center of incident parent faces. Note the latter is NOT in terms of vertices of the parent but potentially vertices in the child originating from faces of the parent. This has been done historically in OpenSubdiv but is finding less use -- particularly when it comes to providing greater support for the Loop scheme -- and is a point needing attention. So the mask queries require the following capabilities assign the number of vertex, edge andor face weights retrieve the number of vertex, edge andor face weights assign individual vertex, edge andor face weights by index retrieve individual vertex, edge andor face weights by index through a set of methods required of all MASK classes. Since the maximum number of weights is typically known based on the topology, usage within Vtr, Far or Hbr is expected to simply define buffers on the stack. Another option is to utilize pre-allocated tables, partitioned into the three sets of weights on construction of a MASK , and populated by the mask queries. A potentially useful side-effect of this is that the client can define their weights to be stored in either single or double-precision. With that possibility in mind, care was taken within the mask queries to make use of a declared type in the MASK interface MASKWeight for intermediate calculations. Having support for double-precision masks in Sdc does enable it at higher levels in OpenSubdiv if later desired, and that support is made almost trivial with MASK being generic. It is important to remember here that these masks are being defined consistent with existing usage within OpenSubdiv both Hbr and the subdivision tables generated by Far . As noted above, the face weights correspond to the centers of incident faces, i.e. vertices on the same level as the vertex for which the mask is being computed, and not relative to vertices in the parent level as with the other sets of weights. It is true that the weights can be translated into a set in terms solely of parent vertices, but in the general case i.e. Catmark subdivision with non-quads in the base mesh this requires additional topological association. In general we would need N-3 weights for the N-3 vertices between the two incident edges, where N is the number of vertices of each face typically 4 even at level 0. Perhaps such a translation method could be provided on the mask class, with an optional indication of the incident face topology for the irregular cases. The Loop scheme does not have face weights , for a vertex-vertex mask, but for an edge-vertex mask it does require weights associated with the faces incident the edge -- either the vertex opposite the edge for each triangle, or its center which has no other use for Loop. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"subdivision_surfaces.html\"\n+ \"loc\": \"sdc_overview.html\"\n }, {\n \"title\": \"Overview of Release 3.5\",\n- \"text\": \"Overview of Release 3.5 Overview of Release 3.5 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE New Features Simplified Surface Evaluation Tessellation Patterns API Additions New classes in namespace Bfr Additions to FarTopologyLevel Additions to nested Options classes in Far Other Changes Deprecation Announcements Improvements Bug Fixes New Features Simplified Surface Evaluation The new Bfr interface provides an alternative to Far and Osd for evaluating the limit surface -- treating a subdivision mesh as a piecewise parameteric surface primitive . The Bfr interface is named for Base Face Representation as all concepts and classes relate to the individual faces of the original unrefined mesh, i.e. its base faces. Each base face has a piece of limit surface associated with it BfrSurface that is parameterized BfrParameterization and so can be evaluated and tessellated The limit surface for each face can be identified and evaluated independently of any other faces without any global pre-processing of a connected mesh. The resulting evaluation interface is much simpler, more flexible and more scalable than those assembled with the table-based class in Far -- providing a preferable alternative for many CPU-based use cases though not all. See the Bfr Overview documentation for more details. Tessellation Patterns In addition to evaluation, the Bfr interface includes a Tessellation class that provides tessellation patterns for all faces of a base mesh. A Tessellation does not encapsulate the entire tessellation process but simply provides topological patterns to assemble such a process. All faces of a mesh have a Parameterization which can be used to define a Tessellation with the specification of one or more tessellation rates. A single tessellation rate defines simple uniform tessellations while tessellation rates for each edge provide more complex non-uniform patterns A small set of options is also available to define the resulting patterns e.g. the preservation of quad faces with quad-based subdivision schemes above left and center and to generate its output in a more favorable form. Tessellation patterns do not support the full flexibility of typical hardware tessellation e.g. no fractional tessellation but do provide some useful alternatives to hardware tessellation e.g. the quad preservation previously noted, and more uniform triangles for triangular patches above right . The Tessellation class is also completely independent of evaluation in Bfr , and so can be used with any other evaluation interface. See the Tessellation subsection of the Bfr Overview for more details. API Additions See associated Doxygen for full details. New classes in namespace Bfr class Parameterization class Surface class SurfaceFactory class SurfaceFactoryCache class SurfaceFactoryMeshAdapter class RefinerSurfaceFactory class Tessellation Additions to FarTopologyLevel TopologyLevelIsVertexCorner TopologyLevelIsVertexRegular TopologyLevelIsVertexInfSharp TopologyLevelIsVertexSemiSharp TopologyLevelIsEdgeInfSharp TopologyLevelIsEdgeSemiSharp Additions to nested Options classes in Far PatchTableFactoryOptionsSetMaxIsolationLevel TopologyRefinerUniformOptionsSetRefinementLevel TopologyRefinerAdaptiveOptionsSetMaxIsolationLevel TopologyRefinerAdaptiveOptionsSetMaxSecondaryLevel Other Changes Deprecation Announcements Hbr is deprecated and will be removed from subsequent releases Improvements Suppression of GCC compiler warnings GitHub 1253, 1254, 1270 Additional methods for FarTopologyLevel GitHub 1227, 1255 Improved mixed partial derivative at Gregory patch corners GitHub 1252 Minor improvements to Far tutorials GitHub 1226, 1241 Added CMake config GitHub 1242 Updated CMake minimum version to 3.12 GitHub 1237, 1261 Updated documentation build scripts for Python 3 1265, 1266 Updated stringify build tool for improved cross compilation support GitHub 1267 Added NO_MACOS_FRAMEWORKS build option GitHub 1238 Updated Azure pipelines agents for Unbuntu and macOS GitHub 1247, 1256 Removed obsolete AppVeyor and Travis CI scripts GitHub 1259 Bug Fixes Cache active program for OsdGLComputeEvaluator GitHub 1244 Fixed member initialization warnings in OsdD3D11ComputeEvaluator GitHub 1239 Fixed GLSL shader source to remove storage qualifiers from struct members GitHub 1271 Fixed use of CMake variables for Apple builds GitHub 1235 Fixed build errors when using OpenGL without GLFW GitHub 1257 Fixed links to embedded videos GitHub 1231 Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"Overview of Release 3.5 Overview of Release 3.5 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE New Features Simplified Surface Evaluation Tessellation Patterns API Additions New classes in namespace Bfr Additions to FarTopologyLevel Additions to nested Options classes in Far Other Changes Deprecation Announcements Improvements Bug Fixes New Features Simplified Surface Evaluation The new Bfr interface provides an alternative to Far and Osd for evaluating the limit surface -- treating a subdivision mesh as a piecewise parameteric surface primitive . The Bfr interface is named for Base Face Representation as all concepts and classes relate to the individual faces of the original unrefined mesh, i.e. its base faces. Each base face has a piece of limit surface associated with it BfrSurface that is parameterized BfrParameterization and so can be evaluated and tessellated The limit surface for each face can be identified and evaluated independently of any other faces without any global pre-processing of a connected mesh. The resulting evaluation interface is much simpler, more flexible and more scalable than those assembled with the table-based class in Far -- providing a preferable alternative for many CPU-based use cases though not all. See the Bfr Overview documentation for more details. Tessellation Patterns In addition to evaluation, the Bfr interface includes a Tessellation class that provides tessellation patterns for all faces of a base mesh. A Tessellation does not encapsulate the entire tessellation process but simply provides topological patterns to assemble such a process. All faces of a mesh have a Parameterization which can be used to define a Tessellation with the specification of one or more tessellation rates. A single tessellation rate defines simple uniform tessellations while tessellation rates for each edge provide more complex non-uniform patterns A small set of options is also available to define the resulting patterns e.g. the preservation of quad faces with quad-based subdivision schemes above left and center and to generate its output in a more favorable form. Tessellation patterns do not support the full flexibility of typical hardware tessellation e.g. no fractional tessellation but do provide some useful alternatives to hardware tessellation e.g. the quad preservation previously noted, and more uniform triangles for triangular patches above right . The Tessellation class is also completely independent of evaluation in Bfr , and so can be used with any other evaluation interface. See the Tessellation subsection of the Bfr Overview for more details. API Additions See associated Doxygen for full details. New classes in namespace Bfr class Parameterization class Surface class SurfaceFactory class SurfaceFactoryCache class SurfaceFactoryMeshAdapter class RefinerSurfaceFactory class Tessellation Additions to FarTopologyLevel TopologyLevelIsVertexCorner TopologyLevelIsVertexRegular TopologyLevelIsVertexInfSharp TopologyLevelIsVertexSemiSharp TopologyLevelIsEdgeInfSharp TopologyLevelIsEdgeSemiSharp Additions to nested Options classes in Far PatchTableFactoryOptionsSetMaxIsolationLevel TopologyRefinerUniformOptionsSetRefinementLevel TopologyRefinerAdaptiveOptionsSetMaxIsolationLevel TopologyRefinerAdaptiveOptionsSetMaxSecondaryLevel Other Changes Deprecation Announcements Hbr is deprecated and will be removed from subsequent releases Improvements Suppression of GCC compiler warnings GitHub 1253, 1254, 1270 Additional methods for FarTopologyLevel GitHub 1227, 1255 Improved mixed partial derivative at Gregory patch corners GitHub 1252 Minor improvements to Far tutorials GitHub 1226, 1241 Added CMake config GitHub 1242 Updated CMake minimum version to 3.12 GitHub 1237, 1261 Updated documentation build scripts for Python 3 1265, 1266 Updated stringify build tool for improved cross compilation support GitHub 1267 Added NO_MACOS_FRAMEWORKS build option GitHub 1238 Updated Azure pipelines agents for Unbuntu and macOS GitHub 1247, 1256 Removed obsolete AppVeyor and Travis CI scripts GitHub 1259 Bug Fixes Cache active program for OsdGLComputeEvaluator GitHub 1244 Fixed member initialization warnings in OsdD3D11ComputeEvaluator GitHub 1239 Fixed GLSL shader source to remove storage qualifiers from struct members GitHub 1271 Fixed use of CMake variables for Apple builds GitHub 1235 Fixed build errors when using OpenGL without GLFW GitHub 1257 Fixed links to embedded videos GitHub 1231 Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"release_35.html\"\n }, {\n \"title\": \"Overview of Release 3.4\",\n- \"text\": \"Overview of Release 3.4 Overview of Release 3.4 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE New Features Triangular Patches for Loop Subdivision Major Improvements to Introductory Documentation Sparse Patch Tables Support for Double Precision in Far API Additions Far extensions for triangular patches Construction and refinement of topology Construction and interface of FarPatchTable Construction and use of Far stencil tables Far class templates for double precision Far member functions converted to templates for double precision OsdMeshBits OsdPatchArray Osd extensions for patch evaluation common to all shaders Osd extensions for patch tessellation common to all shaders Other Changes Improvements Bug Fixes New Features Triangular Patches for Loop Subdivision Support for the drawing and evaluation of Loop subdivision meshes with triangular patches was added. This includes the full set of Far and Osd interfaces for both evaluation and drawing. The feature set supported for Loop subdivision now matches that of Catmark, including creases, face-varying patches, non-manifold topology, etc. The long standing requirement that Loop meshes be purely triangular remains, as Loop subdivision is not defined for non-triangular faces. And as is the case with the use of the Catmark scheme, application of Loop subdivision to dense, poorly modeled meshes may lead to unexpectedly poor performance andor surface quality. The patch representation used for Loop subdivision is intended to exactly match the underlying limit surface where regular, and so uses quartic triangular Box-splines. This is in contrast to approaches that use simpler patches to approximate the Loop limit surface everywhere. As with Catmark, Gregory patches are used to approximate irregular areas. Though other choices are available that compromise surface quality in favor of improved performance, they may be less effective with Loop than they are with Catmark. Major Improvements to Introductory Documentation A significant rewrite of the Subdivision Surfaces page is included in this release. The new documentation emphasizes the piecewise parametric surface nature of subdivision surfaces and the implications of supporting arbitary topology . As a true surface primitive, the distinction between the control points and the limit surface and the corresponding operations of subdivision and tessellation that are applied to them is made clear. Sparse Patch Tables Interfaces in Far for the construction of PatchTables and the required adaptive refinement have been extended to apply to an arbitrary subset of faces. This allows patches for either large meshes or meshes that may otherwise benefit from some kind of partioning e.g. areas of static and dynamic topology to be managed in an arbitrary number of groups. In the extreme, a PatchTable forming the tree of patches for a single base face can be constructed. Client data buffers for the base mesh do not need to be partitioned and base mesh topology can be shared by multiple instances of FarTopologyRefiner used to create corresponding instances of FarPatchTables. See the new Far tutorial 5.2 for a simple example. Support for Double Precision in Far Classes and interfaces in Far have been adapted to allow support for double precision via templates parameterized for float or double. Class templates for major classes such as FarStencilTable have been introduced and the original classes preserved for compatibility. Other classes such as FarPatchTable have had methods overloaded or replaced with template functions to support both single and double precision. Internally, all use of floating point constants and math library functions has been adapted to maximize accuracy appropriate to the precision of the template instance. Interfaces in Osd have not been extended. The extensions in Far provide the basis for extensions in Osd, but demand is limited. For those benefiting from such Osd extensions, contributions are welcomed. See the revised Far tutorial 5.1 that constructs a FarPatchTable for a simple example. API Additions See associated Doxygen for full details. Far extensions for triangular patches enum PatchDescriptorTypeGREGORY_TRIANGLE PatchParamNormalizeTriangle PatchParamUnnormalizeTriangle PatchParamIsTriangleRotated Construction and refinement of topology overloaded TopologyRefinerFactoryCreate extensions to TopologyRefinerRefineAdaptive Construction and interface of FarPatchTable overloaded PatchTableFactoryCreate PatchTableFactoryGetRefineAdaptiveOptions member PatchTableFactoryOptionsincludeBaseLevelIndices member PatchTableFactoryOptionsincludeFVarBaseLevelIndices member PatchTableFactoryOptionsgenerateVaryingTables member PatchTableFactoryOptionsgenerateVaryingLocalPoints member PatchTableFactoryOptionssetPatchPrecisionDouble member PatchTableFactoryOptionssetFVarPatchPrecisionDouble PatchTableGetFVarPatchDescriptorRegular PatchTableGetFVarPatchDescriptorIrregular PatchTableGetFVarValueStride Construction and use of Far stencil tables overloaded StencilTableUpdateValues enum LimitStencilTableFactoryMode member LimitStencilTableFactoryOptionsinterpolationMode member LimitStencilTableFactoryOptionsfvarChannel Far class templates for double precision class StencilReal class StencilTableReal class StencilTableFactoryReal class LimitStencilReal class LimitStencilTableReal class LimitStencilTableFactoryReal class PrimvarRefinerReal Far member functions converted to templates for double precision PatchParamNormalize PatchParamUnnormalize PatchTableEvaluateBasis PatchTableEvaluateBasisVarying PatchTableEvaluateBasisFaceVarying PatchTableGetLocalPointStencilTable PatchTableGetLocalPointVaryingStencilTable PatchTableGetLocalPointFaceVaryingStencilTable PatchMapFindPatch OsdMeshBits enumeration MeshEndCapBilinearBasis OsdPatchArray GetDescriptorRegular GetDescriptorIrregular GetPatchTyperRegular GetPatchTyperIrregular GetStride Osd extensions for patch evaluation common to all shaders struct OsdPatchArray and OsdPatchArrayInit struct OsdPatchCoord and OsdPatchCoordInit struct OsdPatchParam and OsdPatchParamInit OsdPatchParamGetFaceId OsdPatchParamGetU OsdPatchParamGetV OsdPatchParamGetTransition OsdPatchParamGetBoundary OsdPatchParamGetNonQuadRoot OsdPatchParamGetDepth OsdPatchParamGetParamFraction OsdPatchParamIsRegular OsdPatchParamIsTriangleRotated OsdPatchParamNormalize OsdPatchParamUnnormalize OsdPatchParamNormalizeTriangle OsdPatchParamUnnormalizeTriangle OsdEvaluatePatchBasisNormalized OsdEvaluatePatchBasis Osd extensions for patch tessellation common to all shaders OsdInterpolatePatchCoordTriangle OsdComputePerPatchVertexBoxSplineTriangle OsdEvalPatchBezierTriangle OsdEvalPatchGregoryTriangle OsdGetTessLevelsUniformTriangle OsdEvalPatchBezierTessLevels OsdEvalPatchBezierTriangleTessLevels OsdGetTessParameterizationTriangle Other Changes Improvements Added new build script GitHub 1068 Added support for newer DirectX SDKs GitHub 1066 Patch arrays extended to support combined regular and irregular types GitHub 995 FarPatchTables and adaptive refinement supported for Bilinear scheme GitHub 1035 New FarPatchTableFactory method to determine adaptive refinement options GitHub 1047 New FarPatchTableFactory options to align primvar buffers of uniform tables GitHub 986 FarStencilTableUpdateValues overloaded to support separate base buffer GitHub 1011 FarLimitStencilTableFactory updated to create face-varying tables GitHub 1012 Regular patches on boundaries no longer require additional isolation GitHub 1025 Inclusion of OpenSubdiv header files in source code now consistent GitHub 767 Re-organization of and additions to Far tutorials GitHub 1083 Examples now use common command-line conventions and parsing GitHub 1056 Bug Fixes Fixed FarPrimvarRefiner internal limitFVar prototype GitHub 979 Fixed FarStencilTable append when base StencilTable empty GitHub 982 Patches around non-manifold vertices now free of cracks GitHub 1013 Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"Overview of Release 3.4 Overview of Release 3.4 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE New Features Triangular Patches for Loop Subdivision Major Improvements to Introductory Documentation Sparse Patch Tables Support for Double Precision in Far API Additions Far extensions for triangular patches Construction and refinement of topology Construction and interface of FarPatchTable Construction and use of Far stencil tables Far class templates for double precision Far member functions converted to templates for double precision OsdMeshBits OsdPatchArray Osd extensions for patch evaluation common to all shaders Osd extensions for patch tessellation common to all shaders Other Changes Improvements Bug Fixes New Features Triangular Patches for Loop Subdivision Support for the drawing and evaluation of Loop subdivision meshes with triangular patches was added. This includes the full set of Far and Osd interfaces for both evaluation and drawing. The feature set supported for Loop subdivision now matches that of Catmark, including creases, face-varying patches, non-manifold topology, etc. The long standing requirement that Loop meshes be purely triangular remains, as Loop subdivision is not defined for non-triangular faces. And as is the case with the use of the Catmark scheme, application of Loop subdivision to dense, poorly modeled meshes may lead to unexpectedly poor performance andor surface quality. The patch representation used for Loop subdivision is intended to exactly match the underlying limit surface where regular, and so uses quartic triangular Box-splines. This is in contrast to approaches that use simpler patches to approximate the Loop limit surface everywhere. As with Catmark, Gregory patches are used to approximate irregular areas. Though other choices are available that compromise surface quality in favor of improved performance, they may be less effective with Loop than they are with Catmark. Major Improvements to Introductory Documentation A significant rewrite of the Subdivision Surfaces page is included in this release. The new documentation emphasizes the piecewise parametric surface nature of subdivision surfaces and the implications of supporting arbitary topology . As a true surface primitive, the distinction between the control points and the limit surface and the corresponding operations of subdivision and tessellation that are applied to them is made clear. Sparse Patch Tables Interfaces in Far for the construction of PatchTables and the required adaptive refinement have been extended to apply to an arbitrary subset of faces. This allows patches for either large meshes or meshes that may otherwise benefit from some kind of partioning e.g. areas of static and dynamic topology to be managed in an arbitrary number of groups. In the extreme, a PatchTable forming the tree of patches for a single base face can be constructed. Client data buffers for the base mesh do not need to be partitioned and base mesh topology can be shared by multiple instances of FarTopologyRefiner used to create corresponding instances of FarPatchTables. See the new Far tutorial 5.2 for a simple example. Support for Double Precision in Far Classes and interfaces in Far have been adapted to allow support for double precision via templates parameterized for float or double. Class templates for major classes such as FarStencilTable have been introduced and the original classes preserved for compatibility. Other classes such as FarPatchTable have had methods overloaded or replaced with template functions to support both single and double precision. Internally, all use of floating point constants and math library functions has been adapted to maximize accuracy appropriate to the precision of the template instance. Interfaces in Osd have not been extended. The extensions in Far provide the basis for extensions in Osd, but demand is limited. For those benefiting from such Osd extensions, contributions are welcomed. See the revised Far tutorial 5.1 that constructs a FarPatchTable for a simple example. API Additions See associated Doxygen for full details. Far extensions for triangular patches enum PatchDescriptorTypeGREGORY_TRIANGLE PatchParamNormalizeTriangle PatchParamUnnormalizeTriangle PatchParamIsTriangleRotated Construction and refinement of topology overloaded TopologyRefinerFactoryCreate extensions to TopologyRefinerRefineAdaptive Construction and interface of FarPatchTable overloaded PatchTableFactoryCreate PatchTableFactoryGetRefineAdaptiveOptions member PatchTableFactoryOptionsincludeBaseLevelIndices member PatchTableFactoryOptionsincludeFVarBaseLevelIndices member PatchTableFactoryOptionsgenerateVaryingTables member PatchTableFactoryOptionsgenerateVaryingLocalPoints member PatchTableFactoryOptionssetPatchPrecisionDouble member PatchTableFactoryOptionssetFVarPatchPrecisionDouble PatchTableGetFVarPatchDescriptorRegular PatchTableGetFVarPatchDescriptorIrregular PatchTableGetFVarValueStride Construction and use of Far stencil tables overloaded StencilTableUpdateValues enum LimitStencilTableFactoryMode member LimitStencilTableFactoryOptionsinterpolationMode member LimitStencilTableFactoryOptionsfvarChannel Far class templates for double precision class StencilReal class StencilTableReal class StencilTableFactoryReal class LimitStencilReal class LimitStencilTableReal class LimitStencilTableFactoryReal class PrimvarRefinerReal Far member functions converted to templates for double precision PatchParamNormalize PatchParamUnnormalize PatchTableEvaluateBasis PatchTableEvaluateBasisVarying PatchTableEvaluateBasisFaceVarying PatchTableGetLocalPointStencilTable PatchTableGetLocalPointVaryingStencilTable PatchTableGetLocalPointFaceVaryingStencilTable PatchMapFindPatch OsdMeshBits enumeration MeshEndCapBilinearBasis OsdPatchArray GetDescriptorRegular GetDescriptorIrregular GetPatchTyperRegular GetPatchTyperIrregular GetStride Osd extensions for patch evaluation common to all shaders struct OsdPatchArray and OsdPatchArrayInit struct OsdPatchCoord and OsdPatchCoordInit struct OsdPatchParam and OsdPatchParamInit OsdPatchParamGetFaceId OsdPatchParamGetU OsdPatchParamGetV OsdPatchParamGetTransition OsdPatchParamGetBoundary OsdPatchParamGetNonQuadRoot OsdPatchParamGetDepth OsdPatchParamGetParamFraction OsdPatchParamIsRegular OsdPatchParamIsTriangleRotated OsdPatchParamNormalize OsdPatchParamUnnormalize OsdPatchParamNormalizeTriangle OsdPatchParamUnnormalizeTriangle OsdEvaluatePatchBasisNormalized OsdEvaluatePatchBasis Osd extensions for patch tessellation common to all shaders OsdInterpolatePatchCoordTriangle OsdComputePerPatchVertexBoxSplineTriangle OsdEvalPatchBezierTriangle OsdEvalPatchGregoryTriangle OsdGetTessLevelsUniformTriangle OsdEvalPatchBezierTessLevels OsdEvalPatchBezierTriangleTessLevels OsdGetTessParameterizationTriangle Other Changes Improvements Added new build script GitHub 1068 Added support for newer DirectX SDKs GitHub 1066 Patch arrays extended to support combined regular and irregular types GitHub 995 FarPatchTables and adaptive refinement supported for Bilinear scheme GitHub 1035 New FarPatchTableFactory method to determine adaptive refinement options GitHub 1047 New FarPatchTableFactory options to align primvar buffers of uniform tables GitHub 986 FarStencilTableUpdateValues overloaded to support separate base buffer GitHub 1011 FarLimitStencilTableFactory updated to create face-varying tables GitHub 1012 Regular patches on boundaries no longer require additional isolation GitHub 1025 Inclusion of OpenSubdiv header files in source code now consistent GitHub 767 Re-organization of and additions to Far tutorials GitHub 1083 Examples now use common command-line conventions and parsing GitHub 1056 Bug Fixes Fixed FarPrimvarRefiner internal limitFVar prototype GitHub 979 Fixed FarStencilTable append when base StencilTable empty GitHub 982 Patches around non-manifold vertices now free of cracks GitHub 1013 Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"release_34.html\"\n }, {\n- \"title\": \"Overview of Release 3.1\",\n- \"text\": \"Overview of Release 3.1 Overview of Release 3.1 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE New Features Bicubic Face-Varying Patches Varying and Face-Varying Evaluation Second Order Derivative Evaluation Separate Levels of Feature Isolation Sharp Patches for Infinitely Sharp Features API Additions OsdCpuEvaluator, GLComputeEvaluator, etc OsdCpuPatchTable, GLPatchTable, etc OsdMeshBits OsdPatchParam OsdGLSLPatchShaderSource, HLSLPatchShaderSource FarLimitStencil FarLimitStencilTable FarLimitStencilTableFactoryOptions FarPatchParam FarPatchTable FarPatchTableFactoryOptions FarTopologyLevel FarTopologyRefinerAdaptiveOptions Other Changes Improvements Bug Fixes New Features Bicubic Face-Varying Patches The motivation for this feature is to improve drawing and evaluation of face-varying primvar values for all face-varying interpolation options . A common use of non-linear face-varying data is to capture a UV projection on the surface. The following example shows a simple shape with the face-varying interpolation option assigned to a non-linear choice to achieve the desired projection in this case FVAR_LINEAR_NONE Linear Face-Varying Patches Bicubic Face-Varying Patches The result on the left shows the old linearly interpolated patches, which ignores any non-linear settings. The result on the right shows the new use of bicubic face-varying patches to accurately interpolate the desired projection. Generation of a full face-varying patch representation can be enabled using a new option in FarPatchTableFactoryOptions. Additionally, topological refinement can be improved to consider fvar channel topology using a new option in FarTopologyRefinerAdaptiveOptions. See the API additions below and their associated Doxygen text for more details. Evaluation of patch basis weights for all patch types as been added to the GPU shader source provided by OsdGLSLPatchShaderSource, and OsdHLSLPatchShaderSource. Use of non-linear face-varying patches increases the storage size of the patch table and may also require additional data access and computation while drawing. Varying and Face-Varying Evaluation This feature extends existing evaluation methods to support evaluation of varying and face-varying primvar data at arbitrary limit surface locations. Varying Primvar Evaluation Face-Varying Primvar Evaluation The image on the left shows evaluation of varying primvar values and the image on the right shows evaluation of face-varying primvar values. The EvaluateBasis API of FarPatchTable has been extended as well as the OSD Evaluator API. Second Order Derivative Evaluation This feature extends the FarLimitStencilTable and FarPatchTable interfaces to support evaluation of 2nd order partial derivatives. The generation of derivative values for FarLimitStencilTable is controlled by new options that can be specified when creating the stencil table. Additionally, the implementation exposes a more accurate method to compute derivatives for Gregory basis patches. This can be enabled using the CMake configuration and compile time definition OPENSUBDIV_GREGORY_EVAL_TRUE_DERIVATIVES. Separate Levels of Feature Isolation The primary motivation for this feature is to reduce the number of patches generated by adaptive refinement when they can be adequately represented at a lower level. A single level of isolation must be as high as the greatest semi-sharp feature to properly resolve the shape of that feature. That high isolation level generates many unnecessary patches for smooth extra-ordinary vertices. In the following example, a single semi-sharp vertex is refined to level 5 Single Isolation Level 5 Primary Level 5, Secondary Level 2 Single isolation to level 5 on the left results in 312 patches. The right shows the semi-sharp feature isolated to 5, but with the new secondary level set to 2, the number of patches is reduced to 123. The second specified level of adaptive refinement is used to halt isolation for features that typically do not require the specified maximum. These include interior and boundary extra-ordinary vertices and those infinitely sharp patches that correspond to boundary extra-ordinary patches. The secondary level is available as a new option in FarTopologyRefinerAdaptiveOptions. Sharp Patches for Infinitely Sharp Features The motivation for sharp patches is to accurately represent the limit surface of infinitely sharp features, which otherwise can only be approximated by very high levels of adaptive refinement, resulting in many unnecessary patches. The true limit surface for regular faces along an infinitely sharp crease is a regular B-Spline patch -- the same as regular faces along a boundary. Similarly, the limit surface for faces around an extra-ordinary vertex on an infinitely sharp crease is the same as that of faces around an extra-ordinary vertex on a boundary. So these patches are identified and isolated to the same degree -- the regular patches as soon as possible, and the irregular patches to the depth specified. Consider the following regressionshapecatmark_cube_creases2 Level 5 without Sharp Patches Level 5 with Sharp Patches Level 2 with Sharp Patches Without use of sharp patches on the left, isolating to level 5 generates 1764 patches and does still not capture the sharp edges. With sharp patches in the center, isolating to the same degree level 5 reduces the number of patches to 96 and captures the sharp edges. The sharp features can be captured at a lower degree with comparable accuracy as illustrated on the right where isolation to level 2 further reduces the number of patches to 42. The use of infinitely sharp patches can be enabled both at a high level as an new option to OsdMesh, or more directly when adaptively refining or construction the patch tables in FarTopologyRefinerAdaptiveOptions and FarPatchTableFactoryOptions. Given the improved accuracy and reduced patches by the use of simple regular patches, we would prefer that this be the default behavior, but it was made an explicit option in order to avoid disrupting existing usage. In a future major release this feature will hopefully be the norm. API Additions See associated Doxygen for full details. OsdCpuEvaluator, GLComputeEvaluator, etc EvalPatchesVarying EvalPatchesFaceVarying OsdCpuPatchTable, GLPatchTable, etc GetFVarPatchArrayBuffer GetFVarPatchIndexBuffer GetFVarPatchIndexSize GetFVarPatchParamBuffer GetFVarPatchParamSize GetNumFVarChannels GetVaryingPatchArrayBuffer GetVaryingPatchIndexBuffer GetVaryingPatchIndexSize OsdMeshBits member MeshFVarAdaptive member MeshUseInfSharpPatch OsdPatchParam IsRegular Unnormalize extensions to Set OsdGLSLPatchShaderSource, HLSLPatchShaderSource GetPatchBasisShaderSource FarLimitStencil GetDuuWeights GetDuvWeights GetDvvWeights extensions to LimitStencil FarLimitStencilTable GetDuuWeights GetDuvWeights GetDvvWeights Update2ndDerivs extensions to LimitStencilTable FarLimitStencilTableFactoryOptions member generate1stDerivatives member generate1stDerivatives extensions to LimitTableStencilFactoryCreate FarPatchParam IsRegular Unnormalize extensions to Set FarPatchTable ComputeLocalPointValuesFaceVarying ComputeLocalPointValuesVarying GetFVarPatchDescriptor GetFVarPatchParam GetNumLocalPointsFaceVarying GetNumLocalPointsVarying GetPatchArrayVaryingVertices GetPatchArrayFVarPatchParam GetPatchArrayFVarValues GetPatchFVarPatchParam GetPatchVaryingVertices GetVaryingPatchDescriptor GetVaryingVertices EvaluateBasisFaceVarying EvaluateBasisVarying extensions to EvaluateBasis FarPatchTableFactoryOptions member useInfSharpPatch member genenerateFVarLegacyLinearPatches FarTopologyLevel DoesEdgeFVarTopologyMatch DoesFaceFVarTopologyMatch DoesVertexFVarTopologyMatch IsEdgeBoundary IsEdgeNonManifold IsVertexBoundary IsVertexNonManifold FarTopologyRefinerAdaptiveOptions member secondaryLevel member useInfSharpPatch member considerFVarChannels Other Changes Improvements Enabled the use of CMakes folder feature Removed the use of iso646 alternative keywords and, or, not, etc. to improve portability Added numerical valued preprocessor directives OPENSUBDIV_VERSION_MAJOR, etc. to opensubdivversion.h Improved documentation for FarPatchParam and added Unnormalize to complement Normalize Added additional topology queries to FarTopologyLevel Updated glFVarViewer and glEvalLimit viewer to make use of bicubic face-varying patches Updated glViewer and dxViewer to add a toggle for InfSharpPatch Updated dxPtexViewer for improved feature parity with glPtexViewer Improved far_regression to exercise shapes independent of Hbr compatibility Added support for Appveyor continuous integration testing Removed cmakeFindIlmBase Removed mayaPolySmooth example Bug Fixes Fixed Ptex version parsing and compatibility issues Fixed compatibility issues with VS2015 Fixed bug interpolating face-varying data with Bilinear scheme Fixed bug with refinement using Chaikin creasing Fixed bugs with HUD sliders in the example viewers Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n- \"tags\": \"\",\n- \"loc\": \"release_31.html\"\n- }, {\n \"title\": \"Overview of Release 3.2\",\n- \"text\": \"Overview of Release 3.2 Overview of Release 3.2 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE New Features Face-Varying Stencil Evaluation 1st and 2nd Derivative Evaluation Smooth Corner Patch API Additions OsdCpuEvaluator, GLComputeEvaluator, etc OsdMesh OsdMeshBits FarPatchTableFactoryOptions FarStencilTableFactory FarStencilTableFactoryOptions Other Changes Improvements Bug Fixes New Features Face-Varying Stencil Evaluation Face-Varying primvar values may now be refined using stencil tables. The stencil table for a face-varying channel is created by specifying the desired fvarChannel and setting the FarStencilTableFactoryOption interpolationMode to INTERPOLATE_FACE_VARYING when creating the stencil table. 1st and 2nd Derivative Evaluation The Osd Evaluator API has been extended to support 1st derivative and 2nd partial derivative evaluation for stencils and patches. 1st Derivative Surface Normal 2nd Derivative Surface Curvature On the left is an example of computing a surface normal at each point using the evaluated 1st derivatives, while on the right is an example of computing surface curvature at each point using the evaluated 2nd partial derivatives. Smooth Corner Patch An option has been added to disable the legacy behavior of generating a sharp-corner patch at a smooth corner. Corners which are actually sharp will continue to generate sharp-corner patches. The differences between the two methods is most apparent at low-levels of feature isolation. This feature is controlled by the generateLegacySharpCornerPatches option added to FarPatchTableFactoryOptions. Sharp Corner Patch legacy behavior Smooth Corner Patch On the left is the legacy behavior of generating sharp corner patches at smooth corners. The image on the right shows the correct smooth corner patches generated when this legacy behavior is disabled. API Additions See associated Doxygen for full details. OsdCpuEvaluator, GLComputeEvaluator, etc Create EvalStencils EvalPatches EvalPatchesVarying EvalPatchesFaceVarying OsdMesh Create OsdMeshBits member MeshUseSmoothCornerPatch FarPatchTableFactoryOptions member generateLegacySharpCornerPatches FarStencilTableFactory enumeration ModeINTERPOLATE_FACE_VARYING AppendLocalPointStencilTableFaceVarying FarStencilTableFactoryOptions member fvarChannel Other Changes Improvements Corrected numerous spelling errors in doxygen comments Updated glFVarViewer with improved error detection and command line parsing Added option to build using MSVC with static CRT Bug Fixes Fixed a double delete of GL program in OsdGLComputeEvaluator Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"Overview of Release 3.2 Overview of Release 3.2 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE New Features Face-Varying Stencil Evaluation 1st and 2nd Derivative Evaluation Smooth Corner Patch API Additions OsdCpuEvaluator, GLComputeEvaluator, etc OsdMesh OsdMeshBits FarPatchTableFactoryOptions FarStencilTableFactory FarStencilTableFactoryOptions Other Changes Improvements Bug Fixes New Features Face-Varying Stencil Evaluation Face-Varying primvar values may now be refined using stencil tables. The stencil table for a face-varying channel is created by specifying the desired fvarChannel and setting the FarStencilTableFactoryOption interpolationMode to INTERPOLATE_FACE_VARYING when creating the stencil table. 1st and 2nd Derivative Evaluation The Osd Evaluator API has been extended to support 1st derivative and 2nd partial derivative evaluation for stencils and patches. 1st Derivative Surface Normal 2nd Derivative Surface Curvature On the left is an example of computing a surface normal at each point using the evaluated 1st derivatives, while on the right is an example of computing surface curvature at each point using the evaluated 2nd partial derivatives. Smooth Corner Patch An option has been added to disable the legacy behavior of generating a sharp-corner patch at a smooth corner. Corners which are actually sharp will continue to generate sharp-corner patches. The differences between the two methods is most apparent at low-levels of feature isolation. This feature is controlled by the generateLegacySharpCornerPatches option added to FarPatchTableFactoryOptions. Sharp Corner Patch legacy behavior Smooth Corner Patch On the left is the legacy behavior of generating sharp corner patches at smooth corners. The image on the right shows the correct smooth corner patches generated when this legacy behavior is disabled. API Additions See associated Doxygen for full details. OsdCpuEvaluator, GLComputeEvaluator, etc Create EvalStencils EvalPatches EvalPatchesVarying EvalPatchesFaceVarying OsdMesh Create OsdMeshBits member MeshUseSmoothCornerPatch FarPatchTableFactoryOptions member generateLegacySharpCornerPatches FarStencilTableFactory enumeration ModeINTERPOLATE_FACE_VARYING AppendLocalPointStencilTableFaceVarying FarStencilTableFactoryOptions member fvarChannel Other Changes Improvements Corrected numerous spelling errors in doxygen comments Updated glFVarViewer with improved error detection and command line parsing Added option to build using MSVC with static CRT Bug Fixes Fixed a double delete of GL program in OsdGLComputeEvaluator Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"release_32.html\"\n }, {\n \"title\": \"Overview of Release 3.3\",\n- \"text\": \"Overview of Release 3.3 Overview of Release 3.3 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE New Features Metal Compute and Tessellation API Additions OsdMTLContext OsdMTLComputeEvaluator OsdMTLStencilTable OsdMTLMeshInterface OsdMTLPatchTable OsdMTLLegacyGregoryPatchTable OsdPatchShaderSource OsdCPUMTLVertexBuffer Other Changes Improvements New Features Metal Compute and Tessellation Added support for drawing and evaluation using Apples Metal API for high performance and low-overhead GPU access. This includes the full set of Osd interfaces needed to draw using Metal graphics rendering including tessellation shaders and evaluate stencils and patch tables using Metal compute processing. Also includes an example mtlViewer that can be built to run on either macOS or iOS. Metal Graphics Rendering with Tessellation Metal Patch Evaluation using Compute Processing API Additions See associated Doxygen for full details. OsdMTLContext device and commandQueue member data OsdMTLComputeEvaluator Create EvalStencils EvalPatches EvalPatchesVarying EvalPatchesFaceVarying Compile Synchronize OsdMTLStencilTable Create GetSizesBuffer GetIndicesBuffer GetDuWeightsBuffer GetDvWeightsBuffer GetDuuWeightsBuffer GetDuvWeightsBuffer GetDvvWeightsBuffer GetNumStencils OsdMTLMeshInterface GetNumVertices GetMaxValence UpdateVertexBuffer UpdateVaryingBuffer Refine Synchronize GetPatchTable GetFarPatchTable BindVertexBuffer BindVaryingBuffer OsdMTLPatchTable Create GetPatchArrays GetPatchIndexBuffer GetPatchParamBuffer GetVaryingPatchArrays GetVaryingPatchIndexBuffer GetNumFVarChannels GetFVarPatchArrays GetFVarPatchIndexBuffer GetFVarPatchParamBuffer OsdMTLLegacyGregoryPatchTable UpdateVertexBuffer GetVertexBuffer GetVertexValenceBuffer GetQuadOffsetsBuffer OsdPatchShaderSource GetCommonShaderSource GetPatchBasisShaderSource GetVertexShaderSource GetHullShaderSource GetDomainShaderSource OsdCPUMTLVertexBuffer Create UpdateData GetNumElements GetNumVertices BindCpuBuffer BindMTLBuffer BindVBO Other Changes Improvements Fixed several instances of local variable shadowing that could cause build warnings Updated continuous-integration build scripts and added testing on macOS Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"Overview of Release 3.3 Overview of Release 3.3 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE New Features Metal Compute and Tessellation API Additions OsdMTLContext OsdMTLComputeEvaluator OsdMTLStencilTable OsdMTLMeshInterface OsdMTLPatchTable OsdMTLLegacyGregoryPatchTable OsdPatchShaderSource OsdCPUMTLVertexBuffer Other Changes Improvements New Features Metal Compute and Tessellation Added support for drawing and evaluation using Apples Metal API for high performance and low-overhead GPU access. This includes the full set of Osd interfaces needed to draw using Metal graphics rendering including tessellation shaders and evaluate stencils and patch tables using Metal compute processing. Also includes an example mtlViewer that can be built to run on either macOS or iOS. Metal Graphics Rendering with Tessellation Metal Patch Evaluation using Compute Processing API Additions See associated Doxygen for full details. OsdMTLContext device and commandQueue member data OsdMTLComputeEvaluator Create EvalStencils EvalPatches EvalPatchesVarying EvalPatchesFaceVarying Compile Synchronize OsdMTLStencilTable Create GetSizesBuffer GetIndicesBuffer GetDuWeightsBuffer GetDvWeightsBuffer GetDuuWeightsBuffer GetDuvWeightsBuffer GetDvvWeightsBuffer GetNumStencils OsdMTLMeshInterface GetNumVertices GetMaxValence UpdateVertexBuffer UpdateVaryingBuffer Refine Synchronize GetPatchTable GetFarPatchTable BindVertexBuffer BindVaryingBuffer OsdMTLPatchTable Create GetPatchArrays GetPatchIndexBuffer GetPatchParamBuffer GetVaryingPatchArrays GetVaryingPatchIndexBuffer GetNumFVarChannels GetFVarPatchArrays GetFVarPatchIndexBuffer GetFVarPatchParamBuffer OsdMTLLegacyGregoryPatchTable UpdateVertexBuffer GetVertexBuffer GetVertexValenceBuffer GetQuadOffsetsBuffer OsdPatchShaderSource GetCommonShaderSource GetPatchBasisShaderSource GetVertexShaderSource GetHullShaderSource GetDomainShaderSource OsdCPUMTLVertexBuffer Create UpdateData GetNumElements GetNumVertices BindCpuBuffer BindMTLBuffer BindVBO Other Changes Improvements Fixed several instances of local variable shadowing that could cause build warnings Updated continuous-integration build scripts and added testing on macOS Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"release_33.html\"\n }, {\n- \"title\": \"Overview of Release 3.0\",\n- \"text\": \"Overview of Release 3.0 Overview of Release 3.0 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Release 3.0 Subdivision Core Sdc Topology and Refinement Limit Properties and Patches Faster Evaluation and Display Updated Source-Code Style Documentation and Tutorials Additional Resources Porting Guide Subdivision Compatibility Release 3.0 OpenSubdiv 3.0 represents a landmark release, with profound changes to the core algorithms, simplified APIs, and streamlined GPU execution. Providing faster, more efficient, and more flexible subdivision code remains our principal goal. To achieve this, OpenSubdiv 3.0 introduces many improvements that constitute a fairly radical departure from previous versions. This document highlights some of the major changes that have gone in to the 3.0 release. Subdivision Core Sdc In consideration of past, present and future topological representations, all low-level details fundamental to subdivision and the specific subdivision schemes have been factored into a new low-level layer called Sdc SubDivision Core. This layer encapsulates the full set of applicable options, the formulae required to support semi-sharp creasing, the formulae for the refinement masks of each subdivision scheme, etc. As initially conceived, its goal was often expressed as separating the math from the mesh. Sdc provides the low-level nuts and bolts to provide a subdivision implementation consistent with OpenSubdiv. It is used by OpenSubdivs libraries and may also be useful in providing an existing clients implementation with the details necessary to make that implementation consistent with OpenSubdiv. Topology and Refinement OpenSubdiv 3.0 introduces a new intermediate internal topological representation named Vtr Vectorized Topology Representation. Compared to the Hbr library used in previous versions, Vtr is much more efficient for the kinds of topological analysis required by Far and is more flexible. While Hbr is no longer used by OpenSubdiv, it will remain in the source distribution for legacy and regression purposes. Faster Subdivision A major focus of the 3.0 release is performance, and the improvement to the initial refinement of a mesh required for topological analysis is close to an order magnitude often much more for uniform, but less for adaptive. Supporting for Non-manifold Topology With topology conversion no longer constrained by Hbr, OpenSubdiv is no longer restricted to meshes of manifold topology. With one exception non-triangles with Loop subdivision, any set of faces and vertices that can be represented in common container formats such as Obj or Alembic can be represented and subdivided. With future efforts to bring the functionality for the Loop scheme up to par with Catmark, that last remaining topological restriction will be removed. Simpler Conversion of Topology Several entry-points are now available for client topology, rather than the single incremental assembly of an HbrMesh that previously existed. The new topological relationships can be populated using either a high-level interface where simplicity has been emphasized, or a more complex lower-level interface for enhanced efficiency. Face Varying Topology Previously, face-varying data was assigned by value to the vertex for each face, and whether or not the set of values around a vertex was continuous was determined by comparing these values later. In some cases this could result in two values that were not meant to be shared being welded together. Face-varying data is now specified topologically just as the vertex topology is defined from a set of vertices and integer references indices to these vertices for the corner of each face, face-varying topology is defined from a set of values and integer references indices to these values for the corner of each face. So if values are to be considered distinct around a vertex, they are given distinct indices and no comparison of any data is ever performed. Note that the number of vertices and values will typically differ, but since indices are assigned to the corners of all faces for both, the total number of indices assigned to all faces will be the same. This ensures that OpenSubdivs face-varying topology matches what is often specified in common geometry container formats like Obj, Alembic and USD. Multiple channels of face-varying data can be defined and each is topologically independent of the others. Limit Properties and Patches A fundamental goal of OpenSubdiv is to provide an accurate and reliable representation of the limit surface. Improvements have been made both to the properties positions and tangents at discrete points in the subdivision hierarchy, as well as to the representations of patches used for the continuous limit surface between them. Removed Fixed Valence Tables Limit properties of extra-ordinary vertices are computed for arbitrary valence and new patch types no longer rely on small table sizes. All tables that restricted the valence of a vertex to some relatively small table size have now been removed. The only restriction on valence that exists is within the new topology representation, which restricts it to the size of an unsigned 16-bit integer 65,535. This limit could also be removed, by recompiling with a certain size changed from 16- to 32-bits, but doing so would increase the memory cost for all common cases. We feel the 16-bit limit is a reasonable compromise. Single Crease Patch OpenSubdiv 3.0 newly implements efficient evaluation of semi-smooth creases using single crease patches. With this optimization, high-order edge sharpness tags can be handled very efficiently for both computation time and memory consumption. Niessner et al., Efficient Evaluation of Semi-Smooth Creases in Catmull-Clark Subdivision Surfaces. Eurographics Short Papers. 2012. httpresearch.microsoft.comen-usumpeoplecloopEG2012.pdf New Irregular Patch Approximations While legacy Gregory patch support is still available, we have introduced several new options for representing irregular patches Legacy Gregory, fast Gregory Basis stencils, and BSpline patches. Gregory basis stencils provide the same high quality approximation of Legacy Gregory patches, but execute considerably faster with a simpler GPU representation. While BSpline patches are not as close an approximation as Gregory patches, they enable an entire adaptively refined mesh to be drawn with screen space tessellation via a single global shader configuration Gregory Basis patches require one additional global shader configuration. The new implementations of the GregoryBasis and BSpline approximations relax the previous max valence limit. Legacy Gregory patch still has a limitation of max valence typically 24, depending on the hardware capability of GL_MAX_VARYING_VECTORS. Users are still encouraged to use models with vertices of low valence for both improved model quality and performance. Faster Evaluation and Display OpenSubdiv 3.0 also introduces new data structures and algorithms that greatly enhance performance for the common case of repeated evaluation both on the CPU and GPU. Introducing Stencil Tables OpenSubdiv 3.0 replaces the serialized subdivision tables with factorized stencil tables. The SubdivisionTables class of earlier releases contained a large number of data inter-dependencies, which incurred penalties from fences or force additional kernel launches. Most of these dependencies have now been factorized away in the pre-computation stage, yielding stencil tables FarStencilTable instead. Stencils remove all data dependencies and simplify all the computations into a single trivial kernel. This simplification results in a faster pre-computation stage, faster execution on GPU, with less driver overhead. The new stencil tables Compute back-end is supported on all the same platforms as previous releases except GCD. Faster, Simpler GPU Kernels On the GPU side, the replacement of subdivision tables with stencils greatly reduces bottlenecks in compute, yielding as much as a 4x interpolation speed-up. At the same time, stencils reduce the complexity of interpolation to a single kernel launch per primitive, a critical improvement for mobile platforms. As a result of these changes, compute batching is now trivial, which in turn enabled API simplifications in the Osd layer. Unified Adaptive Shaders Adaptive tessellation shader configurations have been greatly simplified. The number of shader configurations has been reduced from a combinatorial per-patch explosion down to a constant two global configurations. This massive improvement over the 2.x code base results in significantly faster load times and a reduced per-frame cost for adaptive drawing. Similar to compute kernel simplification, this shader simplification has resulted in additional simplifications in the Osd layer. Updated Source-Code Style OpenSubdiv 3.0 replaces naming prefixes with C namespaces for all API layers, bringing the source style more in line with contemporary specifications mostly inspired from the Google C Style Guide . The large-scale changes introduced in this release generally break compatibility with existing client-code. However, this gives us the opportunity to effect some much needed updates to our code-style guidelines and general conventions, throughout the entire OpenSubdiv code-base. We are hoping to drastically improve the quality, consistency and readability of the source code. Documentation and Tutorials The documentation has been reorganized and fleshed out. This release introduces a number of new tutorials . The tutorials provide an easier entry point for learning the API than do the programs provided in examples. The examples provide more fleshed out solutions and are a good next step after the tutorials are mastered. Additional Resources Porting Guide Please see the Porting Guide for help on how to port existing code written for OpenSubdiv 2.x to the new 3.0 release. Subdivision Compatibility The 3.0 release has made some minor changes to the subdivision specification and rules. See Subdivision Compatibility for a complete list. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"Overview of Release 3.1\",\n+ \"text\": \"Overview of Release 3.1 Overview of Release 3.1 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE New Features Bicubic Face-Varying Patches Varying and Face-Varying Evaluation Second Order Derivative Evaluation Separate Levels of Feature Isolation Sharp Patches for Infinitely Sharp Features API Additions OsdCpuEvaluator, GLComputeEvaluator, etc OsdCpuPatchTable, GLPatchTable, etc OsdMeshBits OsdPatchParam OsdGLSLPatchShaderSource, HLSLPatchShaderSource FarLimitStencil FarLimitStencilTable FarLimitStencilTableFactoryOptions FarPatchParam FarPatchTable FarPatchTableFactoryOptions FarTopologyLevel FarTopologyRefinerAdaptiveOptions Other Changes Improvements Bug Fixes New Features Bicubic Face-Varying Patches The motivation for this feature is to improve drawing and evaluation of face-varying primvar values for all face-varying interpolation options . A common use of non-linear face-varying data is to capture a UV projection on the surface. The following example shows a simple shape with the face-varying interpolation option assigned to a non-linear choice to achieve the desired projection in this case FVAR_LINEAR_NONE Linear Face-Varying Patches Bicubic Face-Varying Patches The result on the left shows the old linearly interpolated patches, which ignores any non-linear settings. The result on the right shows the new use of bicubic face-varying patches to accurately interpolate the desired projection. Generation of a full face-varying patch representation can be enabled using a new option in FarPatchTableFactoryOptions. Additionally, topological refinement can be improved to consider fvar channel topology using a new option in FarTopologyRefinerAdaptiveOptions. See the API additions below and their associated Doxygen text for more details. Evaluation of patch basis weights for all patch types as been added to the GPU shader source provided by OsdGLSLPatchShaderSource, and OsdHLSLPatchShaderSource. Use of non-linear face-varying patches increases the storage size of the patch table and may also require additional data access and computation while drawing. Varying and Face-Varying Evaluation This feature extends existing evaluation methods to support evaluation of varying and face-varying primvar data at arbitrary limit surface locations. Varying Primvar Evaluation Face-Varying Primvar Evaluation The image on the left shows evaluation of varying primvar values and the image on the right shows evaluation of face-varying primvar values. The EvaluateBasis API of FarPatchTable has been extended as well as the OSD Evaluator API. Second Order Derivative Evaluation This feature extends the FarLimitStencilTable and FarPatchTable interfaces to support evaluation of 2nd order partial derivatives. The generation of derivative values for FarLimitStencilTable is controlled by new options that can be specified when creating the stencil table. Additionally, the implementation exposes a more accurate method to compute derivatives for Gregory basis patches. This can be enabled using the CMake configuration and compile time definition OPENSUBDIV_GREGORY_EVAL_TRUE_DERIVATIVES. Separate Levels of Feature Isolation The primary motivation for this feature is to reduce the number of patches generated by adaptive refinement when they can be adequately represented at a lower level. A single level of isolation must be as high as the greatest semi-sharp feature to properly resolve the shape of that feature. That high isolation level generates many unnecessary patches for smooth extra-ordinary vertices. In the following example, a single semi-sharp vertex is refined to level 5 Single Isolation Level 5 Primary Level 5, Secondary Level 2 Single isolation to level 5 on the left results in 312 patches. The right shows the semi-sharp feature isolated to 5, but with the new secondary level set to 2, the number of patches is reduced to 123. The second specified level of adaptive refinement is used to halt isolation for features that typically do not require the specified maximum. These include interior and boundary extra-ordinary vertices and those infinitely sharp patches that correspond to boundary extra-ordinary patches. The secondary level is available as a new option in FarTopologyRefinerAdaptiveOptions. Sharp Patches for Infinitely Sharp Features The motivation for sharp patches is to accurately represent the limit surface of infinitely sharp features, which otherwise can only be approximated by very high levels of adaptive refinement, resulting in many unnecessary patches. The true limit surface for regular faces along an infinitely sharp crease is a regular B-Spline patch -- the same as regular faces along a boundary. Similarly, the limit surface for faces around an extra-ordinary vertex on an infinitely sharp crease is the same as that of faces around an extra-ordinary vertex on a boundary. So these patches are identified and isolated to the same degree -- the regular patches as soon as possible, and the irregular patches to the depth specified. Consider the following regressionshapecatmark_cube_creases2 Level 5 without Sharp Patches Level 5 with Sharp Patches Level 2 with Sharp Patches Without use of sharp patches on the left, isolating to level 5 generates 1764 patches and does still not capture the sharp edges. With sharp patches in the center, isolating to the same degree level 5 reduces the number of patches to 96 and captures the sharp edges. The sharp features can be captured at a lower degree with comparable accuracy as illustrated on the right where isolation to level 2 further reduces the number of patches to 42. The use of infinitely sharp patches can be enabled both at a high level as an new option to OsdMesh, or more directly when adaptively refining or construction the patch tables in FarTopologyRefinerAdaptiveOptions and FarPatchTableFactoryOptions. Given the improved accuracy and reduced patches by the use of simple regular patches, we would prefer that this be the default behavior, but it was made an explicit option in order to avoid disrupting existing usage. In a future major release this feature will hopefully be the norm. API Additions See associated Doxygen for full details. OsdCpuEvaluator, GLComputeEvaluator, etc EvalPatchesVarying EvalPatchesFaceVarying OsdCpuPatchTable, GLPatchTable, etc GetFVarPatchArrayBuffer GetFVarPatchIndexBuffer GetFVarPatchIndexSize GetFVarPatchParamBuffer GetFVarPatchParamSize GetNumFVarChannels GetVaryingPatchArrayBuffer GetVaryingPatchIndexBuffer GetVaryingPatchIndexSize OsdMeshBits member MeshFVarAdaptive member MeshUseInfSharpPatch OsdPatchParam IsRegular Unnormalize extensions to Set OsdGLSLPatchShaderSource, HLSLPatchShaderSource GetPatchBasisShaderSource FarLimitStencil GetDuuWeights GetDuvWeights GetDvvWeights extensions to LimitStencil FarLimitStencilTable GetDuuWeights GetDuvWeights GetDvvWeights Update2ndDerivs extensions to LimitStencilTable FarLimitStencilTableFactoryOptions member generate1stDerivatives member generate1stDerivatives extensions to LimitTableStencilFactoryCreate FarPatchParam IsRegular Unnormalize extensions to Set FarPatchTable ComputeLocalPointValuesFaceVarying ComputeLocalPointValuesVarying GetFVarPatchDescriptor GetFVarPatchParam GetNumLocalPointsFaceVarying GetNumLocalPointsVarying GetPatchArrayVaryingVertices GetPatchArrayFVarPatchParam GetPatchArrayFVarValues GetPatchFVarPatchParam GetPatchVaryingVertices GetVaryingPatchDescriptor GetVaryingVertices EvaluateBasisFaceVarying EvaluateBasisVarying extensions to EvaluateBasis FarPatchTableFactoryOptions member useInfSharpPatch member genenerateFVarLegacyLinearPatches FarTopologyLevel DoesEdgeFVarTopologyMatch DoesFaceFVarTopologyMatch DoesVertexFVarTopologyMatch IsEdgeBoundary IsEdgeNonManifold IsVertexBoundary IsVertexNonManifold FarTopologyRefinerAdaptiveOptions member secondaryLevel member useInfSharpPatch member considerFVarChannels Other Changes Improvements Enabled the use of CMakes folder feature Removed the use of iso646 alternative keywords and, or, not, etc. to improve portability Added numerical valued preprocessor directives OPENSUBDIV_VERSION_MAJOR, etc. to opensubdivversion.h Improved documentation for FarPatchParam and added Unnormalize to complement Normalize Added additional topology queries to FarTopologyLevel Updated glFVarViewer and glEvalLimit viewer to make use of bicubic face-varying patches Updated glViewer and dxViewer to add a toggle for InfSharpPatch Updated dxPtexViewer for improved feature parity with glPtexViewer Improved far_regression to exercise shapes independent of Hbr compatibility Added support for Appveyor continuous integration testing Removed cmakeFindIlmBase Removed mayaPolySmooth example Bug Fixes Fixed Ptex version parsing and compatibility issues Fixed compatibility issues with VS2015 Fixed bug interpolating face-varying data with Bilinear scheme Fixed bug with refinement using Chaikin creasing Fixed bugs with HUD sliders in the example viewers Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"release_30.html\"\n+ \"loc\": \"release_31.html\"\n }, {\n- \"title\": \"OSD Overview\",\n- \"text\": \"OSD Overview OSD Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE OpenSubdiv Osd Refinement Limit Stencil Evaluation Limit Evaluation with PatchTable OpenGLDX11Metal Drawing with Hardware Tessellation InterleavedBatched Buffer Configuration Cross-Platform Implementation OpenSubdiv Osd Osd contains device dependent code that makes Far structures available on various backends such as TBB, CUDA, OpenCL, GLSL, etc. The main roles of Osd are Refinement Compute stencil-based uniformadaptive subdivision on CPUGPU backends Limit Stencil Evaluation Compute limit surfaces by limit stencils on CPUGPU backends Limit Evaluation with PatchTable Compute limit surfaces by patch evaluation on CPUGPU backends OpenGLDX11Metal Drawing with hardware tessellation Provide GLSLHLSLMetal tessellation functions for patch table InterleavedBatched buffer configuration Provide consistent buffer descriptor to deal with arbitrary buffer layout. Cross-Platform Implementation Provide convenient classes to interop between compute and draw APIs These are independently used by clients. For example, a client can use only the limit stencil evaluation, or a client can refine subdivision surfaces and draw them with the PatchTable and Osd tessellation shaders. All device specific evaluation kernels are implemented in the Evaluator classes. Since Evaluators dont own vertex buffers, clients should provide their own buffers as a source and destination. There are some interop classes defined in Osd for convenience. OpenSubdiv utilizes a series of regression tests to compare and enforce identical results across different computational devices. Refinement Osd supports both uniform and feature adaptive subdivision. Once clients create a FarStencilTable for the topology, they can convert it into device-specific stencil tables if necessary. The following table shows which evaluator classes and stencil table interfaces can be used together. Note that while Osd provides these stencil table classes which can be easily constructed from FarStencilTable, clients arent required to use these table classes. Clients may have their own entities as a stencil tables as long as EvaluatorEvalStencils can access the necessary interfaces. Backend Evaluator class compatible stencil table CPU CPU single-threaded CpuEvaluator FarStencilTable TBB CPU multi-threaded TbbEvaluator FarStencilTable OpenMP CPU multi-threaded OmpEvaluator FarStencilTable CUDA GPU CudaEvaluator CudaStencilTable OpenCL CPUGPU CLEvaluator CLStencilTable GL ComputeShader GPU GLComputeEvaluator GLStencilTableSSBO GL Transform Feedback GPU GLXFBEvaluator GLStencilTableTBO DX11 ComputeShader GPU D3D11ComputeEvaluator D3D11StencilTable Metal GPU MTLComputeEvaluator MTLStencilTable Limit Stencil Evaluation Limit stencil evaluation is quite similar to refinement in Osd . At first clients create FarLimitStencilTable for the locations to evaluate the limit surfaces, then convert it into an evaluator compatible stencil table and call EvaluatorEvalStencils. Limit Evaluation with PatchTable Another way to evaluate the limit surfaces is to use the PatchTable. Once all control vertices and local points are resolved by the stencil evaluation, Osd can evaluate the limit surfaces through the PatchTable. Backend Evaluator class compatible patch table CPU CPU single-threaded CpuEvaluator CpuPatchTable TBB CPU multi-threaded TbbEvaluator CpuPatchTable OpenMP CPU multi-threaded OmpEvaluator CpuPatchTable CUDA GPU CudaEvaluator CudaPatchTable OpenCL CPUGPU CLEvaluator CLPatchTable GL ComputeShader GPU GLComputeEvaluator GLPatchTable GL Transform Feedback GPU GLXFBEvaluator GLPatchTable DX11 ComputeShader GPU D3D11ComputeEvaluator not yet supported D3D11PatchTable Metal ComputeShader GPU MTLComputeEvaluator MTLPatchTable Release Notes 3.x Osd evaluation backends EvaluatorEvalPatches do not support evaluation of single-crease or Legacy Gregory patch types. OpenGLDX11Metal Drawing with Hardware Tessellation One of the most interesting use cases of the Osd layer is realtime drawing of subdivision surfaces using hardware tessellation. This is somewhat similar to limit evaluation with PatchTable described above. Drawing differs from limit evaluation in that Osd provides shader snippets for patch evaluation and clients will inject them into their own shader source. See shader interface for a more detailed discussion of the shader interface. InterleavedBatched Buffer Configuration All Osd layer APIs assume that each primitive variables to be computed points, colors, uvs ... are contiguous arrays of 32bit floating point values. The Osd API refers to such an array as a buffer. A buffer can exist on CPU memory or GPU memory. Osd Evaluators typically take one source buffer and one destination buffer, or three destination buffers if derivatives are being computed. Osd Evaluators also take BufferDescriptors, that are used to specify the layout of the source and destination buffers. A BufferDescriptor is a struct of 3 integers which specify an offset, length and stride. For example Vertex 0 Vertex 1 ... X Y Z X Y Z ... The layout of this buffer can be described as Osd BufferDescriptor desc offset 0 , length 3 , stride 3 BufferDescriptor can be used for an interleaved buffer too. Vertex 0 Vertex 1 ... X Y Z R G B A X Y Z R G B A ... Osd BufferDescriptor xyzDesc 0 , 3 , 7 Osd BufferDescriptor rgbaDesc 3 , 4 , 7 Although the source and destination buffers dont need to be the same buffer for EvalStencils, adaptive patch tables are constructed to first index the coarse vertices and the refined vertices immediately afterward. In this case, the BufferDescriptor for the destination should include the offset as the number of coarse vertices to be skipped. Coarse vertices n Src Refined vertices Dst Vertex 0 Vertex 1 ... Vertex n Vertex n1 X Y Z X Y Z ... X Y Z X Y Z ... Osd BufferDescriptor srcDesc 0 , 3 , 3 Osd BufferDescriptor dstDesc n 3 , 3 , 3 Also note that the source descriptor doesnt have to start with offset 0. This is useful when a client has a big buffer with multiple objects batched together. Cross-Platform Implementation One of the key goals of OpenSubdiv is to achieve as much cross-platform flexibility as possible and leverage all optimized hardware paths where available. This can be very challenging as there is a very large variety of plaftorms and APIs available, with very distinct capabilities. In Osd , Evaluators dont care about interops between those APIs. All Evaluators have two kinds of APIs for both EvalStencils and EvalPatches. Explicit signatures which directly take device-specific buffer representation e.g., pointer for CpuEvaluator, GLuint buffer for GLComputeEvaluator, etc. Generic signatures which take arbitrary buffer classes. The buffer class is required to have a certain method to return the device-specific buffer representation. The later interface is useful if the client supports multiple backends at the same time. The methods that need to be implemented for the Evaluators are Evaluator class object method CpuEvaluator TbbEvaluator OmpEvaluator pointer to cpu memory BindCpuBuffer CudaEvaluator pointer to cuda memory BindCudaBuffer CLEvaluator cl_mem BindCLBuffer GLComputeEvaluator GLXFBEvaluator GL buffer object BindVBO D3D11ComputeEvaluator D3D11 UAV BindD3D11UAV MTLComputeEvaluator MTLBuffer BindMTLBuffer The buffers can use these methods as a trigger of interop. Osd provides a default implementation of interop buffer for most of the backend combinations. For example, if the client wants to use CUDA as a computation backend and use OpenGL as the drawing API, OsdCudaGLVertexBuffer fits the case since it implements BindCudaBuffer and BindVBO. Again, clients can implement their own buffer class and pass it to the Evaluators. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"Overview of Release 3.0\",\n+ \"text\": \"Overview of Release 3.0 Overview of Release 3.0 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Release 3.0 Subdivision Core Sdc Topology and Refinement Limit Properties and Patches Faster Evaluation and Display Updated Source-Code Style Documentation and Tutorials Additional Resources Porting Guide Subdivision Compatibility Release 3.0 OpenSubdiv 3.0 represents a landmark release, with profound changes to the core algorithms, simplified APIs, and streamlined GPU execution. Providing faster, more efficient, and more flexible subdivision code remains our principal goal. To achieve this, OpenSubdiv 3.0 introduces many improvements that constitute a fairly radical departure from previous versions. This document highlights some of the major changes that have gone in to the 3.0 release. Subdivision Core Sdc In consideration of past, present and future topological representations, all low-level details fundamental to subdivision and the specific subdivision schemes have been factored into a new low-level layer called Sdc SubDivision Core. This layer encapsulates the full set of applicable options, the formulae required to support semi-sharp creasing, the formulae for the refinement masks of each subdivision scheme, etc. As initially conceived, its goal was often expressed as separating the math from the mesh. Sdc provides the low-level nuts and bolts to provide a subdivision implementation consistent with OpenSubdiv. It is used by OpenSubdivs libraries and may also be useful in providing an existing clients implementation with the details necessary to make that implementation consistent with OpenSubdiv. Topology and Refinement OpenSubdiv 3.0 introduces a new intermediate internal topological representation named Vtr Vectorized Topology Representation. Compared to the Hbr library used in previous versions, Vtr is much more efficient for the kinds of topological analysis required by Far and is more flexible. While Hbr is no longer used by OpenSubdiv, it will remain in the source distribution for legacy and regression purposes. Faster Subdivision A major focus of the 3.0 release is performance, and the improvement to the initial refinement of a mesh required for topological analysis is close to an order magnitude often much more for uniform, but less for adaptive. Supporting for Non-manifold Topology With topology conversion no longer constrained by Hbr, OpenSubdiv is no longer restricted to meshes of manifold topology. With one exception non-triangles with Loop subdivision, any set of faces and vertices that can be represented in common container formats such as Obj or Alembic can be represented and subdivided. With future efforts to bring the functionality for the Loop scheme up to par with Catmark, that last remaining topological restriction will be removed. Simpler Conversion of Topology Several entry-points are now available for client topology, rather than the single incremental assembly of an HbrMesh that previously existed. The new topological relationships can be populated using either a high-level interface where simplicity has been emphasized, or a more complex lower-level interface for enhanced efficiency. Face Varying Topology Previously, face-varying data was assigned by value to the vertex for each face, and whether or not the set of values around a vertex was continuous was determined by comparing these values later. In some cases this could result in two values that were not meant to be shared being welded together. Face-varying data is now specified topologically just as the vertex topology is defined from a set of vertices and integer references indices to these vertices for the corner of each face, face-varying topology is defined from a set of values and integer references indices to these values for the corner of each face. So if values are to be considered distinct around a vertex, they are given distinct indices and no comparison of any data is ever performed. Note that the number of vertices and values will typically differ, but since indices are assigned to the corners of all faces for both, the total number of indices assigned to all faces will be the same. This ensures that OpenSubdivs face-varying topology matches what is often specified in common geometry container formats like Obj, Alembic and USD. Multiple channels of face-varying data can be defined and each is topologically independent of the others. Limit Properties and Patches A fundamental goal of OpenSubdiv is to provide an accurate and reliable representation of the limit surface. Improvements have been made both to the properties positions and tangents at discrete points in the subdivision hierarchy, as well as to the representations of patches used for the continuous limit surface between them. Removed Fixed Valence Tables Limit properties of extra-ordinary vertices are computed for arbitrary valence and new patch types no longer rely on small table sizes. All tables that restricted the valence of a vertex to some relatively small table size have now been removed. The only restriction on valence that exists is within the new topology representation, which restricts it to the size of an unsigned 16-bit integer 65,535. This limit could also be removed, by recompiling with a certain size changed from 16- to 32-bits, but doing so would increase the memory cost for all common cases. We feel the 16-bit limit is a reasonable compromise. Single Crease Patch OpenSubdiv 3.0 newly implements efficient evaluation of semi-smooth creases using single crease patches. With this optimization, high-order edge sharpness tags can be handled very efficiently for both computation time and memory consumption. Niessner et al., Efficient Evaluation of Semi-Smooth Creases in Catmull-Clark Subdivision Surfaces. Eurographics Short Papers. 2012. httpresearch.microsoft.comen-usumpeoplecloopEG2012.pdf New Irregular Patch Approximations While legacy Gregory patch support is still available, we have introduced several new options for representing irregular patches Legacy Gregory, fast Gregory Basis stencils, and BSpline patches. Gregory basis stencils provide the same high quality approximation of Legacy Gregory patches, but execute considerably faster with a simpler GPU representation. While BSpline patches are not as close an approximation as Gregory patches, they enable an entire adaptively refined mesh to be drawn with screen space tessellation via a single global shader configuration Gregory Basis patches require one additional global shader configuration. The new implementations of the GregoryBasis and BSpline approximations relax the previous max valence limit. Legacy Gregory patch still has a limitation of max valence typically 24, depending on the hardware capability of GL_MAX_VARYING_VECTORS. Users are still encouraged to use models with vertices of low valence for both improved model quality and performance. Faster Evaluation and Display OpenSubdiv 3.0 also introduces new data structures and algorithms that greatly enhance performance for the common case of repeated evaluation both on the CPU and GPU. Introducing Stencil Tables OpenSubdiv 3.0 replaces the serialized subdivision tables with factorized stencil tables. The SubdivisionTables class of earlier releases contained a large number of data inter-dependencies, which incurred penalties from fences or force additional kernel launches. Most of these dependencies have now been factorized away in the pre-computation stage, yielding stencil tables FarStencilTable instead. Stencils remove all data dependencies and simplify all the computations into a single trivial kernel. This simplification results in a faster pre-computation stage, faster execution on GPU, with less driver overhead. The new stencil tables Compute back-end is supported on all the same platforms as previous releases except GCD. Faster, Simpler GPU Kernels On the GPU side, the replacement of subdivision tables with stencils greatly reduces bottlenecks in compute, yielding as much as a 4x interpolation speed-up. At the same time, stencils reduce the complexity of interpolation to a single kernel launch per primitive, a critical improvement for mobile platforms. As a result of these changes, compute batching is now trivial, which in turn enabled API simplifications in the Osd layer. Unified Adaptive Shaders Adaptive tessellation shader configurations have been greatly simplified. The number of shader configurations has been reduced from a combinatorial per-patch explosion down to a constant two global configurations. This massive improvement over the 2.x code base results in significantly faster load times and a reduced per-frame cost for adaptive drawing. Similar to compute kernel simplification, this shader simplification has resulted in additional simplifications in the Osd layer. Updated Source-Code Style OpenSubdiv 3.0 replaces naming prefixes with C namespaces for all API layers, bringing the source style more in line with contemporary specifications mostly inspired from the Google C Style Guide . The large-scale changes introduced in this release generally break compatibility with existing client-code. However, this gives us the opportunity to effect some much needed updates to our code-style guidelines and general conventions, throughout the entire OpenSubdiv code-base. We are hoping to drastically improve the quality, consistency and readability of the source code. Documentation and Tutorials The documentation has been reorganized and fleshed out. This release introduces a number of new tutorials . The tutorials provide an easier entry point for learning the API than do the programs provided in examples. The examples provide more fleshed out solutions and are a good next step after the tutorials are mastered. Additional Resources Porting Guide Please see the Porting Guide for help on how to port existing code written for OpenSubdiv 2.x to the new 3.0 release. Subdivision Compatibility The 3.0 release has made some minor changes to the subdivision specification and rules. See Subdivision Compatibility for a complete list. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"osd_overview.html\"\n+ \"loc\": \"release_30.html\"\n }, {\n- \"title\": \"References\",\n- \"text\": \"References References 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE References References Analytic Displacement Mapping using Hardware Tessellation Matthias Niessner, Charles Loop ACM Transactions on Graphics, Vol. 32 No. 3 Article 26 June 2013 httpresearch.microsoft.comen-usumpeoplecloopTOG2013.pdf httpdoi.org10.11452487228.2487234 Feature Adaptive GPU Rendering of Catmull-Clark Subdivision Surfaces Matthias Niessner, Charles Loop, Mark Meyer, and Tony DeRose ACM Transactions on Graphics, Vol. 31 No. 1 Article 6 January 2012 httpresearch.microsoft.comen-usumpeopleclooptog2012.pdf httpdoi.org10.11452077341.2077347 Efficient Evaluation of Semi-Smooth Creases in Catmull-Clark Subdivision Surfaces Matthias Niessner, Charles Loop, and Guenter Greiner. Eurographics Proceedings, Cagliari, 2012 httpresearch.microsoft.comen-usumpeoplecloopEG2012.pdf httpdoi.org10.2312confEG2012short041-044 Approximating Subdivision Surfaces with Gregory Patches for Hardware Tessellation Charles Loop, Scott Schaefer, Tianyun Ni, Ignacio Castano SIGGRAPH Asia Conference Proceedings 2009 httpwww.dgp.toronto.edupeoplestamrealityResearchpdfsig98.pdf httpdoi.org10.11451661412.1618497 GPU Smoothing of Quad Meshes T. L. Ni, Y. Yeo, A. Myles, V. Goel and J. Peters Proc. IEEE SMI 2008 httpwww.cise.ufl.eduresearchSurfLabpaperssmi08.pdf httpdoi.org10.1109SMI.2008.4547938 Fast Parallel Construction of Smooth Surfaces from Meshes with TriQuadPent Facets A. Myles and T. Ni and J. Peters Eurographics Symposium on Geometry Processing 2008 httpswww.cise.ufl.eduresearchSurfLabpapers08poly.pdf httpdoi.org10.1111j.1467-8659.2008.01276.x Approximating Catmull-Clark Subdivision Surfaces with Bicubic Patches Charles Loop, Scott Schaefer ACM Transactions on Graphics, Vol. 27 No. 1 Article 8 March 2008 httpresearch.microsoft.comen-usumpeoplecloopacctog.pdf httpdoi.org10.11451330511.1330519 Rapid Evaluation of Catmull-Clark Subdivision Surfaces Jeffrey Bolz and Peter Schroder Web3D Proceedings 2002 httpwww.multires.caltech.edupubsfastsubd.pdf httpdoi.org10.1145504502.504505 Piecewise Smooth Subdivision Surfaces with Normal Control Henning Biermann, Adi Levin and Denis Zorin SIGGRAPH 2000 Conference Proceedings httpmrl.nyu.edudzorinpapersbiermann2000pss.pdf httpdoi.org10.1145344779.344841 Subdivision for Modeling and Animation Denis Zorin, Peter Schroder Course Notes of SIGGRAPH 1999 httpwww.multires.caltech.edupubssig99notes.pdf Exact Evaluation of Catmull-Clark Subdivision Surfaces at Arbitrary Parameter Values Jos Stam SIGGRAPH 98 Conference Proceedings, Annual Conference Series, July 1998 httpwww.dgp.toronto.edupeoplestamrealityResearchpdfsig98.pdf httpdoi.org10.1145280814.280945 Subdivision Surfaces in Character Animation Tony DeRose, Michael Kass, Tien Truong Proceedings of SIGGRAPH 1998 httpgraphics.pixar.comlibraryGeripaper.pdf httpdoi.org10.1145280814.280826 Efficient, Fair Interpolation Using Catmull-Clark Surfaces Mark Halstead, Michael Kass, Tony DeRose SIGGRAPH 93 Conference Proceedings httpgraphics.pixar.comlibraryFairSubdivisionpaper.pdf httpdoi.org10.1145166117.166121 Recursively generated B-spline surfaces on arbitrary topological meshes Catmull, E. Clark, J. Computer-Aided Design 10 6 1978 httpdoi.org10.10160010-448528782990110-0 Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"Porting Guide 2.x to 3.0\",\n+ \"text\": \"Porting Guide 2.x to 3.0 Porting Guide 2.x to 3.0 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Porting Guide 2.x to 3.0 Source Code Organization Hbr Layer Translation Subdivision Schemes and Options in Sdc Specifying Face Varying Topology and Options Far Layer Translation Ordering of Refined Vertices Osd Layer Translation Controller Objects ComputeContext, DrawContext EvalLimitContext OsdMesh OsdKernelBatch OsdVertex Feature Adaptive Shader Changes End Cap Strategies Changes to Subdivision Build Support for Combining 2.x and 3.0 Porting Guide 2.x to 3.0 This document is a high-level description of how to port exiting OpenSubdiv 2.x code to use OpenSubdiv 3.0. NOTE If your questions are not answered here, please contact us on the OpenSubdiv forum and we will be happy to help Source Code Organization Given the scale of functional changes that were being made to the public interface, we took the opportunity in 3.0 to update the coding style and organization -- most notably making use of namespaces for each library. Subdirectory Namespace Relevance hbr NA Historical, no longer used sdc Sdc New, low-level, public options, constants, etc. vtr Vtr New, internal use, topology representation far Far Revised, similar functionality with new API osd Osd Revised, similar functionality with new API Hbr Layer Translation Client mesh topology is now translated into an instance of FarTopologyRefiner instead of HbrMesh. OpenSubdiv 2.x OpenSubdiv 3.0 HbrMeshVTX_TYPE FarTopologyRefiner The FarTopologyRefiner is now the core representation of topology from which all other major classes in Far and Osd are constructed. It was designed to support efficient refinement uniform or sparse of a base mesh of arbitrary topology no manifold restrictions. Once constructed it can be directly refined to meet some need, or passed to other contexts that will refine it to meet their needs. In contrast to directly assembling an HbrMesh, the TopologyRefiner, like other classes in Far, requires a Factory class for its construction. One of the early goals of these factories was to allow a client to convert their existing boundary representation -- with its full topological traversal abilities -- directly into the TopologyRefiners representation. While this is now possible, this also represents the most complex construction process and is only recommended for usage where this conversion process is critical. Details on how to construct a TopologyRefiner can be found in the Far overview documentation. Additionally, documentation for FarTopologyRefinerFactoryMESH outlines the requirements, and Far tutorial 3.1 tutorialsfartutorial_3_1 provides an example of a factory for directly converting HbrMeshes to TopologyRefiners. Its worth a reminder here that FarTopologyRefiner contains only topological information which does include sharpness, since that is considered relating to subdivision topology and not the positions or other data associated with a mesh. While HbrMeshT required some definition of a vertex type T and dimensions of face-varying data, TopologyRefiner is more clearly separated from the data. So the construction of the TopologyRefiner does not involve data specification at all. Subdivision Schemes and Options in Sdc The creation of a new TopologyRefiner requires specification of a subdivision scheme and a set of options that are applicable to all schemes. With HbrMesh, the scheme was specified by declaring a static instance of a specific subclass of a subdivision object, and the options were specified with a number of methods on the different classes. Such general information about the schemes has now been encapsulated in the Sdc layer for use throughout OpenSubdiv. The subdivision scheme is now a simple enumerated type SdcSchemeType and the entire set of options that can be applied to a scheme is encapsulated in a single simple struct of flags and enumerated types SdcOptions. OpenSubdiv 2.x OpenSubdiv 3.0 HbrMeshTSetInterpolateBoundaryMethod SdcOptionsSetVtxBoundaryInterpolation HbrMeshTSetFVarInterpolateBoundaryMethod SdcOptionsSetFVarLinearInterpolation HbrSubdivisionTSetCreaseSubdivisionMethod SdcOptionsSetCreasingMethod Specifying Face Varying Topology and Options Both the way in which face varying data is associated with a mesh and the options used to control its interpolation have changed. The documentation on Compatibility with OpenSubdiv 2.x details the equivalence of interpolation options between Hbr and the new SdcOptionsFVarLinearInterpolation enum, while the section on Face Varying Interpolation illustrates their effects. Face varying data is now specified by index rather than by value, or as often stated, it is specified topologically. Just as vertices for faces are specified by indices into a potential buffer of positions, face varying values are specified by indices into a potential buffer of values. Both vertices and face varying values frequently referred to as FVarValues in the API are assigned and associated with the corners of all faces. In many cases this will simplify representation as many common geometry container formats such as Obj or Alembic specify texture coordinates the same way. For other cases, where a value per face-corner is provided with no indication of which values incident each vertex should be considered shared, it will be necessary to determine shared indices for values at each vertex if any non-linear interpolation is desired. Far Layer Translation While TopologyRefiner was introduced into Far as the new intermediate topology representation, several other changes were made to classes in Far to provide more modular building blocks for use by the Osd layer or directly. OpenSubdiv 2.x OpenSubdiv 3.0 FarMeshU NA, no longer needed FarSubdivisionTables FarStencilTable FarPatchTables FarPatchTable Ordering of Refined Vertices The FarMesh was previously responsible for refining an HbrMesh -- generating new vertices and faces in successive levels of refinement in the FarSubdivisionTables. Vertices were grouped and reordered from the native ordering of HbrMesh so that vertices requiring similar processing were consecutive. Such grouping alleviated most of the idiosyncrasies of HbrMeshs native ordering but not all. FarToplogyRefiner is inherently a collection of refinement levels, and within each refined level so excluding the base level, all components are still grouped for the same reasons. There are two issues here though the ordering of these groups has changed though an option exists to preserve it the ordering of components within these groups is not guaranteed to have been preserved Vertices in a refined level are grouped according to the type of component in the parent level from which they originated, i.e. some vertices originate from the center of a face face-vertices, some from an edge edge-vertices and some from a vertex vertex-vertices. Note that there is a conflict in terminology here -- face-vertices and edge-vertices most often refer to vertices incident a face or edge -- but for the sake of this discussion, we use them to refer to the component from which a child vertex originates. The following table shows the ordering of these groups in 2.x and the two choices available in 3.0. The option is the orderVerticesFromFacesFirst flag that can be set in the Option structs passed to the uniform and adaptive refinement methods of TopologyRefiner Version and option Vertex group ordering 2.x face-vertices, edge-vertices, vertex-vertices 3.0 default vertex-vertices, face-vertices, edge-vertices 3.0 orderVerticesFromFacesFirst true face-vertices, edge-vertices, vertex-vertices The decision to change the default ordering was based on common feedback the rationale was to allow a trivial mapping from vertices in the cage to their descendants at all refinement levels. While the grouping is fundamental to the refinement process, the ordering of the groups is internally flexible, and the full set of possible orderings can be made publicly available in future if there is demand for such flexibility. The ordering of vertices within these groups was never clearly defined given the way that HbrMesh applied its refinement. For example, for the face-vertices in a level, it was never clear which face-vertices would be first as it depended on the order in which HbrMesh traversed the parent faces and generated them. Given one face, HbrMesh would often visit neighboring faces first before moving to the next intended face. The ordering with FarTopologyRefiner is much clearer and predictable. Using the face-vertices as an example, the order of the face-vertices in level N1 is identical to the order of the parent faces in level N from which they originated. So if we have face-vertices Vi , Vj and Vk at some level, originating from faces Fi , Fj and Fk in the previous level, they will be ordered in increasing order of i , j and k . For uniform refinement the ordering of face vertices Vi will therefore exactly match the ordering of the parent faces Fi . For adaptive or otherwise sparse refinement, the subset of Vi will be ordered similarly, just with components missing from those not refined. The same is true of all vertices, i.e. edge-vertices and vertex-vertices, and also for other components in refined levels, i.e. the child faces and edges. For child faces and edges, more than one will originate from the same parent face or edge. In addition to the overall ordering based on the parent faces or edges, another ordering is imposed on multiple children originating from the same face or edge. They will be ordered based on the corner or end-vertex with which they are associated. In the case of refined faces, another way to view the ordering is to consider the way that faces are originally defined -- by specifying the set of vertices for the corners of each face, often aggregated into a single large array. The ordering of the set of refined faces for each level will correspond directly to such an array of vertices per face in the previous level. Osd Layer Translation Two big changes in the 3.0 API have allowed the Osd layer to be significantly simpler, the first is the move to stencil tables from subdivision tables and the second is shader simplification. With this refactoring, the focus has been to use more meaningful names and to make the data contained within an object more apparent. Controller Objects The API-specific ComputeController has been replaced with the Evaluator . It reflects the fact that stencil compute batches are significantly simpler than subdivision table compute batches. The name Evaluator was chosen with the hope that is more meaningful than the generic ComputeController moniker the Evaluator evaluates stencil and patch tables. In the 2.x code base, subdiv level buffers were always required to be allocated contiguously, however in 3.0 with the shift to stencil tables, this strict allocation scheme is no longer required. As a result, the EvalStencils and EvalPatches methods now accept both a source and a destination descriptor. OpenSubdiv 2.x OpenSubdiv 3.0 ComputeControllerRefine Osd...EvaluatorEvalStencils ComputeControllerSynchronize Osd...EvaluatorSynchronize EvalStencilsControllerUpdateValues Osd...EvaluatorEvalStencils EvalStencilsControllerUpdateDerivs Osd...EvaluatorEvalStencils EvalLimitControllerEvalLimitSample Osd...EvaluatorEvalPatches Also note that OsdVertexDescriptor has been renamed, however its data members and semantic purpose remains the same OpenSubdiv 2.x OpenSubdiv 3.0 OsdVertexBufferDescriptor OsdBufferDescriptor ComputeContext, DrawContext ComputeContext and DrawContext have been replaced with API-specific StencilTable and PatchTable objects, for example OsdGLStencilTableSSBO. OpenSubdiv 2.x OpenSubdiv 3.0 ComputeContext Osd...StencilTable e.g. GLStencilTableTBO EvalStencilsContext Osd...StencilTable DrawContext Osd...PatchTable e.g. GLPatchTable EvalLimitContext The data stored in EvalLimitContext has been merged into the Evaluator class as well. EvalCoords have been moved into their own type, OsdPatchCoords. The primary change here is that the PTex face ID is no longer part of the data structure, rather the client can use a FarPatchMap to convert from PTex face ID to a FarPatchTablePatchHandle. OpenSubdiv 2.x OpenSubdiv 3.0 EvalLimitContext PatchTable EvalLimitContextEvalCoords OsdPatchCoords types.h OsdMesh While not strictly required, OsdMesh is still supported in 3.0 as convenience API for allocating buffers. OsdMesh serves as a simple way to allocate all required data, in the location required by the API for example, GPU buffers for OpenGL. OsdKernelBatch No translation, it is no longer part of the API. OsdVertex No translation, it is no longer part of the API. Feature Adaptive Shader Changes In 3.0, the feature adaptive screen-space tessellation shaders have been dramatically simplified, and the client-facing API has changed dramatically as well. The primary shift is to reduce the total number of shader combinations, and as a result, some of the complexity management mechanisms are no longer necessary. In the discussion below, some key changes are highlighted, but deep integrations may require additional discussion please feel free to send follow up questions to the OpenSubdiv google group. The number of feature adaptive shaders has been reduced from N to exactly 1 or 2, depending on how end-caps are handled. Osd layer no longer compiles shaders, rather it returns shader source for the client to compile. This source is obtained via OsdGLSLHLSLPatchShaderSource. The API exposed in shaders to access patch-based data has been consolidated and formalized, see osdglslPatchCommon.glsl and osdhlslPatchCommon.hlsl for details. Patches are no longer rotated and transition patches have been eliminated, simplifying PatchDescriptor to a 4 bits. Additionally, FarPatchTablesDescriptor has been moved into its own class in the Far namespace. The following table outlines the API translation between 2.x and 3.0 OpenSubdiv 2.x OpenSubdiv 3.0 OsdDrawContextPatchDescriptor NA, no longer needed. OsdDrawContextPatchArray OSdPatchArray types.h FarPatchTablesPatchDescriptor FarPatchDescriptor patchDescriptor.h FarPatchTablesPatchArray made private. End Cap Strategies By default, OpenSubdiv uses Gregory patches to approximate the patches around extraordinary vertices at the maximum isolation level, this process is referred to as end-capping. If ENDCAP_BSPLINE_BASIS is specified to PatchTableFactoryOptions, BSpline patches are used, which gives less accuracy, but it makes possible to render an entire mesh in a single draw call. Both patches require additional control points that are not part of the mesh, we refer to these as local points. In 3.0, the local points of those patches are computed by applying a stencil table to refined vertices to construct a new stencil table for the local points. Since this new stencil table is topologically compatible with the primary stencil table for refinement, it is convenient and efficient to splice those stencil tables together. This splicing can be done in the following way FarStencilTable const refineStencils FarStencilTableFactoryCreatetopologyRefiner FarPatchTable cosnt patchTable FarPatchTableFactoryCreatetopologyRefiner FarStencilTable const localPointStencils patchTable-GetLocalPointStencilTable FarStencilTable const splicedStencils FarStencilTableFactoryAppendLocalPointStencilTablestopologyRefiner, refineStencils, localPointStencils NOTE Once the spliced stencil table is created, the refined stencils can be released, but the local point stencils are owned by patchTable, it should not be released. OpenSubdiv 3.0 also supports 2.x style Gregory patches, if ENDCAP_LEGACY_GREGORY is specified to PatchTableFactoryOptions. In this case, such an extra stencil splicing isnt needed, however clients must still bind additional buffers VertexValence buffer and QuadOffsets buffer. See OsdGLLegacyGregoryPatchTable for additional details. Changes to Subdivision The refactoring of OpenSubdiv 3.0 data representations presented a unique opportunity to revisit some corners of the subdivision specification and remove or update some legacy features -- none of which was taken lightly. More details are provided in Subdivision Compatibility , while the following offers a quick overview All face-varying interpolation options have been combined into a single enum. Vertex interpolation options have been renamed or removed The naming of the standard creasing method has changed from Normal to Uniform . Unused legacy modes of the smoothtriangle option have been removed. The averaging of Chaikin creasing with infinitely sharp edges has changed. Support for Hierarchical Edits has been removed. Build Support for Combining 2.x and 3.0 Running OpenSubdiv 2.0 and 3.0 in a single process is supported, however some special care must be taken to avoid namespace collisions, both in terms of run-time symbols avoid using OpenSubdivOsd, for example and in terms of build-time search paths. To support both OpenSubdiv 2.0 and 3.0 in your build environment, you can prefix the header install directory of OpenSubdiv 3.0. Do this using the build flag CMAKE_INCDIR_BASE when configuring cmake i.e. -DCMAKE_INCDIR_BASEincludeopensubdiv3 and then including files from opensubdiv3... in client code. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"references.html\"\n+ \"loc\": \"porting.html\"\n }, {\n \"title\": \"OSD Tessellation shader Interface\",\n- \"text\": \"OSD Tessellation shader Interface OSD Tessellation shader Interface 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Basic Tessellation Control Shader Example for B-Spline patches Tessellation Evaluation Shader Example for B-Spline patches Basis Conversion B-spline Patch Gregory Basis Patch Box-spline Triangle Patch Gregory Triangle Patch Legacy Gregory Patch 2.x compatibility Tessellation levels Tessellation levels at each tessellated vertex Tessellation levels computed at each patch Uniform Screenspace Basic Starting with 3.0, Osd tessellation shaders can be used as a set of functions from client shader code. In order to tessellate Osd patches, client shader code should perform the following steps regular B-spline patch case In a tessellation control shader fetch a PatchParam for the current patch call OsdComputePerPatchVertexBSpline to compute OsdPerPatchVertexBezier. compute tessellation level. To prevent cracks on transition patches, two vec4 parameters tessOuterHi, tessOuterLo will be needed in addition to built-in gl_TessLevelInnerOuters. In a tessellation evaluation shader call OsdGetTessParameterization to remap gl_TessCoord to a patch parameter at which to evaluate. call OsdEvalPatchBezierOsdEvalPatchGregory to evaluate the current patch. The following is a minimal example of GLSL code explaining how client shader code uses OpenSubdiv shader functions to tessellate patches of a patch table. Tessellation Control Shader Example for B-Spline patches layout vertices 16 out in vec3 position patch out vec4 tessOuterLo , tessOuterHi out OsdPerPatchVertexBezier v void main Get a patch param from texture buffer. ivec3 patchParam OsdGetPatchParam gl_PrimitiveID Compute per-patch vertices. OsdComputePerPatchVertexBSpline patchParam , gl_InvocationID , position , v Compute tessellation factors. if gl_InvocationID 0 vec4 tessLevelOuter vec4 0 vec2 tessLevelInner vec2 0 OsdGetTessLevelsUniform patchParam , tessLevelOuter , tessLevelInner , tessOuterLo , tessOuterHi gl_TessLevelOuter 0 tessLevelOuter 0 gl_TessLevelOuter 1 tessLevelOuter 1 gl_TessLevelOuter 2 tessLevelOuter 2 gl_TessLevelOuter 3 tessLevelOuter 3 gl_TessLevelInner 0 tessLevelInner 0 gl_TessLevelInner 1 tessLevelInner 1 Tessellation Evaluation Shader Example for B-Spline patches layout quads in patch in vec4 tessOuterLo , tessOuterHi in OsdPerPatchVertexBezier v uniform mat4 mvpMatrix void main Compute tesscoord. vec2 UV OsdGetTessParameterization gl_TessCoord . xy , tessOuterLo , tessOuterHi vec3 P vec3 0 , dPu vec3 0 , dPv vec3 0 vec3 N vec3 0 , dNu vec3 0 , dNv vec3 0 ivec3 patchParam inpt 0 . v . patchParam Evaluate patch at the tess coord UV OsdEvalPatchBezier patchParam , UV , v , P , dPu , dPv , N , dNu , dNv Apply model-view-projection matrix. gl_Position mvpMatrix vec4 P , 1 Basis Conversion B-spline Patch The following diagram shows how the Osd shaders process b-spline patches. While regular patches are expressed as b-spline patches in FarPatchTable, the Osd shader converts them into Bezier basis patches for simplicity and efficiency. This conversion is performed in the tessellation control stage. The boundary edge evaluation and single crease matrix evaluation are also resolved during this conversion. OsdComputePerPatchVertexBSpline can be used for this process. The resulting Bezier control vertices are stored in OsdPerPatchVertexBezier struct. void OsdComputePerPatchVertexBSpline ivec3 patchParam , int ID , vec3 cv 16 , out OsdPerPatchVertexBezier result The tessellation evaluation shader takes an array of OsdPerPatchVertexBezier struct, and then evaluates the patch using the OsdEvalPatchBezier function. void OsdEvalPatchBezier ivec3 patchParam , vec2 UV , OsdPerPatchVertexBezier cv 16 , out vec3 P , out vec3 dPu , out vec3 dPv , out vec3 N , out vec3 dNu , out vec3 dNv Gregory Basis Patch In a similar way, Gregory basis patches are processed as follows OsdComputePerPatchVertexGregoryBasis can be used for the Gregory patches although no basis conversion involved for the Gregory patches and the resulting vertices are stored in a OsdPerPatchVertexGreogryBasis struct. void OsdComputePerPatchVertexGregoryBasis ivec3 patchParam , int ID , vec3 cv , out OsdPerPatchVertexGregoryBasis result The tessellation evaluation shader takes an array of OsdPerPatchVertexGregoryBasis struct, and then evaluates the patch using the OsdEvalPatchGregory function. void OsdEvalPatchGregory ivec3 patchParam , vec2 UV , vec3 cv 20 , out vec3 P , out vec3 dPu , out vec3 dPv , out vec3 N , out vec3 dNu , out vec3 dNv Box-spline Triangle Patch While regular triangle patches are expressed as triangular box-spline patches in FarPatchTable, the Osd shader converts them into triangular Bezier patches for consistency. This conversion is performed in the tessellation control stage. The boundary edge evaluation is resolved during this conversion. OsdComputePerPatchVertexBoxSplineTriangle can be used for this process. The resulting Bezier control vertices are stored in OsdPerPatchVertexBezier struct. void OsdComputePerPatchVertexBoxSplineTriangle ivec3 patchParam , int ID , vec3 cv 12 , out OsdPerPatchVertexBezier result The tessellation evaluation shader takes an array of OsdPerPatchVertexBezier struct, and then evaluates the patch using the OsdEvalPatchBezierTriangle function. void OsdEvalPatchBezierTriangle ivec3 patchParam , vec2 UV , OsdPerPatchVertexBezier cv 15 , out vec3 P , out vec3 dPu , out vec3 dPv , out vec3 N , out vec3 dNu , out vec3 dNv Gregory Triangle Patch OsdComputePerPatchVertexGregoryBasis can be used for the quartic triangular Gregory patches although no basis conversion involved for the Gregory triangle patches and the resulting vertices are stored in a OsdPerPatchVertexGreogryBasis struct. void OsdComputePerPatchVertexGregoryBasis ivec3 patchParam , int ID , vec3 cv , out OsdPerPatchVertexGregoryBasis result The tessellation evaluation shader takes an array of OsdPerPatchVertexGregoryBasis struct, and then evaluates the patch using the OsdEvalPatchGregoryTriangle function. void OsdEvalPatchGregoryTriangle ivec3 patchParam , vec2 UV , vec3 cv 18 , out vec3 P , out vec3 dPu , out vec3 dPv , out vec3 N , out vec3 dNu , out vec3 dNv Legacy Gregory Patch 2.x compatibility OpenSubdiv 3.0 also supports 2.x style Gregory patch evaluation see far_overview. In order to evaluate a legacy Gregory patch, client needs to bind extra buffers and to perform extra steps in the vertex shader as shown in the following diagram Tessellation levels Osd provides both uniform and screen-space adaptive tessellation level computation. Because of the nature of feature adaptive subdivision , we need to pay extra attention for a patchs outer tessellation level for the screen-space adaptive case so that cracks dont appear. An edge of the patch marked as a transition edge is split into two segments Hi and Lo. The Osd shaders uses these two segments to ensure the same tessellation along the edge between different levels of subdivision. In the following example, suppose the left hand side patch has determined the tessellation level of its right edge to be 5. gl_TessLevelOuter is set to 5 for the edge, and at the same time we also pass 2 and 3 to the tessellation evaluation shader as separate levels for the two segments of the edge split at the middle. Tessellation levels at each tessellated vertex The tessellation evaluation shader takes gl_TessCoord and those two values, and remaps gl_TessCoord using OsdGetTessParameterization or OsdGetTessLevelParameterizationTriangle to ensure the parameters are consistent across adjacent patches. vec2 OsdGetTessParameterization vec2 uv , vec4 tessOuterLo , vec4 tessOuterHi vec2 OsdGetTessParameterizationTriangle vec3 uvw , vec4 tessOuterLo , vec4 tessOuterHi Tessellation levels computed at each patch These tessellation levels can be computed the corresponding method in the tesselation control shader. Note that these functions potentially requires all bezier control points, you need to call barrier to ensure the conversion is done for all invocations. See osdglslPatchBSpline.glsl for more details. Uniform void OsdGetTessLevelsUniform ivec3 patchParam , out vec4 tessLevelOuter , out vec2 tessLevelInner , out vec4 tessOuterLo , out vec4 tessOuterHi void OsdGetTessLevelsUniformTriangle ivec3 patchParam , out vec4 tessLevelOuter , out vec2 tessLevelInner , out vec4 tessOuterLo , out vec4 tessOuterHi Screenspace void OsdEvalPatchBezierTessLevels OsdPerPatchVertexBezier cpBezier 16 , ivec3 patchParam , out vec4 tessLevelOuter , out vec2 tessLevelInner , out vec4 tessOuterLo , out vec4 tessOuterHi void OsdEvalPatchBezierTriangleTessLevels vec3 cv 15 , ivec3 patchParam , out vec4 tessLevelOuter , out vec2 tessLevelInner , out vec4 tessOuterLo , out vec4 tessOuterHi Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"OSD Tessellation shader Interface OSD Tessellation shader Interface 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Basic Tessellation Control Shader Example for B-Spline patches Tessellation Evaluation Shader Example for B-Spline patches Basis Conversion B-spline Patch Gregory Basis Patch Box-spline Triangle Patch Gregory Triangle Patch Legacy Gregory Patch 2.x compatibility Tessellation levels Tessellation levels at each tessellated vertex Tessellation levels computed at each patch Uniform Screenspace Basic Starting with 3.0, Osd tessellation shaders can be used as a set of functions from client shader code. In order to tessellate Osd patches, client shader code should perform the following steps regular B-spline patch case In a tessellation control shader fetch a PatchParam for the current patch call OsdComputePerPatchVertexBSpline to compute OsdPerPatchVertexBezier. compute tessellation level. To prevent cracks on transition patches, two vec4 parameters tessOuterHi, tessOuterLo will be needed in addition to built-in gl_TessLevelInnerOuters. In a tessellation evaluation shader call OsdGetTessParameterization to remap gl_TessCoord to a patch parameter at which to evaluate. call OsdEvalPatchBezierOsdEvalPatchGregory to evaluate the current patch. The following is a minimal example of GLSL code explaining how client shader code uses OpenSubdiv shader functions to tessellate patches of a patch table. Tessellation Control Shader Example for B-Spline patches layout vertices 16 out in vec3 position patch out vec4 tessOuterLo , tessOuterHi out OsdPerPatchVertexBezier v void main Get a patch param from texture buffer. ivec3 patchParam OsdGetPatchParam gl_PrimitiveID Compute per-patch vertices. OsdComputePerPatchVertexBSpline patchParam , gl_InvocationID , position , v Compute tessellation factors. if gl_InvocationID 0 vec4 tessLevelOuter vec4 0 vec2 tessLevelInner vec2 0 OsdGetTessLevelsUniform patchParam , tessLevelOuter , tessLevelInner , tessOuterLo , tessOuterHi gl_TessLevelOuter 0 tessLevelOuter 0 gl_TessLevelOuter 1 tessLevelOuter 1 gl_TessLevelOuter 2 tessLevelOuter 2 gl_TessLevelOuter 3 tessLevelOuter 3 gl_TessLevelInner 0 tessLevelInner 0 gl_TessLevelInner 1 tessLevelInner 1 Tessellation Evaluation Shader Example for B-Spline patches layout quads in patch in vec4 tessOuterLo , tessOuterHi in OsdPerPatchVertexBezier v uniform mat4 mvpMatrix void main Compute tesscoord. vec2 UV OsdGetTessParameterization gl_TessCoord . xy , tessOuterLo , tessOuterHi vec3 P vec3 0 , dPu vec3 0 , dPv vec3 0 vec3 N vec3 0 , dNu vec3 0 , dNv vec3 0 ivec3 patchParam inpt 0 . v . patchParam Evaluate patch at the tess coord UV OsdEvalPatchBezier patchParam , UV , v , P , dPu , dPv , N , dNu , dNv Apply model-view-projection matrix. gl_Position mvpMatrix vec4 P , 1 Basis Conversion B-spline Patch The following diagram shows how the Osd shaders process b-spline patches. While regular patches are expressed as b-spline patches in FarPatchTable, the Osd shader converts them into Bezier basis patches for simplicity and efficiency. This conversion is performed in the tessellation control stage. The boundary edge evaluation and single crease matrix evaluation are also resolved during this conversion. OsdComputePerPatchVertexBSpline can be used for this process. The resulting Bezier control vertices are stored in OsdPerPatchVertexBezier struct. void OsdComputePerPatchVertexBSpline ivec3 patchParam , int ID , vec3 cv 16 , out OsdPerPatchVertexBezier result The tessellation evaluation shader takes an array of OsdPerPatchVertexBezier struct, and then evaluates the patch using the OsdEvalPatchBezier function. void OsdEvalPatchBezier ivec3 patchParam , vec2 UV , OsdPerPatchVertexBezier cv 16 , out vec3 P , out vec3 dPu , out vec3 dPv , out vec3 N , out vec3 dNu , out vec3 dNv Gregory Basis Patch In a similar way, Gregory basis patches are processed as follows OsdComputePerPatchVertexGregoryBasis can be used for the Gregory patches although no basis conversion involved for the Gregory patches and the resulting vertices are stored in a OsdPerPatchVertexGreogryBasis struct. void OsdComputePerPatchVertexGregoryBasis ivec3 patchParam , int ID , vec3 cv , out OsdPerPatchVertexGregoryBasis result The tessellation evaluation shader takes an array of OsdPerPatchVertexGregoryBasis struct, and then evaluates the patch using the OsdEvalPatchGregory function. void OsdEvalPatchGregory ivec3 patchParam , vec2 UV , vec3 cv 20 , out vec3 P , out vec3 dPu , out vec3 dPv , out vec3 N , out vec3 dNu , out vec3 dNv Box-spline Triangle Patch While regular triangle patches are expressed as triangular box-spline patches in FarPatchTable, the Osd shader converts them into triangular Bezier patches for consistency. This conversion is performed in the tessellation control stage. The boundary edge evaluation is resolved during this conversion. OsdComputePerPatchVertexBoxSplineTriangle can be used for this process. The resulting Bezier control vertices are stored in OsdPerPatchVertexBezier struct. void OsdComputePerPatchVertexBoxSplineTriangle ivec3 patchParam , int ID , vec3 cv 12 , out OsdPerPatchVertexBezier result The tessellation evaluation shader takes an array of OsdPerPatchVertexBezier struct, and then evaluates the patch using the OsdEvalPatchBezierTriangle function. void OsdEvalPatchBezierTriangle ivec3 patchParam , vec2 UV , OsdPerPatchVertexBezier cv 15 , out vec3 P , out vec3 dPu , out vec3 dPv , out vec3 N , out vec3 dNu , out vec3 dNv Gregory Triangle Patch OsdComputePerPatchVertexGregoryBasis can be used for the quartic triangular Gregory patches although no basis conversion involved for the Gregory triangle patches and the resulting vertices are stored in a OsdPerPatchVertexGreogryBasis struct. void OsdComputePerPatchVertexGregoryBasis ivec3 patchParam , int ID , vec3 cv , out OsdPerPatchVertexGregoryBasis result The tessellation evaluation shader takes an array of OsdPerPatchVertexGregoryBasis struct, and then evaluates the patch using the OsdEvalPatchGregoryTriangle function. void OsdEvalPatchGregoryTriangle ivec3 patchParam , vec2 UV , vec3 cv 18 , out vec3 P , out vec3 dPu , out vec3 dPv , out vec3 N , out vec3 dNu , out vec3 dNv Legacy Gregory Patch 2.x compatibility OpenSubdiv 3.0 also supports 2.x style Gregory patch evaluation see far_overview. In order to evaluate a legacy Gregory patch, client needs to bind extra buffers and to perform extra steps in the vertex shader as shown in the following diagram Tessellation levels Osd provides both uniform and screen-space adaptive tessellation level computation. Because of the nature of feature adaptive subdivision , we need to pay extra attention for a patchs outer tessellation level for the screen-space adaptive case so that cracks dont appear. An edge of the patch marked as a transition edge is split into two segments Hi and Lo. The Osd shaders uses these two segments to ensure the same tessellation along the edge between different levels of subdivision. In the following example, suppose the left hand side patch has determined the tessellation level of its right edge to be 5. gl_TessLevelOuter is set to 5 for the edge, and at the same time we also pass 2 and 3 to the tessellation evaluation shader as separate levels for the two segments of the edge split at the middle. Tessellation levels at each tessellated vertex The tessellation evaluation shader takes gl_TessCoord and those two values, and remaps gl_TessCoord using OsdGetTessParameterization or OsdGetTessLevelParameterizationTriangle to ensure the parameters are consistent across adjacent patches. vec2 OsdGetTessParameterization vec2 uv , vec4 tessOuterLo , vec4 tessOuterHi vec2 OsdGetTessParameterizationTriangle vec3 uvw , vec4 tessOuterLo , vec4 tessOuterHi Tessellation levels computed at each patch These tessellation levels can be computed the corresponding method in the tesselation control shader. Note that these functions potentially requires all bezier control points, you need to call barrier to ensure the conversion is done for all invocations. See osdglslPatchBSpline.glsl for more details. Uniform void OsdGetTessLevelsUniform ivec3 patchParam , out vec4 tessLevelOuter , out vec2 tessLevelInner , out vec4 tessOuterLo , out vec4 tessOuterHi void OsdGetTessLevelsUniformTriangle ivec3 patchParam , out vec4 tessLevelOuter , out vec2 tessLevelInner , out vec4 tessOuterLo , out vec4 tessOuterHi Screenspace void OsdEvalPatchBezierTessLevels OsdPerPatchVertexBezier cpBezier 16 , ivec3 patchParam , out vec4 tessLevelOuter , out vec2 tessLevelInner , out vec4 tessOuterLo , out vec4 tessOuterHi void OsdEvalPatchBezierTriangleTessLevels vec3 cv 15 , ivec3 patchParam , out vec4 tessLevelOuter , out vec2 tessLevelInner , out vec4 tessOuterLo , out vec4 tessOuterHi Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"osd_shader_interface.html\"\n }, {\n- \"title\": \"Porting Guide 2.x to 3.0\",\n- \"text\": \"Porting Guide 2.x to 3.0 Porting Guide 2.x to 3.0 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Porting Guide 2.x to 3.0 Source Code Organization Hbr Layer Translation Subdivision Schemes and Options in Sdc Specifying Face Varying Topology and Options Far Layer Translation Ordering of Refined Vertices Osd Layer Translation Controller Objects ComputeContext, DrawContext EvalLimitContext OsdMesh OsdKernelBatch OsdVertex Feature Adaptive Shader Changes End Cap Strategies Changes to Subdivision Build Support for Combining 2.x and 3.0 Porting Guide 2.x to 3.0 This document is a high-level description of how to port exiting OpenSubdiv 2.x code to use OpenSubdiv 3.0. NOTE If your questions are not answered here, please contact us on the OpenSubdiv forum and we will be happy to help Source Code Organization Given the scale of functional changes that were being made to the public interface, we took the opportunity in 3.0 to update the coding style and organization -- most notably making use of namespaces for each library. Subdirectory Namespace Relevance hbr NA Historical, no longer used sdc Sdc New, low-level, public options, constants, etc. vtr Vtr New, internal use, topology representation far Far Revised, similar functionality with new API osd Osd Revised, similar functionality with new API Hbr Layer Translation Client mesh topology is now translated into an instance of FarTopologyRefiner instead of HbrMesh. OpenSubdiv 2.x OpenSubdiv 3.0 HbrMeshVTX_TYPE FarTopologyRefiner The FarTopologyRefiner is now the core representation of topology from which all other major classes in Far and Osd are constructed. It was designed to support efficient refinement uniform or sparse of a base mesh of arbitrary topology no manifold restrictions. Once constructed it can be directly refined to meet some need, or passed to other contexts that will refine it to meet their needs. In contrast to directly assembling an HbrMesh, the TopologyRefiner, like other classes in Far, requires a Factory class for its construction. One of the early goals of these factories was to allow a client to convert their existing boundary representation -- with its full topological traversal abilities -- directly into the TopologyRefiners representation. While this is now possible, this also represents the most complex construction process and is only recommended for usage where this conversion process is critical. Details on how to construct a TopologyRefiner can be found in the Far overview documentation. Additionally, documentation for FarTopologyRefinerFactoryMESH outlines the requirements, and Far tutorial 3.1 tutorialsfartutorial_3_1 provides an example of a factory for directly converting HbrMeshes to TopologyRefiners. Its worth a reminder here that FarTopologyRefiner contains only topological information which does include sharpness, since that is considered relating to subdivision topology and not the positions or other data associated with a mesh. While HbrMeshT required some definition of a vertex type T and dimensions of face-varying data, TopologyRefiner is more clearly separated from the data. So the construction of the TopologyRefiner does not involve data specification at all. Subdivision Schemes and Options in Sdc The creation of a new TopologyRefiner requires specification of a subdivision scheme and a set of options that are applicable to all schemes. With HbrMesh, the scheme was specified by declaring a static instance of a specific subclass of a subdivision object, and the options were specified with a number of methods on the different classes. Such general information about the schemes has now been encapsulated in the Sdc layer for use throughout OpenSubdiv. The subdivision scheme is now a simple enumerated type SdcSchemeType and the entire set of options that can be applied to a scheme is encapsulated in a single simple struct of flags and enumerated types SdcOptions. OpenSubdiv 2.x OpenSubdiv 3.0 HbrMeshTSetInterpolateBoundaryMethod SdcOptionsSetVtxBoundaryInterpolation HbrMeshTSetFVarInterpolateBoundaryMethod SdcOptionsSetFVarLinearInterpolation HbrSubdivisionTSetCreaseSubdivisionMethod SdcOptionsSetCreasingMethod Specifying Face Varying Topology and Options Both the way in which face varying data is associated with a mesh and the options used to control its interpolation have changed. The documentation on Compatibility with OpenSubdiv 2.x details the equivalence of interpolation options between Hbr and the new SdcOptionsFVarLinearInterpolation enum, while the section on Face Varying Interpolation illustrates their effects. Face varying data is now specified by index rather than by value, or as often stated, it is specified topologically. Just as vertices for faces are specified by indices into a potential buffer of positions, face varying values are specified by indices into a potential buffer of values. Both vertices and face varying values frequently referred to as FVarValues in the API are assigned and associated with the corners of all faces. In many cases this will simplify representation as many common geometry container formats such as Obj or Alembic specify texture coordinates the same way. For other cases, where a value per face-corner is provided with no indication of which values incident each vertex should be considered shared, it will be necessary to determine shared indices for values at each vertex if any non-linear interpolation is desired. Far Layer Translation While TopologyRefiner was introduced into Far as the new intermediate topology representation, several other changes were made to classes in Far to provide more modular building blocks for use by the Osd layer or directly. OpenSubdiv 2.x OpenSubdiv 3.0 FarMeshU NA, no longer needed FarSubdivisionTables FarStencilTable FarPatchTables FarPatchTable Ordering of Refined Vertices The FarMesh was previously responsible for refining an HbrMesh -- generating new vertices and faces in successive levels of refinement in the FarSubdivisionTables. Vertices were grouped and reordered from the native ordering of HbrMesh so that vertices requiring similar processing were consecutive. Such grouping alleviated most of the idiosyncrasies of HbrMeshs native ordering but not all. FarToplogyRefiner is inherently a collection of refinement levels, and within each refined level so excluding the base level, all components are still grouped for the same reasons. There are two issues here though the ordering of these groups has changed though an option exists to preserve it the ordering of components within these groups is not guaranteed to have been preserved Vertices in a refined level are grouped according to the type of component in the parent level from which they originated, i.e. some vertices originate from the center of a face face-vertices, some from an edge edge-vertices and some from a vertex vertex-vertices. Note that there is a conflict in terminology here -- face-vertices and edge-vertices most often refer to vertices incident a face or edge -- but for the sake of this discussion, we use them to refer to the component from which a child vertex originates. The following table shows the ordering of these groups in 2.x and the two choices available in 3.0. The option is the orderVerticesFromFacesFirst flag that can be set in the Option structs passed to the uniform and adaptive refinement methods of TopologyRefiner Version and option Vertex group ordering 2.x face-vertices, edge-vertices, vertex-vertices 3.0 default vertex-vertices, face-vertices, edge-vertices 3.0 orderVerticesFromFacesFirst true face-vertices, edge-vertices, vertex-vertices The decision to change the default ordering was based on common feedback the rationale was to allow a trivial mapping from vertices in the cage to their descendants at all refinement levels. While the grouping is fundamental to the refinement process, the ordering of the groups is internally flexible, and the full set of possible orderings can be made publicly available in future if there is demand for such flexibility. The ordering of vertices within these groups was never clearly defined given the way that HbrMesh applied its refinement. For example, for the face-vertices in a level, it was never clear which face-vertices would be first as it depended on the order in which HbrMesh traversed the parent faces and generated them. Given one face, HbrMesh would often visit neighboring faces first before moving to the next intended face. The ordering with FarTopologyRefiner is much clearer and predictable. Using the face-vertices as an example, the order of the face-vertices in level N1 is identical to the order of the parent faces in level N from which they originated. So if we have face-vertices Vi , Vj and Vk at some level, originating from faces Fi , Fj and Fk in the previous level, they will be ordered in increasing order of i , j and k . For uniform refinement the ordering of face vertices Vi will therefore exactly match the ordering of the parent faces Fi . For adaptive or otherwise sparse refinement, the subset of Vi will be ordered similarly, just with components missing from those not refined. The same is true of all vertices, i.e. edge-vertices and vertex-vertices, and also for other components in refined levels, i.e. the child faces and edges. For child faces and edges, more than one will originate from the same parent face or edge. In addition to the overall ordering based on the parent faces or edges, another ordering is imposed on multiple children originating from the same face or edge. They will be ordered based on the corner or end-vertex with which they are associated. In the case of refined faces, another way to view the ordering is to consider the way that faces are originally defined -- by specifying the set of vertices for the corners of each face, often aggregated into a single large array. The ordering of the set of refined faces for each level will correspond directly to such an array of vertices per face in the previous level. Osd Layer Translation Two big changes in the 3.0 API have allowed the Osd layer to be significantly simpler, the first is the move to stencil tables from subdivision tables and the second is shader simplification. With this refactoring, the focus has been to use more meaningful names and to make the data contained within an object more apparent. Controller Objects The API-specific ComputeController has been replaced with the Evaluator . It reflects the fact that stencil compute batches are significantly simpler than subdivision table compute batches. The name Evaluator was chosen with the hope that is more meaningful than the generic ComputeController moniker the Evaluator evaluates stencil and patch tables. In the 2.x code base, subdiv level buffers were always required to be allocated contiguously, however in 3.0 with the shift to stencil tables, this strict allocation scheme is no longer required. As a result, the EvalStencils and EvalPatches methods now accept both a source and a destination descriptor. OpenSubdiv 2.x OpenSubdiv 3.0 ComputeControllerRefine Osd...EvaluatorEvalStencils ComputeControllerSynchronize Osd...EvaluatorSynchronize EvalStencilsControllerUpdateValues Osd...EvaluatorEvalStencils EvalStencilsControllerUpdateDerivs Osd...EvaluatorEvalStencils EvalLimitControllerEvalLimitSample Osd...EvaluatorEvalPatches Also note that OsdVertexDescriptor has been renamed, however its data members and semantic purpose remains the same OpenSubdiv 2.x OpenSubdiv 3.0 OsdVertexBufferDescriptor OsdBufferDescriptor ComputeContext, DrawContext ComputeContext and DrawContext have been replaced with API-specific StencilTable and PatchTable objects, for example OsdGLStencilTableSSBO. OpenSubdiv 2.x OpenSubdiv 3.0 ComputeContext Osd...StencilTable e.g. GLStencilTableTBO EvalStencilsContext Osd...StencilTable DrawContext Osd...PatchTable e.g. GLPatchTable EvalLimitContext The data stored in EvalLimitContext has been merged into the Evaluator class as well. EvalCoords have been moved into their own type, OsdPatchCoords. The primary change here is that the PTex face ID is no longer part of the data structure, rather the client can use a FarPatchMap to convert from PTex face ID to a FarPatchTablePatchHandle. OpenSubdiv 2.x OpenSubdiv 3.0 EvalLimitContext PatchTable EvalLimitContextEvalCoords OsdPatchCoords types.h OsdMesh While not strictly required, OsdMesh is still supported in 3.0 as convenience API for allocating buffers. OsdMesh serves as a simple way to allocate all required data, in the location required by the API for example, GPU buffers for OpenGL. OsdKernelBatch No translation, it is no longer part of the API. OsdVertex No translation, it is no longer part of the API. Feature Adaptive Shader Changes In 3.0, the feature adaptive screen-space tessellation shaders have been dramatically simplified, and the client-facing API has changed dramatically as well. The primary shift is to reduce the total number of shader combinations, and as a result, some of the complexity management mechanisms are no longer necessary. In the discussion below, some key changes are highlighted, but deep integrations may require additional discussion please feel free to send follow up questions to the OpenSubdiv google group. The number of feature adaptive shaders has been reduced from N to exactly 1 or 2, depending on how end-caps are handled. Osd layer no longer compiles shaders, rather it returns shader source for the client to compile. This source is obtained via OsdGLSLHLSLPatchShaderSource. The API exposed in shaders to access patch-based data has been consolidated and formalized, see osdglslPatchCommon.glsl and osdhlslPatchCommon.hlsl for details. Patches are no longer rotated and transition patches have been eliminated, simplifying PatchDescriptor to a 4 bits. Additionally, FarPatchTablesDescriptor has been moved into its own class in the Far namespace. The following table outlines the API translation between 2.x and 3.0 OpenSubdiv 2.x OpenSubdiv 3.0 OsdDrawContextPatchDescriptor NA, no longer needed. OsdDrawContextPatchArray OSdPatchArray types.h FarPatchTablesPatchDescriptor FarPatchDescriptor patchDescriptor.h FarPatchTablesPatchArray made private. End Cap Strategies By default, OpenSubdiv uses Gregory patches to approximate the patches around extraordinary vertices at the maximum isolation level, this process is referred to as end-capping. If ENDCAP_BSPLINE_BASIS is specified to PatchTableFactoryOptions, BSpline patches are used, which gives less accuracy, but it makes possible to render an entire mesh in a single draw call. Both patches require additional control points that are not part of the mesh, we refer to these as local points. In 3.0, the local points of those patches are computed by applying a stencil table to refined vertices to construct a new stencil table for the local points. Since this new stencil table is topologically compatible with the primary stencil table for refinement, it is convenient and efficient to splice those stencil tables together. This splicing can be done in the following way FarStencilTable const refineStencils FarStencilTableFactoryCreatetopologyRefiner FarPatchTable cosnt patchTable FarPatchTableFactoryCreatetopologyRefiner FarStencilTable const localPointStencils patchTable-GetLocalPointStencilTable FarStencilTable const splicedStencils FarStencilTableFactoryAppendLocalPointStencilTablestopologyRefiner, refineStencils, localPointStencils NOTE Once the spliced stencil table is created, the refined stencils can be released, but the local point stencils are owned by patchTable, it should not be released. OpenSubdiv 3.0 also supports 2.x style Gregory patches, if ENDCAP_LEGACY_GREGORY is specified to PatchTableFactoryOptions. In this case, such an extra stencil splicing isnt needed, however clients must still bind additional buffers VertexValence buffer and QuadOffsets buffer. See OsdGLLegacyGregoryPatchTable for additional details. Changes to Subdivision The refactoring of OpenSubdiv 3.0 data representations presented a unique opportunity to revisit some corners of the subdivision specification and remove or update some legacy features -- none of which was taken lightly. More details are provided in Subdivision Compatibility , while the following offers a quick overview All face-varying interpolation options have been combined into a single enum. Vertex interpolation options have been renamed or removed The naming of the standard creasing method has changed from Normal to Uniform . Unused legacy modes of the smoothtriangle option have been removed. The averaging of Chaikin creasing with infinitely sharp edges has changed. Support for Hierarchical Edits has been removed. Build Support for Combining 2.x and 3.0 Running OpenSubdiv 2.0 and 3.0 in a single process is supported, however some special care must be taken to avoid namespace collisions, both in terms of run-time symbols avoid using OpenSubdivOsd, for example and in terms of build-time search paths. To support both OpenSubdiv 2.0 and 3.0 in your build environment, you can prefix the header install directory of OpenSubdiv 3.0. Do this using the build flag CMAKE_INCDIR_BASE when configuring cmake i.e. -DCMAKE_INCDIR_BASEincludeopensubdiv3 and then including files from opensubdiv3... in client code. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"References\",\n+ \"text\": \"References References 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE References References Analytic Displacement Mapping using Hardware Tessellation Matthias Niessner, Charles Loop ACM Transactions on Graphics, Vol. 32 No. 3 Article 26 June 2013 httpresearch.microsoft.comen-usumpeoplecloopTOG2013.pdf httpdoi.org10.11452487228.2487234 Feature Adaptive GPU Rendering of Catmull-Clark Subdivision Surfaces Matthias Niessner, Charles Loop, Mark Meyer, and Tony DeRose ACM Transactions on Graphics, Vol. 31 No. 1 Article 6 January 2012 httpresearch.microsoft.comen-usumpeopleclooptog2012.pdf httpdoi.org10.11452077341.2077347 Efficient Evaluation of Semi-Smooth Creases in Catmull-Clark Subdivision Surfaces Matthias Niessner, Charles Loop, and Guenter Greiner. Eurographics Proceedings, Cagliari, 2012 httpresearch.microsoft.comen-usumpeoplecloopEG2012.pdf httpdoi.org10.2312confEG2012short041-044 Approximating Subdivision Surfaces with Gregory Patches for Hardware Tessellation Charles Loop, Scott Schaefer, Tianyun Ni, Ignacio Castano SIGGRAPH Asia Conference Proceedings 2009 httpwww.dgp.toronto.edupeoplestamrealityResearchpdfsig98.pdf httpdoi.org10.11451661412.1618497 GPU Smoothing of Quad Meshes T. L. Ni, Y. Yeo, A. Myles, V. Goel and J. Peters Proc. IEEE SMI 2008 httpwww.cise.ufl.eduresearchSurfLabpaperssmi08.pdf httpdoi.org10.1109SMI.2008.4547938 Fast Parallel Construction of Smooth Surfaces from Meshes with TriQuadPent Facets A. Myles and T. Ni and J. Peters Eurographics Symposium on Geometry Processing 2008 httpswww.cise.ufl.eduresearchSurfLabpapers08poly.pdf httpdoi.org10.1111j.1467-8659.2008.01276.x Approximating Catmull-Clark Subdivision Surfaces with Bicubic Patches Charles Loop, Scott Schaefer ACM Transactions on Graphics, Vol. 27 No. 1 Article 8 March 2008 httpresearch.microsoft.comen-usumpeoplecloopacctog.pdf httpdoi.org10.11451330511.1330519 Rapid Evaluation of Catmull-Clark Subdivision Surfaces Jeffrey Bolz and Peter Schroder Web3D Proceedings 2002 httpwww.multires.caltech.edupubsfastsubd.pdf httpdoi.org10.1145504502.504505 Piecewise Smooth Subdivision Surfaces with Normal Control Henning Biermann, Adi Levin and Denis Zorin SIGGRAPH 2000 Conference Proceedings httpmrl.nyu.edudzorinpapersbiermann2000pss.pdf httpdoi.org10.1145344779.344841 Subdivision for Modeling and Animation Denis Zorin, Peter Schroder Course Notes of SIGGRAPH 1999 httpwww.multires.caltech.edupubssig99notes.pdf Exact Evaluation of Catmull-Clark Subdivision Surfaces at Arbitrary Parameter Values Jos Stam SIGGRAPH 98 Conference Proceedings, Annual Conference Series, July 1998 httpwww.dgp.toronto.edupeoplestamrealityResearchpdfsig98.pdf httpdoi.org10.1145280814.280945 Subdivision Surfaces in Character Animation Tony DeRose, Michael Kass, Tien Truong Proceedings of SIGGRAPH 1998 httpgraphics.pixar.comlibraryGeripaper.pdf httpdoi.org10.1145280814.280826 Efficient, Fair Interpolation Using Catmull-Clark Surfaces Mark Halstead, Michael Kass, Tony DeRose SIGGRAPH 93 Conference Proceedings httpgraphics.pixar.comlibraryFairSubdivisionpaper.pdf httpdoi.org10.1145166117.166121 Recursively generated B-spline surfaces on arbitrary topological meshes Catmull, E. Clark, J. Computer-Aided Design 10 6 1978 httpdoi.org10.10160010-448528782990110-0 Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"porting.html\"\n+ \"loc\": \"references.html\"\n+ }, {\n+ \"title\": \"OSD Overview\",\n+ \"text\": \"OSD Overview OSD Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE OpenSubdiv Osd Refinement Limit Stencil Evaluation Limit Evaluation with PatchTable OpenGLDX11Metal Drawing with Hardware Tessellation InterleavedBatched Buffer Configuration Cross-Platform Implementation OpenSubdiv Osd Osd contains device dependent code that makes Far structures available on various backends such as TBB, CUDA, OpenCL, GLSL, etc. The main roles of Osd are Refinement Compute stencil-based uniformadaptive subdivision on CPUGPU backends Limit Stencil Evaluation Compute limit surfaces by limit stencils on CPUGPU backends Limit Evaluation with PatchTable Compute limit surfaces by patch evaluation on CPUGPU backends OpenGLDX11Metal Drawing with hardware tessellation Provide GLSLHLSLMetal tessellation functions for patch table InterleavedBatched buffer configuration Provide consistent buffer descriptor to deal with arbitrary buffer layout. Cross-Platform Implementation Provide convenient classes to interop between compute and draw APIs These are independently used by clients. For example, a client can use only the limit stencil evaluation, or a client can refine subdivision surfaces and draw them with the PatchTable and Osd tessellation shaders. All device specific evaluation kernels are implemented in the Evaluator classes. Since Evaluators dont own vertex buffers, clients should provide their own buffers as a source and destination. There are some interop classes defined in Osd for convenience. OpenSubdiv utilizes a series of regression tests to compare and enforce identical results across different computational devices. Refinement Osd supports both uniform and feature adaptive subdivision. Once clients create a FarStencilTable for the topology, they can convert it into device-specific stencil tables if necessary. The following table shows which evaluator classes and stencil table interfaces can be used together. Note that while Osd provides these stencil table classes which can be easily constructed from FarStencilTable, clients arent required to use these table classes. Clients may have their own entities as a stencil tables as long as EvaluatorEvalStencils can access the necessary interfaces. Backend Evaluator class compatible stencil table CPU CPU single-threaded CpuEvaluator FarStencilTable TBB CPU multi-threaded TbbEvaluator FarStencilTable OpenMP CPU multi-threaded OmpEvaluator FarStencilTable CUDA GPU CudaEvaluator CudaStencilTable OpenCL CPUGPU CLEvaluator CLStencilTable GL ComputeShader GPU GLComputeEvaluator GLStencilTableSSBO GL Transform Feedback GPU GLXFBEvaluator GLStencilTableTBO DX11 ComputeShader GPU D3D11ComputeEvaluator D3D11StencilTable Metal GPU MTLComputeEvaluator MTLStencilTable Limit Stencil Evaluation Limit stencil evaluation is quite similar to refinement in Osd . At first clients create FarLimitStencilTable for the locations to evaluate the limit surfaces, then convert it into an evaluator compatible stencil table and call EvaluatorEvalStencils. Limit Evaluation with PatchTable Another way to evaluate the limit surfaces is to use the PatchTable. Once all control vertices and local points are resolved by the stencil evaluation, Osd can evaluate the limit surfaces through the PatchTable. Backend Evaluator class compatible patch table CPU CPU single-threaded CpuEvaluator CpuPatchTable TBB CPU multi-threaded TbbEvaluator CpuPatchTable OpenMP CPU multi-threaded OmpEvaluator CpuPatchTable CUDA GPU CudaEvaluator CudaPatchTable OpenCL CPUGPU CLEvaluator CLPatchTable GL ComputeShader GPU GLComputeEvaluator GLPatchTable GL Transform Feedback GPU GLXFBEvaluator GLPatchTable DX11 ComputeShader GPU D3D11ComputeEvaluator not yet supported D3D11PatchTable Metal ComputeShader GPU MTLComputeEvaluator MTLPatchTable Release Notes 3.x Osd evaluation backends EvaluatorEvalPatches do not support evaluation of single-crease or Legacy Gregory patch types. OpenGLDX11Metal Drawing with Hardware Tessellation One of the most interesting use cases of the Osd layer is realtime drawing of subdivision surfaces using hardware tessellation. This is somewhat similar to limit evaluation with PatchTable described above. Drawing differs from limit evaluation in that Osd provides shader snippets for patch evaluation and clients will inject them into their own shader source. See shader interface for a more detailed discussion of the shader interface. InterleavedBatched Buffer Configuration All Osd layer APIs assume that each primitive variables to be computed points, colors, uvs ... are contiguous arrays of 32bit floating point values. The Osd API refers to such an array as a buffer. A buffer can exist on CPU memory or GPU memory. Osd Evaluators typically take one source buffer and one destination buffer, or three destination buffers if derivatives are being computed. Osd Evaluators also take BufferDescriptors, that are used to specify the layout of the source and destination buffers. A BufferDescriptor is a struct of 3 integers which specify an offset, length and stride. For example Vertex 0 Vertex 1 ... X Y Z X Y Z ... The layout of this buffer can be described as Osd BufferDescriptor desc offset 0 , length 3 , stride 3 BufferDescriptor can be used for an interleaved buffer too. Vertex 0 Vertex 1 ... X Y Z R G B A X Y Z R G B A ... Osd BufferDescriptor xyzDesc 0 , 3 , 7 Osd BufferDescriptor rgbaDesc 3 , 4 , 7 Although the source and destination buffers dont need to be the same buffer for EvalStencils, adaptive patch tables are constructed to first index the coarse vertices and the refined vertices immediately afterward. In this case, the BufferDescriptor for the destination should include the offset as the number of coarse vertices to be skipped. Coarse vertices n Src Refined vertices Dst Vertex 0 Vertex 1 ... Vertex n Vertex n1 X Y Z X Y Z ... X Y Z X Y Z ... Osd BufferDescriptor srcDesc 0 , 3 , 3 Osd BufferDescriptor dstDesc n 3 , 3 , 3 Also note that the source descriptor doesnt have to start with offset 0. This is useful when a client has a big buffer with multiple objects batched together. Cross-Platform Implementation One of the key goals of OpenSubdiv is to achieve as much cross-platform flexibility as possible and leverage all optimized hardware paths where available. This can be very challenging as there is a very large variety of plaftorms and APIs available, with very distinct capabilities. In Osd , Evaluators dont care about interops between those APIs. All Evaluators have two kinds of APIs for both EvalStencils and EvalPatches. Explicit signatures which directly take device-specific buffer representation e.g., pointer for CpuEvaluator, GLuint buffer for GLComputeEvaluator, etc. Generic signatures which take arbitrary buffer classes. The buffer class is required to have a certain method to return the device-specific buffer representation. The later interface is useful if the client supports multiple backends at the same time. The methods that need to be implemented for the Evaluators are Evaluator class object method CpuEvaluator TbbEvaluator OmpEvaluator pointer to cpu memory BindCpuBuffer CudaEvaluator pointer to cuda memory BindCudaBuffer CLEvaluator cl_mem BindCLBuffer GLComputeEvaluator GLXFBEvaluator GL buffer object BindVBO D3D11ComputeEvaluator D3D11 UAV BindD3D11UAV MTLComputeEvaluator MTLBuffer BindMTLBuffer The buffers can use these methods as a trigger of interop. Osd provides a default implementation of interop buffer for most of the backend combinations. For example, if the client wants to use CUDA as a computation backend and use OpenGL as the drawing API, OsdCudaGLVertexBuffer fits the case since it implements BindCudaBuffer and BindVBO. Again, clients can implement their own buffer class and pass it to the Evaluators. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"tags\": \"\",\n+ \"loc\": \"osd_overview.html\"\n }, {\n \"title\": \"mtlPtexViewer\",\n- \"text\": \"mtlPtexViewer mtlPtexViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION KEYBOARD CONTROLS SEE ALSO SYNOPSIS open mtlPtexViewer.app --args -yup -u -a -l isolation level ptex color file ptex displacement file DESCRIPTION mtlPtexViewer is a stand-alone application demonstrating shading with color and displacement ptex maps. Multiple controls are available to experiment with the algorithms. KEYBOARD CONTROLS q quit f fit frame - increase decrease tessellation rate SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"mtlPtexViewer mtlPtexViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION KEYBOARD CONTROLS SEE ALSO SYNOPSIS open mtlPtexViewer.app --args -yup -u -a -l isolation level ptex color file ptex displacement file DESCRIPTION mtlPtexViewer is a stand-alone application demonstrating shading with color and displacement ptex maps. Multiple controls are available to experiment with the algorithms. KEYBOARD CONTROLS q quit f fit frame - increase decrease tessellation rate SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"mtlptexviewer.html\"\n }, {\n- \"title\": \"Modeling Tips\",\n- \"text\": \"Modeling Tips Modeling Tips 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Topology Use Fewer Spans Avoid High Valence vertices Edge-Loop Transitions Practical Topology Primer Triangles and N-Gons Semi-Sharp Creases Use crease sets Additional Resources The following sections describe common techniques specific to modeling with subdivision surfaces. Note The following information contains techniques specific to the Catmull-Clark subdivision scheme. Some elements need to be adjusted for Loop surfaces. Topology Well-constructed subdivision meshes have several important properties They consist primarily of regular faces quads for Catmull-Clark, tris for Loop They contain few extraordinary vertices They efficiently describe the intended shape They are topologically manifold Use Fewer Spans While polygon models need to use a large number of spans to approximate smooth curved surfaces, subdivision models require significantly fewer control points. In most situations, 6 spans are enough to create accurate circular shapes, and 4 is often enough to approximate background objects. Avoid High Valence vertices A high valence vertex is a vertex connected to more than 4 adjacent edges. High valence vertices cause several problems when subdivided The Catmull-Clark scheme can produce wavy surfaces when a revolution vertex is surrounded by triangles see here High valence vertices incur fairly large performance hits Currently, OpenSubdiv has a hard constraint imposed by GPU shaders on the maximum valence of a vertex 27 on current hardware Instead, here are some topological strategies to cap revolution shapes Note that all these cylinders use only quad faces, and all the vertices in the caps have a valence of 4 except the bottom left example Edge-Loop Transitions It is often necessary to vary the density of control vertices over the surface mesh areas around the fingers of a hand require more CVs than the comparatively simpler region around the palm. It is important to handle the topology around these transitions efficiently. One strategy is to use extraordinary vertices, such as this example, using a valence 5 vertex to expand 3 edge loops into 5. Practical Topology Primer Some real-world examples showing how to produce detailed shapes with sparse topology, few extraordinary vertices, and no high-valence revolution poles. Triangles and N-Gons Used sparsely, non-quads can be very useful to gather 3 or more diverging edge-loops. These are often encountered in highly deforming areas with curvature saddle-points ex arm-torso connection. The strategic placement of a pentagon in one of these critical spots ensures that the surface remains smooth, while allowing for complex topology to flow around. Semi-Sharp Creases Semi-sharp creases can be a very powerful tool for hard-surface modeling. Both edges and vertices can be tagged with a sharpness value. Crease sharpness values range from 0 smooth to 10 infinitely sharp It is generally cheaper to use creases whenever possible, instead of adding extra edgesedge-loops. However... Creases introduce extra computation costs that are proportional to the sharpness value. So... Sharpness values above 5 should rarely be needed. The following sections introduce some techniques to best leverage them. Use crease sets Complex hard-surface models giant robots, vehicles, buildings... are likely to tag large number of edges it is extremely useful to organize these edgesedge loops into logical sets with descriptive names. Edges or vertices in a crease set group all share the same sharpness value. If you are modeling with Maya, the CreaseSetEditor implements this type of workflow. Additionally, for debugging purposes, it is often very helpful if the name of a set contains the sharpness value ex topDeck_2. Besides authoring convenience, one of the benefits of having many edge-loops share identical sharpness values is that it enables very powerful performance optimizations within the feature adaptive algorithm faster renders less memory. Additional Resources An excellent short tutorial from the Guerrilla CG Project that illustrates many of the common pitfalls of subdivision modeling, and the strategies to overcome them Ivo Kos, Modelling Technical Director at Pixar Animation Studios, shows some of the modeling techniques he uses when modeling props and architecture sets for feature films. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"mtlViewer\",\n+ \"text\": \"mtlViewer mtlViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION KEYBOARD CONTROLS SEE ALSO SYNOPSIS open mtlViewer.app --args -yup -u -a -l refinement level objfiles -catmark -loop -bilinear DESCRIPTION mtlViewer is a stand-alone application that showcases the application of uniform and feature adaptive subdivision schemes to a collection of geometric shapes. Multiple controls are available to experiment with the algorithms. mtlViewer running on macOS mtlViewer running on iOS KEYBOARD CONTROLS q quit f fit frame - increase decrease tessellation rate SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"mod_notes.html\"\n+ \"loc\": \"mtlviewer.html\"\n }, {\n- \"title\": \"License\",\n- \"text\": \"License License 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Open Source License OpenSubdiv is covered by a modified Apache 2.0 license included below, and is free to use for commercial or non-commercial use. All Pixar patents in the area of subdivision surface algorithms have also been released for public use. We welcome any involvement in the development or extension of this code in fact, we would love it. Please contact us if you are interested. For information on Apache licenses, see httpwww.apache.orglicenses The following license text describes the open source policy adopted by Pixar and is included in every source file. Copyright 2013 Pixar Licensed under the Apache License, Version 2.0 the Apache License with the following modification you may not use this file except in compliance with the Apache License and the following modification to it Section 6. Trademarks. is deleted and replaced with 6. Trademarks. This License does not grant permission to use the trade names, trademarks, service marks, or product names of the Licensor and its affiliates, except as required to comply with Section 4c of the License and to reproduce the content of the NOTICE file. You may obtain a copy of the Apache License at httpwww.apache.orglicensesLICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the Apache License with the above modification is distributed on an AS IS BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the Apache License for the specific language governing permissions and limitations under the Apache License. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"Modeling Tips\",\n+ \"text\": \"Modeling Tips Modeling Tips 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Topology Use Fewer Spans Avoid High Valence vertices Edge-Loop Transitions Practical Topology Primer Triangles and N-Gons Semi-Sharp Creases Use crease sets Additional Resources The following sections describe common techniques specific to modeling with subdivision surfaces. Note The following information contains techniques specific to the Catmull-Clark subdivision scheme. Some elements need to be adjusted for Loop surfaces. Topology Well-constructed subdivision meshes have several important properties They consist primarily of regular faces quads for Catmull-Clark, tris for Loop They contain few extraordinary vertices They efficiently describe the intended shape They are topologically manifold Use Fewer Spans While polygon models need to use a large number of spans to approximate smooth curved surfaces, subdivision models require significantly fewer control points. In most situations, 6 spans are enough to create accurate circular shapes, and 4 is often enough to approximate background objects. Avoid High Valence vertices A high valence vertex is a vertex connected to more than 4 adjacent edges. High valence vertices cause several problems when subdivided The Catmull-Clark scheme can produce wavy surfaces when a revolution vertex is surrounded by triangles see here High valence vertices incur fairly large performance hits Currently, OpenSubdiv has a hard constraint imposed by GPU shaders on the maximum valence of a vertex 27 on current hardware Instead, here are some topological strategies to cap revolution shapes Note that all these cylinders use only quad faces, and all the vertices in the caps have a valence of 4 except the bottom left example Edge-Loop Transitions It is often necessary to vary the density of control vertices over the surface mesh areas around the fingers of a hand require more CVs than the comparatively simpler region around the palm. It is important to handle the topology around these transitions efficiently. One strategy is to use extraordinary vertices, such as this example, using a valence 5 vertex to expand 3 edge loops into 5. Practical Topology Primer Some real-world examples showing how to produce detailed shapes with sparse topology, few extraordinary vertices, and no high-valence revolution poles. Triangles and N-Gons Used sparsely, non-quads can be very useful to gather 3 or more diverging edge-loops. These are often encountered in highly deforming areas with curvature saddle-points ex arm-torso connection. The strategic placement of a pentagon in one of these critical spots ensures that the surface remains smooth, while allowing for complex topology to flow around. Semi-Sharp Creases Semi-sharp creases can be a very powerful tool for hard-surface modeling. Both edges and vertices can be tagged with a sharpness value. Crease sharpness values range from 0 smooth to 10 infinitely sharp It is generally cheaper to use creases whenever possible, instead of adding extra edgesedge-loops. However... Creases introduce extra computation costs that are proportional to the sharpness value. So... Sharpness values above 5 should rarely be needed. The following sections introduce some techniques to best leverage them. Use crease sets Complex hard-surface models giant robots, vehicles, buildings... are likely to tag large number of edges it is extremely useful to organize these edgesedge loops into logical sets with descriptive names. Edges or vertices in a crease set group all share the same sharpness value. If you are modeling with Maya, the CreaseSetEditor implements this type of workflow. Additionally, for debugging purposes, it is often very helpful if the name of a set contains the sharpness value ex topDeck_2. Besides authoring convenience, one of the benefits of having many edge-loops share identical sharpness values is that it enables very powerful performance optimizations within the feature adaptive algorithm faster renders less memory. Additional Resources An excellent short tutorial from the Guerrilla CG Project that illustrates many of the common pitfalls of subdivision modeling, and the strategies to overcome them Ivo Kos, Modelling Technical Director at Pixar Animation Studios, shows some of the modeling techniques he uses when modeling props and architecture sets for feature films. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"license.html\"\n+ \"loc\": \"mod_notes.html\"\n }, {\n \"title\": \"HBR Overview\",\n- \"text\": \"HBR Overview HBR Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Hierarchical Boundary Representation Hbr Half-edge Data Structure Half-edge cycles and Manifold Topology Templated Vertex Class Boundary Interpolation Rules Hierarchical Boundary Representation Hbr Hbr is an interconnected topological data representation. The high level of vertex connectivity information makes this representation well suited for creation and editing purposes. It is however inefficient for interactive refinement operations Separate objects are allocated for each vertex and edge with pointers to neighboring vertices and edges. Hbr is also the lowest-level subdivision library in Pixars Photorealistic RenderMan . Note As of OpenSubdiv 3.0, all Hbr dependencies have been removed from the core APIs Sdc , Vtr , Far , Osd . The legacy source code of Hbr is provided purely for regression and legacy purposes. If your code is currently depending on Hbr functionality, we recommend migrating to the newer APIs as we cannot guarantee that this code will be maintained in future releases. For more information see the 3.0 release notes Half-edge Data Structure The current implementation is based on a half-edge data structure. Half-edge cycles and Manifold Topology Because half-edges only carry a reference to their opposite half-edge, a given edge can only access a single neighboring edge cycle. This is a fundamental limitation of the half-edge data structure, in that it cannot represent non-manifold geometry, in particular fan-type topologies. A different approach to topology will probably be necessary in order to accommodate non-manifold geometry. Templated Vertex Class The vertex class has been abstracted into a set of templated function accesses. Providing Hbr with a template vertex class that does not implement these functions allows client-code to use Hbr as a pure topological analysis tool without having to pay any costs for data interpolation. It also allows client-code to remain in complete control of the layout of the vertex data interleaved or non-interleaved. Boundary Interpolation Rules Hbr recognizes 4 rule-sets of boundary interpolation Interpolation Rule-Sets k_InterpolateBoundaryNone k_InterpolateBoundaryEdgeOnly k_InterpolateBoundaryEdgeAndCorner k_InterpolateBoundaryAlwaysSharp This enum is shared for both vertex and face-varying interpolation, with the following distinctions vertex boundaries the BoundaryNone rule skips all boundary vertices results are undefined the AlwaysSharp rule does not apply face-varying boundaries the BoundaryNone rule selects bilinear face-varying interpolation Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"HBR Overview HBR Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Hierarchical Boundary Representation Hbr Half-edge Data Structure Half-edge cycles and Manifold Topology Templated Vertex Class Boundary Interpolation Rules Hierarchical Boundary Representation Hbr Hbr is an interconnected topological data representation. The high level of vertex connectivity information makes this representation well suited for creation and editing purposes. It is however inefficient for interactive refinement operations Separate objects are allocated for each vertex and edge with pointers to neighboring vertices and edges. Hbr is also the lowest-level subdivision library in Pixars Photorealistic RenderMan . Note As of OpenSubdiv 3.0, all Hbr dependencies have been removed from the core APIs Sdc , Vtr , Far , Osd . The legacy source code of Hbr is provided purely for regression and legacy purposes. If your code is currently depending on Hbr functionality, we recommend migrating to the newer APIs as we cannot guarantee that this code will be maintained in future releases. For more information see the 3.0 release notes Half-edge Data Structure The current implementation is based on a half-edge data structure. Half-edge cycles and Manifold Topology Because half-edges only carry a reference to their opposite half-edge, a given edge can only access a single neighboring edge cycle. This is a fundamental limitation of the half-edge data structure, in that it cannot represent non-manifold geometry, in particular fan-type topologies. A different approach to topology will probably be necessary in order to accommodate non-manifold geometry. Templated Vertex Class The vertex class has been abstracted into a set of templated function accesses. Providing Hbr with a template vertex class that does not implement these functions allows client-code to use Hbr as a pure topological analysis tool without having to pay any costs for data interpolation. It also allows client-code to remain in complete control of the layout of the vertex data interleaved or non-interleaved. Boundary Interpolation Rules Hbr recognizes 4 rule-sets of boundary interpolation Interpolation Rule-Sets k_InterpolateBoundaryNone k_InterpolateBoundaryEdgeOnly k_InterpolateBoundaryEdgeAndCorner k_InterpolateBoundaryAlwaysSharp This enum is shared for both vertex and face-varying interpolation, with the following distinctions vertex boundaries the BoundaryNone rule skips all boundary vertices results are undefined the AlwaysSharp rule does not apply face-varying boundaries the BoundaryNone rule selects bilinear face-varying interpolation Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"hbr_overview.html\"\n }, {\n- \"title\": \"Introduction\",\n- \"text\": \"Introduction Introduction 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Introduction Why Fast Subdivision Research Heritage Licensing Contributing External Resources Introduction OpenSubdiv is a set of open source libraries that implement high performance subdivision surface subdiv evaluation on massively parallel CPU and GPU architectures. This code path is optimized for drawing deforming surfaces with static topology at interactive framerates. OpenSubdiv is an API ready to be integrated into 3rd party digital content creation tools. It is not an application, nor a tool that can be used directly to create digital assets. Why Fast Subdivision Subdivision surfaces are commonly used for final rendering of character shapes for a smooth and controllable limit surfaces. However, subdivision surfaces in interactive apps are typically drawn as their polygonal control hulls because of performance. The polygonal control hull is an approximation that is offset from the true limit surface. Looking at an approximation in the interactive app makes it difficult to see exact contact, like fingers touching a potion bottle or hands touching a cheek. It also makes it difficult to see poke-throughs in cloth simulation if the skin and cloth are both approximations. This problem is particularly bad when one character is much larger than another and unequal subdiv face sizes cause approximation errors to be magnified. Maya and Pixars proprietary Presto animation system can take 100ms to subdivide a character of 30,000 polygons to the second level of subdivision 500,000 polygons. Being able to perform the same operation in less than 3ms allows the user to interact with the smooth, accurate limit surface at all times. Research The new GPU technology behind OpenSubdiv is the result of a joint research effort between Pixar and Microsoft. Feature Adaptive GPU Rendering of Catmull-Clark Subdivision Surfaces Matthias Niessner, Charles Loop, Mark Meyer, and Tony DeRose ACM Transactions on Graphics, Vol. 31 No. 1 Article 6 January 2012 httpgraphics.pixar.comlibraryGPUSubdivRenderingApaper.pdf Efficient Evaluation of Semi-Smooth Creases in Catmull-Clark Subdivision Surfaces Matthias Niessner, Charles Loop, and Guenter Greiner. Eurographics Proceedings, Cagliari, 2012 httpsniessnerlab.orgpapers20121semismoothniessner2012efficient.pdf Analytic Displacement Mapping using Hardware Tessellation Matthias Niessner, Charles Loop ACM Transactions on Graphics, Vol. 32 No. 3 Article 26 June 2013 httpsniessnerlab.orgpapers20133analyticniessner2013analytic.pdf Heritage This is the fifth-generation subdiv library in use by Pixars proprietary animation system in a lineage that started with code written by Tony DeRose and Tien Truong for Geris Game in 1996. Each generation has been a from-scratch rewrite that has built upon our experience using subdivision surfaces to make animated films. This code is live, so Pixars changes to OpenSubdiv for current and future films will be released as open source at the same time they are rolled out to Pixar animation production. Subdivision for Modeling and Animation Denis Zorin, Peter Schroder Course Notes of SIGGRAPH 1999 httpwww.multires.caltech.edupubssig99notes.pdf Subdivision Surfaces in Character Animation Tony DeRose, Michael Kass, Tien Truong Proceedings of SIGGRAPH 1998 httpgraphics.pixar.comlibraryGeripaper.pdf Recursively generated B-spline surfaces on arbitrary topological meshes Catmull, E. Clark, J. Computer-Aided Design 10 6 1978 Licensing OpenSubdiv is covered by the Apache License, and is free to use for commercial or non-commercial use. This is the same code that Pixar uses internally for animated film production. Our intent is to encourage a geometry standard for subdivision surfaces, by providing consistent i.e. yielding the same limit surface, high performance implementations on a variety of platforms. Why Apache We were looking for a commercial-friendly license that would convey our patents to the end users. This quickly narrowed the field to Microsoft Public License or Apache. Initially we chose MSPL because it handled trademarks better. But at the request of several companies we gave Apache another look, and decided to go with Apache with a very slight modification that simply says you cannot use any contributors trademarks. In other words, you can use OpenSubdiv to make a product, but you cannot use a Luxo Lamp or other character, etc. when marketing your product. License Header Contributing For details on how to contribute to OpenSubdiv, see the page on Contributing External Resources Pixar Research Pixar RD Portal Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n- \"tags\": \"\",\n- \"loc\": \"intro.html\"\n- }, {\n- \"title\": \"mtlViewer\",\n- \"text\": \"mtlViewer mtlViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION KEYBOARD CONTROLS SEE ALSO SYNOPSIS open mtlViewer.app --args -yup -u -a -l refinement level objfiles -catmark -loop -bilinear DESCRIPTION mtlViewer is a stand-alone application that showcases the application of uniform and feature adaptive subdivision schemes to a collection of geometric shapes. Multiple controls are available to experiment with the algorithms. mtlViewer running on macOS mtlViewer running on iOS KEYBOARD CONTROLS q quit f fit frame - increase decrease tessellation rate SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"License\",\n+ \"text\": \"License License 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Open Source License OpenSubdiv is covered by a modified Apache 2.0 license included below, and is free to use for commercial or non-commercial use. All Pixar patents in the area of subdivision surface algorithms have also been released for public use. We welcome any involvement in the development or extension of this code in fact, we would love it. Please contact us if you are interested. For information on Apache licenses, see httpwww.apache.orglicenses The following license text describes the open source policy adopted by Pixar and is included in every source file. Copyright 2013 Pixar Licensed under the Apache License, Version 2.0 the Apache License with the following modification you may not use this file except in compliance with the Apache License and the following modification to it Section 6. Trademarks. is deleted and replaced with 6. Trademarks. This License does not grant permission to use the trade names, trademarks, service marks, or product names of the Licensor and its affiliates, except as required to comply with Section 4c of the License and to reproduce the content of the NOTICE file. You may obtain a copy of the Apache License at httpwww.apache.orglicensesLICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the Apache License with the above modification is distributed on an AS IS BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the Apache License for the specific language governing permissions and limitations under the Apache License. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"mtlviewer.html\"\n+ \"loc\": \"license.html\"\n }, {\n \"title\": \"Hierarchical Edits\",\n- \"text\": \"Hierarchical Edits Hierarchical Edits 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Hierarchical Edits Hierarchical Edits Paths Vertex Edits Edge Edits Face Edits Hierarchical Edits To understand the hierarchical aspect of subdivision, we realize that subdivision itself leads to a natural hierarchy after the first level of subdivision, each face in a subdivision mesh subdivides to four quads in the Catmull-Clark scheme, or four triangles in the Loop scheme. This creates a parent and child relationship between the original face and the resulting four subdivided faces, which in turn leads to a hierarchy of subdivision as each child in turn subdivides. A hierarchical edit is an edit made to any one of the faces, edges, or vertices that arise anywhere during subdivision. Normally these subdivision components inherit values from their parents based on a set of subdivision rules that depend on the subdivision scheme. A hierarchical edit overrides these values. This allows for a compact specification of localized detail on a subdivision surface, without having to express information about the rest of the subdivision surface at the same level of detail. Release Notes 3.0.0 Hierarchical Edits have been marked as extended specification and support for hierarchical features has been removed from the 3.0 release. This decision allows for great simplifications of many areas of the subdivision algorithms. If we can identify legitimate use-cases for hierarchical tags, we will consider re-implementing them in future releases, as time and resources allow. Hierarchical Edits Paths In order to perform a hierarchical edit, we need to be able to name the subdivision component we are interested in, no matter where it may occur in the subdivision hierarchy. This leads us to a hierarchical path specification for faces, since once we have a face we can navigate to an incident edge or vertex by association. We note that in a subdivision mesh, a face always has incident vertices, which are labelled in relation to the face with an integer index starting at zero and in consecutive order according to the usual winding rules for subdivision surfaces. Faces also have incident edges, and these are labelled according to the origin vertex of the edge. In this diagram, the indices of the vertices of the base face are marked in red so on the left we have an extraordinary Catmull-Clark face with five vertices labeled 0-4 and on the right we have a regular Catmull-Clark face with four vertices labelled 0-3 . The indices of the child faces are blue note that in both the extraordinary and regular cases, the child faces are indexed the same way, i.e. the sub-face labeled n has one incident vertex that is the result of the subdivision of the parent vertex also labeled n in the parent face. Specifically, we note that the sub-face 1 in both the regular and extraordinary face is nearest to the vertex labelled 1 in the parent. The indices of the vertices of the child faces are labeled green , and this is where the difference lies between the extraordinary and regular case in the extraordinary case, vertex to vertex subdivision always results in a vertex labeled 0 , while in the regular case, vertex to vertex subdivision assigns the same index to the child vertex. Again, specifically, we note that the parent vertex indexed 1 in the extraordinary case has a child vertex 0 , while in the regular case the parent vertex indexed 1 actually has a child vertex that is indexed 1 . Note that this indexing scheme was chosen to maintain the property that the vertex labeled 0 always has the lowest uv parametric value on the face. By appending a vertex index to a face index, we can create a vertex path specification. For example, 655 2 3 0 specifies the 1st. vertex of the 3 rd. child face of the 2 nd. child face of the of the 655 th. face of the subdivision mesh. Vertex Edits Vertex hierarchical edits can modify the value or the sharpness of primitive variables for vertices and sub-vertices anywhere in the subdivision hierarchy. The edits are performed using either an add or a set operator. set indicates the primitive variable value or sharpness is to be set directly to the values specified. add adds a value to the normal result computed via standard subdivision rules. In other words, this operation allows value offsets to be applied to the mesh at any level of the hierarchy. Edge Edits Edge hierarchical edits can only modify the sharpness of primitive variables for edges and sub-edges anywhere in the subdivision hierarchy. Face Edits Face hierarchical edits can modify several properties of faces and sub-faces anywhere in the subdivision hierarchy. Modifiable properties include The set or add operators modify the value of primitive variables associated with faces. The hole operation introduces holes missing faces into the subdivision mesh at any level in the subdivision hierarchy. The faces will be deleted, and none of their children will appear you cannot unhole a face if any ancestor is a hole. This operation takes no float or string arguments. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"Hierarchical Edits Hierarchical Edits 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Hierarchical Edits Hierarchical Edits Paths Vertex Edits Edge Edits Face Edits Hierarchical Edits To understand the hierarchical aspect of subdivision, we realize that subdivision itself leads to a natural hierarchy after the first level of subdivision, each face in a subdivision mesh subdivides to four quads in the Catmull-Clark scheme, or four triangles in the Loop scheme. This creates a parent and child relationship between the original face and the resulting four subdivided faces, which in turn leads to a hierarchy of subdivision as each child in turn subdivides. A hierarchical edit is an edit made to any one of the faces, edges, or vertices that arise anywhere during subdivision. Normally these subdivision components inherit values from their parents based on a set of subdivision rules that depend on the subdivision scheme. A hierarchical edit overrides these values. This allows for a compact specification of localized detail on a subdivision surface, without having to express information about the rest of the subdivision surface at the same level of detail. Release Notes 3.0.0 Hierarchical Edits have been marked as extended specification and support for hierarchical features has been removed from the 3.0 release. This decision allows for great simplifications of many areas of the subdivision algorithms. If we can identify legitimate use-cases for hierarchical tags, we will consider re-implementing them in future releases, as time and resources allow. Hierarchical Edits Paths In order to perform a hierarchical edit, we need to be able to name the subdivision component we are interested in, no matter where it may occur in the subdivision hierarchy. This leads us to a hierarchical path specification for faces, since once we have a face we can navigate to an incident edge or vertex by association. We note that in a subdivision mesh, a face always has incident vertices, which are labelled in relation to the face with an integer index starting at zero and in consecutive order according to the usual winding rules for subdivision surfaces. Faces also have incident edges, and these are labelled according to the origin vertex of the edge. In this diagram, the indices of the vertices of the base face are marked in red so on the left we have an extraordinary Catmull-Clark face with five vertices labeled 0-4 and on the right we have a regular Catmull-Clark face with four vertices labelled 0-3 . The indices of the child faces are blue note that in both the extraordinary and regular cases, the child faces are indexed the same way, i.e. the sub-face labeled n has one incident vertex that is the result of the subdivision of the parent vertex also labeled n in the parent face. Specifically, we note that the sub-face 1 in both the regular and extraordinary face is nearest to the vertex labelled 1 in the parent. The indices of the vertices of the child faces are labeled green , and this is where the difference lies between the extraordinary and regular case in the extraordinary case, vertex to vertex subdivision always results in a vertex labeled 0 , while in the regular case, vertex to vertex subdivision assigns the same index to the child vertex. Again, specifically, we note that the parent vertex indexed 1 in the extraordinary case has a child vertex 0 , while in the regular case the parent vertex indexed 1 actually has a child vertex that is indexed 1 . Note that this indexing scheme was chosen to maintain the property that the vertex labeled 0 always has the lowest uv parametric value on the face. By appending a vertex index to a face index, we can create a vertex path specification. For example, 655 2 3 0 specifies the 1st. vertex of the 3 rd. child face of the 2 nd. child face of the of the 655 th. face of the subdivision mesh. Vertex Edits Vertex hierarchical edits can modify the value or the sharpness of primitive variables for vertices and sub-vertices anywhere in the subdivision hierarchy. The edits are performed using either an add or a set operator. set indicates the primitive variable value or sharpness is to be set directly to the values specified. add adds a value to the normal result computed via standard subdivision rules. In other words, this operation allows value offsets to be applied to the mesh at any level of the hierarchy. Edge Edits Edge hierarchical edits can only modify the sharpness of primitive variables for edges and sub-edges anywhere in the subdivision hierarchy. Face Edits Face hierarchical edits can modify several properties of faces and sub-faces anywhere in the subdivision hierarchy. Modifiable properties include The set or add operators modify the value of primitive variables associated with faces. The hole operation introduces holes missing faces into the subdivision mesh at any level in the subdivision hierarchy. The faces will be deleted, and none of their children will appear you cannot unhole a face if any ancestor is a hole. This operation takes no float or string arguments. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"hedits.html\"\n }, {\n+ \"title\": \"Introduction\",\n+ \"text\": \"Introduction Introduction 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Introduction Why Fast Subdivision Research Heritage Licensing Contributing External Resources Introduction OpenSubdiv is a set of open source libraries that implement high performance subdivision surface subdiv evaluation on massively parallel CPU and GPU architectures. This code path is optimized for drawing deforming surfaces with static topology at interactive framerates. OpenSubdiv is an API ready to be integrated into 3rd party digital content creation tools. It is not an application, nor a tool that can be used directly to create digital assets. Why Fast Subdivision Subdivision surfaces are commonly used for final rendering of character shapes for a smooth and controllable limit surfaces. However, subdivision surfaces in interactive apps are typically drawn as their polygonal control hulls because of performance. The polygonal control hull is an approximation that is offset from the true limit surface. Looking at an approximation in the interactive app makes it difficult to see exact contact, like fingers touching a potion bottle or hands touching a cheek. It also makes it difficult to see poke-throughs in cloth simulation if the skin and cloth are both approximations. This problem is particularly bad when one character is much larger than another and unequal subdiv face sizes cause approximation errors to be magnified. Maya and Pixars proprietary Presto animation system can take 100ms to subdivide a character of 30,000 polygons to the second level of subdivision 500,000 polygons. Being able to perform the same operation in less than 3ms allows the user to interact with the smooth, accurate limit surface at all times. Research The new GPU technology behind OpenSubdiv is the result of a joint research effort between Pixar and Microsoft. Feature Adaptive GPU Rendering of Catmull-Clark Subdivision Surfaces Matthias Niessner, Charles Loop, Mark Meyer, and Tony DeRose ACM Transactions on Graphics, Vol. 31 No. 1 Article 6 January 2012 httpgraphics.pixar.comlibraryGPUSubdivRenderingApaper.pdf Efficient Evaluation of Semi-Smooth Creases in Catmull-Clark Subdivision Surfaces Matthias Niessner, Charles Loop, and Guenter Greiner. Eurographics Proceedings, Cagliari, 2012 httpsniessnerlab.orgpapers20121semismoothniessner2012efficient.pdf Analytic Displacement Mapping using Hardware Tessellation Matthias Niessner, Charles Loop ACM Transactions on Graphics, Vol. 32 No. 3 Article 26 June 2013 httpsniessnerlab.orgpapers20133analyticniessner2013analytic.pdf Heritage This is the fifth-generation subdiv library in use by Pixars proprietary animation system in a lineage that started with code written by Tony DeRose and Tien Truong for Geris Game in 1996. Each generation has been a from-scratch rewrite that has built upon our experience using subdivision surfaces to make animated films. This code is live, so Pixars changes to OpenSubdiv for current and future films will be released as open source at the same time they are rolled out to Pixar animation production. Subdivision for Modeling and Animation Denis Zorin, Peter Schroder Course Notes of SIGGRAPH 1999 httpwww.multires.caltech.edupubssig99notes.pdf Subdivision Surfaces in Character Animation Tony DeRose, Michael Kass, Tien Truong Proceedings of SIGGRAPH 1998 httpgraphics.pixar.comlibraryGeripaper.pdf Recursively generated B-spline surfaces on arbitrary topological meshes Catmull, E. Clark, J. Computer-Aided Design 10 6 1978 Licensing OpenSubdiv is covered by the Apache License, and is free to use for commercial or non-commercial use. This is the same code that Pixar uses internally for animated film production. Our intent is to encourage a geometry standard for subdivision surfaces, by providing consistent i.e. yielding the same limit surface, high performance implementations on a variety of platforms. Why Apache We were looking for a commercial-friendly license that would convey our patents to the end users. This quickly narrowed the field to Microsoft Public License or Apache. Initially we chose MSPL because it handled trademarks better. But at the request of several companies we gave Apache another look, and decided to go with Apache with a very slight modification that simply says you cannot use any contributors trademarks. In other words, you can use OpenSubdiv to make a product, but you cannot use a Luxo Lamp or other character, etc. when marketing your product. License Header Contributing For details on how to contribute to OpenSubdiv, see the page on Contributing External Resources Pixar Research Pixar RD Portal Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"tags\": \"\",\n+ \"loc\": \"intro.html\"\n+ }, {\n \"title\": \"glViewer\",\n- \"text\": \"glViewer glViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS glViewer -f -yup -u -a -l refinement level -c animation loops objfiles -anim -catmark -loop -bilinear DESCRIPTION glViewer is a stand-alone application that showcases the application of uniform and feature adaptive subdivision schemes to a collection of geometric shapes. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"glViewer glViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS glViewer -f -yup -u -a -l refinement level -c animation loops objfiles -anim -catmark -loop -bilinear DESCRIPTION glViewer is a stand-alone application that showcases the application of uniform and feature adaptive subdivision schemes to a collection of geometric shapes. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"glviewer.html\"\n }, {\n \"title\": \"glStencilViewer\",\n- \"text\": \"glStencilViewer glStencilViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS glStencilViewer -f -yup -u -a -l refinement level objfiles -catmark -loop -bilinear DESCRIPTION glStencilViewer is a stand-alone application that showcases the application of pre-computed stencil tables to a collection of geometric test shapes. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"glStencilViewer glStencilViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS glStencilViewer -f -yup -u -a -l refinement level objfiles -catmark -loop -bilinear DESCRIPTION glStencilViewer is a stand-alone application that showcases the application of pre-computed stencil tables to a collection of geometric test shapes. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"glstencilviewer.html\"\n }, {\n \"title\": \"glShareTopology\",\n- \"text\": \"glShareTopology glShareTopology 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS glShareTopology -u -a -l refinement level DESCRIPTION glShareTopology is a stand-alone application that showcases the implementation of topology instancing across Compute contexts. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"glShareTopology glShareTopology 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS glShareTopology -u -a -l refinement level DESCRIPTION glShareTopology is a stand-alone application that showcases the implementation of topology instancing across Compute contexts. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"glsharetopology.html\"\n }, {\n \"title\": \"glPtexViewer\",\n- \"text\": \"glPtexViewer glPtexViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS KEYBOARD CONTROLS SEE ALSO SYNOPSIS glPtexViewer -f -yup -u -a -l isolation level -c animation loops -e environment map -d HDR diffuse map -s HDR specular map --disp displacement scale --bump bump scale ptex color file ptex displacement file ptex occlusion file ptex specular file objfiles DESCRIPTION glPtexViewer is a stand-alone application that showcases advanced HDR shading with color, displacement, occlusion and specular ptex maps. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. -e environment map A low dynamic range spherical environment map used as a background. Ideally, a color-normalized version of the HDR light probe. -d HDR diffuse map An HDR file containing a diffuse environment map typically they are low resolution blurry hemispherical convolutions of the environment light probe. -s environment map An HDR file containing a specular environment map. --disp displacement scale A scalar multiplier for the shader displacement values. --bump displacement scale A scalar multiplier for the shader bump values. ptex color file A ptex file containing RGB channels read as material albedo color. ptex displacement file A single-channel ptex file preferably float precision containing the displacement values. ptex occlusion file A single-channel ptex file preferably 8 bits precision containing a pre-computed ambient occlusion signal. ptex specular file A single-channel ptex file preferably 8 bits precision applied to modulate the specular reflectance of the material KEYBOARD CONTROLS q quit esc hide GUI x save screenshot f fit frame - increase decrease tessellation rate r reload and re - compile the shader files e draw normals SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"glPtexViewer glPtexViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS KEYBOARD CONTROLS SEE ALSO SYNOPSIS glPtexViewer -f -yup -u -a -l isolation level -c animation loops -e environment map -d HDR diffuse map -s HDR specular map --disp displacement scale --bump bump scale ptex color file ptex displacement file ptex occlusion file ptex specular file objfiles DESCRIPTION glPtexViewer is a stand-alone application that showcases advanced HDR shading with color, displacement, occlusion and specular ptex maps. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. -e environment map A low dynamic range spherical environment map used as a background. Ideally, a color-normalized version of the HDR light probe. -d HDR diffuse map An HDR file containing a diffuse environment map typically they are low resolution blurry hemispherical convolutions of the environment light probe. -s environment map An HDR file containing a specular environment map. --disp displacement scale A scalar multiplier for the shader displacement values. --bump displacement scale A scalar multiplier for the shader bump values. ptex color file A ptex file containing RGB channels read as material albedo color. ptex displacement file A single-channel ptex file preferably float precision containing the displacement values. ptex occlusion file A single-channel ptex file preferably 8 bits precision containing a pre-computed ambient occlusion signal. ptex specular file A single-channel ptex file preferably 8 bits precision applied to modulate the specular reflectance of the material KEYBOARD CONTROLS q quit esc hide GUI x save screenshot f fit frame - increase decrease tessellation rate r reload and re - compile the shader files e draw normals SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"glptexviewer.html\"\n }, {\n- \"title\": \"FAR Overview\",\n- \"text\": \"FAR Overview FAR Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Feature Adaptive Representation Far Topology Refinement Primvar Refinement FarTopologyRefiner FarTopologyRefinerFactory FarPrimvarRefiner FarPatchTable Patch Arrays Patch Types Patch Parameterization Single-Crease Patches Local Points Legacy Gregory Patches FarStencilTable Advantages Principles Cascading Stencils Limit Stencils Sample Location On Extraordinary Faces Code example Feature Adaptive Representation Far Far is the primary API layer for processing client-supplied mesh data into subdivided surfaces. The Far interface may be used directly and also may be used to prepare mesh data for further processing by Osd . The two main aspects of the subdivision process are Topology Refinement and Primvar Refinement . Topology Refinement Topology refinement is the process of splitting the mesh topology according to the specified subdivison rules to generate new topological vertices, edges, and faces. This process is purely topological and does not depend on the speciific values of any primvar data point positions, etc. Topology refinement can be either uniform or adaptive, where extraordinary features are automatically isolated see feature adaptive subdivision . The Far topology classes present a public interface for the refinement functionality provided in Vtr , The main classes in Far related to topology refinement are TopologyRefiner A class encapsulating mesh refinement. TopologyLevel A class representing one level of refinement within a TopologyRefiner. TopologyRefinerFactoryMESH A factory class template specialized in terms of the applications mesh representation used to construct TopologyRefiner instances. Primvar Refinement Primvar refinement is the process of computing values for primvar data points, colors, normals, texture coordinates, etc by applying weights determined by the specified subdivision rules. There are many advantages gained by distinguishing between topology refinement and primvar interpolation including the ability to apply a single static topological refinement to multiple primvar instances or to different animated primvar time samples. Far supports methods to refine primvar data at the locations of topological vertices and at arbitrary locations on the subdivision limit surface. The main classes in Far related to primvar refinement are PrimvarRefiner A class implementing refinement of primvar data at the locations of topological vertices. PatchTable A representation of the refined surface topology that can be used for efficient evaluation of primvar data at arbitrary locations. StencilTable A representation of refinement weights suitable for efficient parallel processing of primvar refinement. LimitStencilTable A representation of refinement weights suitable for efficient parallel processing of primvar refinement at arbitrary limit surface locations. FarTopologyRefiner TopologyRefiner is the building block for many other useful classes in Far . It performs refinement of an arbitrary mesh and provides access to the refined mesh topology. It can be used for primvar refinement directly using PrimvarRefiner or indirectly by being used to create a stencil table , or a patch table , etc. TopologyRefiner provides the public refinement methods RefineUniform and RefineAdapative which perform refinement operations using Vtr. TopologyRefiner provides access to the refined topology via TopologyLevel instances. FarTopologyRefinerFactory Consistent with other classes in Far, instances of TopologyRefiner are created by a factory class -- in this case FarTopologyRefinerFactory. Here we outline several approaches for converting mesh topology into the required FarTopologyRefiner. Additional documentation is provided with the FarTopologyRefinerFactoryMESH class template used by all, and each has a concrete example provided in one of the tutorials or in the Far code itself. There are three ways to create TopologyRefiners use the existing TopologyRefinerFactoryTopologyDescriptor with a populated instance of TopologyDescriptor specialize TopologyRefinerFactoryclass MESH for more efficient conversion, using only face-vertex information fully specialize TopologyRefinerFactorclass MESH for most control over conversion Use the FarTopologyDescriptor FarTopologyDescriptor is a simple struct that can be initialized to refer to raw mesh topology information -- primarily a face-vertex list -- and then passed to a provided factory class to create a TopologyRefiner from each. Topologically, the minimal requirement consists of the number of vertices and faces of the mesh an array containing the number of vertices per face an array containing the vertices assigned to each face These last two define one of the six topological relations that are needed internally by Vtr, but this one relation is sufficient to construct the rest. Additional members are available to assign sharpness values per edge andor vertex, hole tags to faces, or to define multiple sets channels of face-varying data. Almost all of the Far tutorials i.e. tutorialsfartutorial_ illustrate use of the TopologyDescriptor and its factory for creating TopologyRefiners, i.e. TopologyRefinerFactoryTopologyDescriptor. For situations when users have raw mesh data and have not yet constructed a boundary representation of their own, it is hoped that this will suffice. Options have even been provided to indicate that raw topology information has been defined in a left-hand winding order and the factory will handle the conversion to right-hand counter-clockwise winding on-the-fly to avoid unnecessary data duplication. Custom Factory for Face Vertices If the nature of the TopologyDescriptors data expectations is not helpful, and so conversion to large temporary arrays would be necessary to properly make use of it, it may be worth writing a custom factory. Specialization of TopologyRefinerFactoryclass MESH should be done with care as the goal here is to maximize the performance of the conversion and so minimize overhead due to runtime validation. The template provides the high-level construction of the required topology vectors of the underlying Vtr. There are two ways to write such a factory provide only the face-vertex information for topology and let the factory infer all edges and other relationships, or provide the complete edge list and all other topological relationships directly. The latter is considerably more involved and described in a following section. The definition of TopologyRefinerFactoryTopologyDescriptor provides a clear and complete example of constructing a TopologyRefiner with minimal topology information, i.e. the face-vertex list. The class template TopologyRefinerFactoryMESH documents the needs here and the TopologyDescriptor instantiation and specialization should illustrate that. Custom Factory for Direct Conversion Fully specializing a factory for direct conversion is needed only for those requiring ultimate control and is not generally recommended. It is recommended that one of the previous two methods initially be used to convert your mesh topology into a TopologyRefiner. If the conversion performance is critical, or significant enough to warrant improvement, then it is worth writing a factory for full topological conversion. Writing a custom factory requires the specificationspecialization of two methods with the following purpose specify the sizes of topological data so that vectors can be pre-allocated assign the topological data to the newly allocated vectors As noted above, the assumption here is that the clients boundary-rep knows best how to retrieve the data that we require most efficiently. After the factory class gathers sizing information and allocates appropriate memory, the factory provides the client with locations of the appropriate tables to be populated using the same Array classes and interface used to access the tables. The client is expected to load a complete topological description along with additional optional data, i.e. the six topological relations required by Vtr, oriented when manifold sharpness values for edges andor vertices optional additional tags related to the components, e.g. holes optional values-per-face for face-varying channels optional This approach requires dealing directly with edges, unlike the other two. In order to convert edges into a TopologyRefiners representation, the edges need to be expressed as a collection of known size N -- each of which is referred to directly by indices 0,N-1. This can be awkward for representations such as half-edge or quad-edge that do not treat the instance of an edge uniquely. Particular care is also necessary when representing non-manifold features. The previous two approaches will construct non-manifold features as required from the face-vertex list -- dealing with degenerate edges and other non-manifold features as encountered. When directly translating full topology it is necessary to tag non-manifold features, and also to ensure that certain edge relationships are satisfied in their presence. More details are available with the assembly methods of the factory class template. While there is plenty of opportunity for user error here, that is no different from any other conversion process. Given that Far controls the construction process through the Factory class, we do have ample opportunity to insert runtime validation, and to vary that level of validation at any time on an instance of the Factory. The factory does provide run-time validation on the topology constructed that can be used for debugging purposes. A common base class has been created for the factory class, i.e. template class MESH class TopologyRefinerFactory public TopologyRefinerFactoryBase both to provide common code independent of MESH and also potentially to protect core code from unwanted specialization. FarPrimvarRefiner PrimvarRefiner supports refinement of arbitrary primvar data at the locations of topological vertices. A PrimvarRefiner accesses topology data directly from a TopologyRefiner. Different methods are provided to support three different classes of primvar interpolation. These methods may be used to refine primvar data to a specified refinement level. Interpolate... Interpolate using vertex weights InterpolateVarying... Interpolate using linear weights InterpolateFaceVarying... Interpolate using face-varying weights Additional methods allow primvar data to be interpolated to the final limit surface including the calculation of first derivative tangents. Limitdst Interpolate to the limit surface using vertex weights Limitdst, dstTan1, dstTan2 Interpolate including first derivatives to the limit surface using vertex weights LimitFaceVarying... Interpolate to the limit surface using face-varying weights PrimarRefiner provides a straightforward interface for refining primvar data, but depending on the application use case, it can be more efficient to create and use a StencilTable , or PatchTable , to refine primvar data. FarPatchTable PatchTable is the collection of patches derived from the refined faces of a particular mesh topology. This collection is created using FarPatchTableFactory from an instance of FarTopologyRefiner after refinement has been applied. Patch Arrays The PatchTable is organized into patch arrays. All patches in each array have the same type except for face-varying patch arrays which may have a mix of regular and irregular patch types. The PatchDescriptor provides the fundamental description of a patch, including the number of control points per patch as well as the basis for patch evaluation. Each patch in the array is associated with a PatchParam which specifies additional information about the individual patch. Patch Types The following are the different patch types that can be represented in the PatchTable Patch Type CVs Description NON_PATCH na Undefined patch type POINTS 1 Points useful for cage drawing LINES 2 Lines useful for cage drawing QUADS 4 Bi-linear quadrilaterals TRIANGLES 3 Linear triangles LOOP 12 Quartic triangular Box-spline patches REGULAR 16 Bi-cubic B-spline patches GREGORY 4 Legacy Gregory patches GREGORY_BOUNDARY 4 Legacy Gregory Boundary patches GREGORY_BASIS 20 Bi-cubic quadrilateral Gregory patches GREGORY_TRIANGLE 18 Quartic triangular Gregory patches The type of a patch dictates the number of control vertices expected in the table as well as the method used to evaluate values. Patch Parameterization Here we describe the encoding of the patch parameterization for quadrilateral patches. The encoding for triangular patches is similar, please see the API documentation of FarPatchParam for details. Each patch represents a specific portion of the parametric space of the coarse topological face identified by the PatchParam FaceId. As topological refinement progresses through successive levels, each resulting patch corresponds to a smaller and smaller subdomain of the face. The PatchParam UV origin describes the mapping from the uv domain of the patch to the uv subdomain of the topological face. We encode this uv origin using log2 integer values for compactness and efficiency. It is important to note that this uv parameterization is the intrinsic parameterization within a given patch or coarse face and is distinct from any client specified face-varying channel data. Patches which result from irregular coarse faces non-quad faces in the Catmark scheme are offset by the one additional level needed to quadrangulate the irregular face. It is the indices of these offset faces that are stored in the PatchParam and used in other classes such as the FarPatchMap. These offset indices can be identified from the coarse face using the FarPtexIndices class when needed. A patch along an interpolated boundary edge is supported by an incomplete sets of control vertices. For consistency, patches in the PatchTable always have a full set of control vertex indices and the PatchParam Boundary bitmask identifies which control vertices are incomplete the incomplete control vertex indices are assigned values which duplicate the first valid index. Each bit in the boundary bitmask corresponds to one edge of the patch starting from the edge from the first vertex and continuing around the patch. With feature adaptive refinement, regular B-spline basis patches along interpolated boundaries will fall into one of the eight cases four boundary and four corner illustrated below Transition edges occur during feature adaptive refinement where a patch at one level of refinement is adjacent to pairs of patches at the next level of refinement. These T-junctions do not pose a problem when evaluating primvar data on patches, but they must be taken into consideration when tessellating patches e.g. while drawing in order to avoid cracks. The PatchParam Transition bitmask identifies the transition edges of a patch. Each bit in the bitmask corresponds to one edge of the patch just like the encoding of boundary edges. After refining an arbitrary mesh, any of the 16 possible transition edge configurations might occur. The method of handling transition edges is delegated to patch drawing code. Single-Crease Patches Using single-crease patches allows a mesh with creases to be represented with many fewer patches than would be needed otherwise. A single-crease patch is a variation of a regular BSpline patch with one additional crease sharpness parameter. Release Notes 3.x Evaluation of single-crease patches is currently only implemented for OSD patch drawing, but we expect to implement support in all of the evaluation code paths for future releases. Local Points The control vertices represented by a PatchTable are primarily refined points, i.e. points which result from applying the subdivision scheme uniformly or adaptively to the points of the coarse mesh. However, the final patches generated from irregular faces, e.g. patches incident on an extraordinary vertex might have a representation which requires additional local points. Legacy Gregory Patches Using Gregory patches to approximate the surface at the final patches generated from irregular faces is an alternative representation which does not require any additional local points to be computed. Instead, when Legacy Gregory patches are used, the PatchTable must also have an alternative representation of the mesh topology encoded as a vertex valence table and a quad offsets table. FarStencilTable The base container for stencil data is the StencilTable class. As with most other Far entities, it has an associated StencilTableFactory that requires a TopologyRefiner Advantages Stencils are used to factorize the interpolation calculations that subdivision schema apply to vertices of smooth surfaces. If the topology being subdivided remains constant, factorizing the subdivision weights into stencils during a pre-compute pass yields substantial amortizations at run-time when re-posing the control cage. Factorizing the subdivision weights also allows to express each subdivided vertex as a weighted sum of vertices from the control cage. This step effectively removes any data inter-dependency between subdivided vertices the computations of subdivision interpolation can be applied to each vertex in parallel without any barriers or constraint. The Osd classes leverage these properties by exploiting CPU and GPU parallelism. Principles Iterative subdivision algorithms converge towards the limit surface by successively refining the vertices of the coarse control cage. Each successive iteration interpolates the new vertices by applying polynomial weights to a basis of supporting vertices . The interpolation calculations for any given vertex can be broken down into sequences of multiply-add operations applied to the supporting vertices. Stencil table encodes a factorization of these weighted sums each stencils is created by combining the list of control vertices from the 1-ring. With iterative subdivision, each refinement step is dependent upon the previous subdivision step being completed, and a substantial number of steps may be required in order approximate the limit each subdivision step incurs an O4 n growing amount of computations. Instead, once the weights of the contributing coarse control vertices for a given refined vertex have been factorized, it is possible to apply the stencil and directly obtain the interpolated vertex data without having to process the data for the intermediate refinement levels. Cascading Stencils Client-code can control the amount of factorization of the stencils the tables can be generated with contributions all the way from a basis of coarse vertices, or reduced only to contributions from vertices from the previous level of refinement. The latter mode allows client-code to access and insert modifications to the vertex data at set refinement levels -- creating what are often referred to as hierarchical edits . Once the edits have been applied by the client-code, another set of stencils can be used to smooth the vertex data to a higher level of refinement. See implementation details, see the Far cascading stencil tutorial Limit Stencils Stencil tables can be trivially extended from discrete subdivided vertices to arbitrary locations on the limit surface. Aside from extraordinary points, every location on the limit surface can be expressed as a closed-form weighted average of a set of coarse control vertices from the 1-ring surrounding the face. The weight accumulation process is similar the control cage is adaptively subdivided around extraordinary locations. A stencil is then generated for each limit location simply by factorizing the bi-cubic Bspline patch weights over those of the contributing basis of control-vertices. The use of bi-cubic patches also allows the accumulation of analytical derivatives, so limit stencils carry a set of weights for tangent vectors. Once the stencil table has been generated, limit stencils are the most direct and efficient method of evaluation of specific locations on the limit of a subdivision surface, starting from the coarse vertices of the control cage. Also just as discrete stencils, limit stencils that are factorized from coarse control vertices do not have inter-dependencies and can be evaluated in parallel. For implementation details, see the glStencilViewer code example. Sample Location On Extraordinary Faces Each stencil is associated with a singular parametric location on the coarse mesh. The parametric location is defined as face location and local 0.0 - 1.0 u,v triplet In the case of face that are not quads, a parametric sub-face quadrant needs to be identified. This can be done either explicitly or implicitly by using the unique ptex face indices for instance. Code example When the control vertices controlPoints move in space, the limit locations can be very efficiently recomputed simply by applying the blending weights to the series of coarse control vertices class StencilType public void Clear memset x , 0 , sizeof StencilType void AddWithWeight StencilType const cv , float weight x cv . x weight y cv . y weight z cv . z weight float x , y , z std vector StencilType controlPoints , points , utan , vtan Update points by applying stencils controlStencils . UpdateValues StencilType controlPoints 0 , points 0 Update tangents by applying derivative stencils controlStencils . UpdateDerivs StencilType controlPoints 0 , utan 0 , vtan 0 Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n- \"tags\": \"\",\n- \"loc\": \"far_overview.html\"\n- }, {\n \"title\": \"glPaintTest\",\n- \"text\": \"glPaintTest glPaintTest 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS KEYBOARD CONTROLS SEE ALSO SYNOPSIS glPaintTest -f -yup -l adaptive refinement level objfiles -catmark -loop DESCRIPTION glPaintTest is a small stand-alone application showing the potential of using GPU limit tessellation for painting and sculpting applications. OPTIONS See the description of the common comand line options for the subset of common options supported here. KEYBOARD CONTROLS c use texture as color d use texture as displacement SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"glPaintTest glPaintTest 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS KEYBOARD CONTROLS SEE ALSO SYNOPSIS glPaintTest -f -yup -l adaptive refinement level objfiles -catmark -loop DESCRIPTION glPaintTest is a small stand-alone application showing the potential of using GPU limit tessellation for painting and sculpting applications. OPTIONS See the description of the common comand line options for the subset of common options supported here. KEYBOARD CONTROLS c use texture as color d use texture as displacement SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"glpainttest.html\"\n }, {\n \"title\": \"glFVarViewer\",\n- \"text\": \"glFVarViewer glFVarViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS glFVarViewer -f -u -a -l refinement level -c animation loops objfiles -catmark -loop -bilinear DESCRIPTION glFVarViewer is a stand-alone application that allows the inspection of face-varying data interpolation. The window displays 2 views left side regular 3D view of the model, with a procedural u,v texture right side a 2D view of the face-varying u,v pair The HUD allows interactive manipulation of the various face-varying boundary interpolation rules and tags. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"glFVarViewer glFVarViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS glFVarViewer -f -u -a -l refinement level -c animation loops objfiles -catmark -loop -bilinear DESCRIPTION glFVarViewer is a stand-alone application that allows the inspection of face-varying data interpolation. The window displays 2 views left side regular 3D view of the model, with a procedural u,v texture right side a 2D view of the face-varying u,v pair The HUD allows interactive manipulation of the various face-varying boundary interpolation rules and tags. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"glfvarviewer.html\"\n }, {\n \"title\": \"glEvalLimit\",\n- \"text\": \"glEvalLimit glEvalLimit 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS glEvalLimit -f -yup -u -a -l refinement level objfiles -catmark -loop -bilinear DESCRIPTION glEvalLimit is a stand-alone application that showcases the limit surface Eval module. On the given shape, random samples are generated in local s,t space. Vertex, varying and face-varying data is then computed on the surface limit and displayed as colors. In order to emphasize the dynamic nature of the EvalLimit API, where the locations can be arbitrarily updated before each evaluation, the glEvalLimit example treats each sample as a ST particle. ST Particles are a simplified parametric-space particle dynamics simulation each particle is assigned a location on the subdivision surface limit that is composed of a unique ptex face index, with a local s,t parametric pair. The system also generates an array of parametric velocities ds, dt for each particle. An Update function then applies the velocities to the locations and moves the points along the parametric space. Face boundaries are managed using a ptex adjacency table obtained from the FarTopologyRefiner. Every time a particle moves outside of the 0.0f, 1.0f parametric range, a warp function moves it to the neighboring face, or bounces it, if the edge happens to be a boundary. Note currently the adjacency code does not handle diagonal crossings, nor crossings between quad and non-quad faces. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"glEvalLimit glEvalLimit 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS glEvalLimit -f -yup -u -a -l refinement level objfiles -catmark -loop -bilinear DESCRIPTION glEvalLimit is a stand-alone application that showcases the limit surface Eval module. On the given shape, random samples are generated in local s,t space. Vertex, varying and face-varying data is then computed on the surface limit and displayed as colors. In order to emphasize the dynamic nature of the EvalLimit API, where the locations can be arbitrarily updated before each evaluation, the glEvalLimit example treats each sample as a ST particle. ST Particles are a simplified parametric-space particle dynamics simulation each particle is assigned a location on the subdivision surface limit that is composed of a unique ptex face index, with a local s,t parametric pair. The system also generates an array of parametric velocities ds, dt for each particle. An Update function then applies the velocities to the locations and moves the points along the parametric space. Face boundaries are managed using a ptex adjacency table obtained from the FarTopologyRefiner. Every time a particle moves outside of the 0.0f, 1.0f parametric range, a warp function moves it to the neighboring face, or bounces it, if the edge happens to be a boundary. Note currently the adjacency code does not handle diagonal crossings, nor crossings between quad and non-quad faces. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"glevallimit.html\"\n }, {\n- \"title\": \"dxPtexViewer\",\n- \"text\": \"dxPtexViewer dxPtexViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS KEYBOARD CONTROLS SEE ALSO SYNOPSIS dxPtexViewer -f -yup -u -a -l isolation level -c animation loops -e environment map -d HDR diffuse map -s HDR specular map --disp displacement scale ptex color file ptex displacement file ptex occlusion file ptex specular file objfiles DESCRIPTION dxPtexViewer is a stand-alone application that showcases advanced HDR shading with color, displacement, occlusion and specular ptex maps. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. -y Swap Z-up geometry to Y-UP. -e environment map A low dynamic range spherical environment map used as a background. Ideally, a color-normalized version of the HDR light probe. -d HDR diffuse map An HDR file containing a diffuse environment map typically they are low resolution blurry hemispherical convolutions of the environment light probe. -s environment map An HDR file containing a specular environment map. --disp displacement scale A scalar multiplier for the shader displacement values. ptex color file A ptex file containing RGB channels read as material albedo color. ptex displacement file A single-channel ptex file preferrably float precision containing the displacement values. ptex occlusion file A single-channel ptex file preferrably 8 bits precision containing a pre-computed ambient occlusion signal. ptex specular file A single-channel ptex file preferrably 8 bits precision applied to modulate the specular reflectance of the material objfiles A sequence of obj files used as an animation loop the topology has to match the data contained in all the ptex files KEYBOARD CONTROLS q quit esc hide GUI f fit frame - increase decrese tessellation rate SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"Getting Started\",\n+ \"text\": \"Getting Started Getting Started 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Downloading the code Branches Git Flow Checking out branches API Versions Making Changes Code Overview Downloading the code The code is hosted on a Github public repository. Download and setup information for Git tools can be found here . You can access the OpenSubdiv Git repository at httpsgithub.comPixarAnimationStudiosOpenSubdiv From there, there are several ways of downloading the OpenSubdiv source code. Zip archive downloaded from here Using a GUI client you can find a list here Please refer to the documentation of your preferred application. From the GitShell, Cygwin or the CLI assuming that you have the Git tools installed, you can clone the OpenSubdiv repository directly with the following command git clone https github.comPixarAnimationStudiosOpenSubdiv.git These methods only pull static archives, which is are not under the version control system and therefore cannot pull updates or push changes back. If you intend on contributing features or fixes to the main trunk of the code, you will need to create a free Github account and clone a fork of the OpenSubdiv repository. Submissions to the main code trunk can be sent using Gits pull-request mechanisms. Please note that we are using the git flow tools so all changes should be made to our dev branch. Before we can accept submissions however, we will need a signed Contributors License Agreement . Branches Git Flow Since version 1.1.0, OpenSubdiv has adopted the Git Flow branching model . Our active development branch is named dev all new features and bug fixes should be submitted to this branch. The changes submitted to the dev branch are periodically patched to the release branch as new versions are released. Checking out branches The Git Flow tools are not a requisite for working with the OpenSubdiv code base, but new work should always be performed in the dev branch, or dedicated feature-branches. By default, a cloned repository will be pointing to the release branch. You can switch to the dev branch using the following command git checkout dev You can check that the branch has now been switched simply with git branch Which should return dev release API Versions OpenSubdiv maintains an internal API versioning system. The version number can be read from the file .opensubdivversion.h . Following the Git-Flow pattern, our releases are indexed using Gits tagging system. List of the existing tags git tag -- list Checking out version 1.2.0 git checkout v1_2_0 Making Changes Direct push access to the OpenSubdiv repository is currently limited to a small internal development team. External code should be submitted by sending Git pull-requests from forks of our dev branch. Code Overview The OpenSubdiv code base contains the following main areas Directory Contents .opensubdiv The main subdivision APIs Sdc, Vtr, Far and Osd. .examples A small collection of standalone applications that illustrate how to deploy the various features and optimizations of the OpenSubdiv APIs. The GL-based examples rely on the cross-platform GLFW API for interactive window management, while the DirectX ones are OS-native. .tutorials Tutorials showing how to manipulate the APIs of OpenSubdiv. .documentation The reStructuredText source files along with python scripts that generate the HTML documentation site. .regression Standalone regression tests and baseline data to help maintain the integrity of our APIs. If GPU SDKs are detected, some tests will attempt to run computations on those GPUs. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"dxptexviewer.html\"\n+ \"loc\": \"getting_started.html\"\n }, {\n- \"title\": \"Contributing to OpenSubdiv\",\n- \"text\": \"Contributing to OpenSubdiv Contributing to OpenSubdiv 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Contributor License Agreement Understand Git Recommended Git Workflow Sending a Pull Request Contributor License Agreement In order for us to accept code submissions merge git pull-requests, contributors need to sign the Contributor License Agreement CLA. There are two CLAs, one for individuals and one for corporations. As for the end-user license, both are based on Apache. They are found in the code repository individual form , corporate form . Please email the signed CLA to opensubdiv-clapixar.com . Understand Git First, you should familiarize yourself with the Git data model and commands. For small changes you may not need to understand Git deeply, but for larger changes or working with the codebase over a long period of time, it becomes critical to understand more of whats going on under the hood. There are many free resources on the internet, one which weve found useful is the following e-book httpsgithub.compluralsightgit-internals-pdfreleases Recommended Git Workflow Once you have a local development tree cloned and working, you can start making changes. You will need to integrate changes from the source tree as you work the following outlines the workflow used by core OpenSubdiv engineers at Pixar and DreamWorks Fork the repository into your own local copy. This can be done via the GitHub website using the fork button. Clone your fork locally git clone your_fork_url OpenSubdiv.your_name e.g. git clone httpsgithub.comyourusernameOpenSubdiv.git OpenSubdiv.yourusername Setup two remotes, origin and upstream . Origin will be setup as a result of cloning your remote repository, but upstream must be setup manually git remote add upstream httpsgithub.comPixarAnimationStudiosOpenSubdiv.git Verify your remotes are setup correctly git remote -v Which should look something like origin httpsgithub.comyourusernameOpenSubdiv.git fetch origin httpsgithub.comyourusernameOpenSubdiv.git push upstream httpsgithub.comPixarAnimationStudiosOpenSubdiv.git fetch upstream httpsgithub.comPixarAnimationStudiosOpenSubdiv.git push Finally, fetch the upstream content this is required for the next step git fetch upstream Setup a new branch for each change. Working with branches in Git is its greatest pleasure, we strongly suggest setting up a new branch for each change which you plan to pull-request. All work is done in the dev branch, so be sure to keep your change in sync with this upstream branch. To begin, start your new branch from the dev branch git checkout -b dev-feature upstreamdev As you are working on your feature, new changes will be merged into the upstream repository, to sync these changes down and preserve your local edits, you can continually rebase your local work git pull --rebase upstream dev Notice the --rebase option here. It updates the current branch to the upstreamdev branch and rebases all edits so they are at the head of your local feature branch. Alternatively, you can rebase all your work at once when your feature is complete. Sending a Pull Request First, rebase and squash your changes appropriately to produce a clean set of changes at the head of your tree. We require changes to be grouped locally to ensure that rolling back changes can be done easily. If youve followed the steps above, your pending change should already be queued up as required. If you have not, you may need to rebase and squash changes at this point. Once the change is clean, push your changes to origin and go to the GitHub website to submit your pull request. Be sure to submit your request against the dev branch. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"FAR Overview\",\n+ \"text\": \"FAR Overview FAR Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Feature Adaptive Representation Far Topology Refinement Primvar Refinement FarTopologyRefiner FarTopologyRefinerFactory FarPrimvarRefiner FarPatchTable Patch Arrays Patch Types Patch Parameterization Single-Crease Patches Local Points Legacy Gregory Patches FarStencilTable Advantages Principles Cascading Stencils Limit Stencils Sample Location On Extraordinary Faces Code example Feature Adaptive Representation Far Far is the primary API layer for processing client-supplied mesh data into subdivided surfaces. The Far interface may be used directly and also may be used to prepare mesh data for further processing by Osd . The two main aspects of the subdivision process are Topology Refinement and Primvar Refinement . Topology Refinement Topology refinement is the process of splitting the mesh topology according to the specified subdivison rules to generate new topological vertices, edges, and faces. This process is purely topological and does not depend on the speciific values of any primvar data point positions, etc. Topology refinement can be either uniform or adaptive, where extraordinary features are automatically isolated see feature adaptive subdivision . The Far topology classes present a public interface for the refinement functionality provided in Vtr , The main classes in Far related to topology refinement are TopologyRefiner A class encapsulating mesh refinement. TopologyLevel A class representing one level of refinement within a TopologyRefiner. TopologyRefinerFactoryMESH A factory class template specialized in terms of the applications mesh representation used to construct TopologyRefiner instances. Primvar Refinement Primvar refinement is the process of computing values for primvar data points, colors, normals, texture coordinates, etc by applying weights determined by the specified subdivision rules. There are many advantages gained by distinguishing between topology refinement and primvar interpolation including the ability to apply a single static topological refinement to multiple primvar instances or to different animated primvar time samples. Far supports methods to refine primvar data at the locations of topological vertices and at arbitrary locations on the subdivision limit surface. The main classes in Far related to primvar refinement are PrimvarRefiner A class implementing refinement of primvar data at the locations of topological vertices. PatchTable A representation of the refined surface topology that can be used for efficient evaluation of primvar data at arbitrary locations. StencilTable A representation of refinement weights suitable for efficient parallel processing of primvar refinement. LimitStencilTable A representation of refinement weights suitable for efficient parallel processing of primvar refinement at arbitrary limit surface locations. FarTopologyRefiner TopologyRefiner is the building block for many other useful classes in Far . It performs refinement of an arbitrary mesh and provides access to the refined mesh topology. It can be used for primvar refinement directly using PrimvarRefiner or indirectly by being used to create a stencil table , or a patch table , etc. TopologyRefiner provides the public refinement methods RefineUniform and RefineAdapative which perform refinement operations using Vtr. TopologyRefiner provides access to the refined topology via TopologyLevel instances. FarTopologyRefinerFactory Consistent with other classes in Far, instances of TopologyRefiner are created by a factory class -- in this case FarTopologyRefinerFactory. Here we outline several approaches for converting mesh topology into the required FarTopologyRefiner. Additional documentation is provided with the FarTopologyRefinerFactoryMESH class template used by all, and each has a concrete example provided in one of the tutorials or in the Far code itself. There are three ways to create TopologyRefiners use the existing TopologyRefinerFactoryTopologyDescriptor with a populated instance of TopologyDescriptor specialize TopologyRefinerFactoryclass MESH for more efficient conversion, using only face-vertex information fully specialize TopologyRefinerFactorclass MESH for most control over conversion Use the FarTopologyDescriptor FarTopologyDescriptor is a simple struct that can be initialized to refer to raw mesh topology information -- primarily a face-vertex list -- and then passed to a provided factory class to create a TopologyRefiner from each. Topologically, the minimal requirement consists of the number of vertices and faces of the mesh an array containing the number of vertices per face an array containing the vertices assigned to each face These last two define one of the six topological relations that are needed internally by Vtr, but this one relation is sufficient to construct the rest. Additional members are available to assign sharpness values per edge andor vertex, hole tags to faces, or to define multiple sets channels of face-varying data. Almost all of the Far tutorials i.e. tutorialsfartutorial_ illustrate use of the TopologyDescriptor and its factory for creating TopologyRefiners, i.e. TopologyRefinerFactoryTopologyDescriptor. For situations when users have raw mesh data and have not yet constructed a boundary representation of their own, it is hoped that this will suffice. Options have even been provided to indicate that raw topology information has been defined in a left-hand winding order and the factory will handle the conversion to right-hand counter-clockwise winding on-the-fly to avoid unnecessary data duplication. Custom Factory for Face Vertices If the nature of the TopologyDescriptors data expectations is not helpful, and so conversion to large temporary arrays would be necessary to properly make use of it, it may be worth writing a custom factory. Specialization of TopologyRefinerFactoryclass MESH should be done with care as the goal here is to maximize the performance of the conversion and so minimize overhead due to runtime validation. The template provides the high-level construction of the required topology vectors of the underlying Vtr. There are two ways to write such a factory provide only the face-vertex information for topology and let the factory infer all edges and other relationships, or provide the complete edge list and all other topological relationships directly. The latter is considerably more involved and described in a following section. The definition of TopologyRefinerFactoryTopologyDescriptor provides a clear and complete example of constructing a TopologyRefiner with minimal topology information, i.e. the face-vertex list. The class template TopologyRefinerFactoryMESH documents the needs here and the TopologyDescriptor instantiation and specialization should illustrate that. Custom Factory for Direct Conversion Fully specializing a factory for direct conversion is needed only for those requiring ultimate control and is not generally recommended. It is recommended that one of the previous two methods initially be used to convert your mesh topology into a TopologyRefiner. If the conversion performance is critical, or significant enough to warrant improvement, then it is worth writing a factory for full topological conversion. Writing a custom factory requires the specificationspecialization of two methods with the following purpose specify the sizes of topological data so that vectors can be pre-allocated assign the topological data to the newly allocated vectors As noted above, the assumption here is that the clients boundary-rep knows best how to retrieve the data that we require most efficiently. After the factory class gathers sizing information and allocates appropriate memory, the factory provides the client with locations of the appropriate tables to be populated using the same Array classes and interface used to access the tables. The client is expected to load a complete topological description along with additional optional data, i.e. the six topological relations required by Vtr, oriented when manifold sharpness values for edges andor vertices optional additional tags related to the components, e.g. holes optional values-per-face for face-varying channels optional This approach requires dealing directly with edges, unlike the other two. In order to convert edges into a TopologyRefiners representation, the edges need to be expressed as a collection of known size N -- each of which is referred to directly by indices 0,N-1. This can be awkward for representations such as half-edge or quad-edge that do not treat the instance of an edge uniquely. Particular care is also necessary when representing non-manifold features. The previous two approaches will construct non-manifold features as required from the face-vertex list -- dealing with degenerate edges and other non-manifold features as encountered. When directly translating full topology it is necessary to tag non-manifold features, and also to ensure that certain edge relationships are satisfied in their presence. More details are available with the assembly methods of the factory class template. While there is plenty of opportunity for user error here, that is no different from any other conversion process. Given that Far controls the construction process through the Factory class, we do have ample opportunity to insert runtime validation, and to vary that level of validation at any time on an instance of the Factory. The factory does provide run-time validation on the topology constructed that can be used for debugging purposes. A common base class has been created for the factory class, i.e. template class MESH class TopologyRefinerFactory public TopologyRefinerFactoryBase both to provide common code independent of MESH and also potentially to protect core code from unwanted specialization. FarPrimvarRefiner PrimvarRefiner supports refinement of arbitrary primvar data at the locations of topological vertices. A PrimvarRefiner accesses topology data directly from a TopologyRefiner. Different methods are provided to support three different classes of primvar interpolation. These methods may be used to refine primvar data to a specified refinement level. Interpolate... Interpolate using vertex weights InterpolateVarying... Interpolate using linear weights InterpolateFaceVarying... Interpolate using face-varying weights Additional methods allow primvar data to be interpolated to the final limit surface including the calculation of first derivative tangents. Limitdst Interpolate to the limit surface using vertex weights Limitdst, dstTan1, dstTan2 Interpolate including first derivatives to the limit surface using vertex weights LimitFaceVarying... Interpolate to the limit surface using face-varying weights PrimarRefiner provides a straightforward interface for refining primvar data, but depending on the application use case, it can be more efficient to create and use a StencilTable , or PatchTable , to refine primvar data. FarPatchTable PatchTable is the collection of patches derived from the refined faces of a particular mesh topology. This collection is created using FarPatchTableFactory from an instance of FarTopologyRefiner after refinement has been applied. Patch Arrays The PatchTable is organized into patch arrays. All patches in each array have the same type except for face-varying patch arrays which may have a mix of regular and irregular patch types. The PatchDescriptor provides the fundamental description of a patch, including the number of control points per patch as well as the basis for patch evaluation. Each patch in the array is associated with a PatchParam which specifies additional information about the individual patch. Patch Types The following are the different patch types that can be represented in the PatchTable Patch Type CVs Description NON_PATCH na Undefined patch type POINTS 1 Points useful for cage drawing LINES 2 Lines useful for cage drawing QUADS 4 Bi-linear quadrilaterals TRIANGLES 3 Linear triangles LOOP 12 Quartic triangular Box-spline patches REGULAR 16 Bi-cubic B-spline patches GREGORY 4 Legacy Gregory patches GREGORY_BOUNDARY 4 Legacy Gregory Boundary patches GREGORY_BASIS 20 Bi-cubic quadrilateral Gregory patches GREGORY_TRIANGLE 18 Quartic triangular Gregory patches The type of a patch dictates the number of control vertices expected in the table as well as the method used to evaluate values. Patch Parameterization Here we describe the encoding of the patch parameterization for quadrilateral patches. The encoding for triangular patches is similar, please see the API documentation of FarPatchParam for details. Each patch represents a specific portion of the parametric space of the coarse topological face identified by the PatchParam FaceId. As topological refinement progresses through successive levels, each resulting patch corresponds to a smaller and smaller subdomain of the face. The PatchParam UV origin describes the mapping from the uv domain of the patch to the uv subdomain of the topological face. We encode this uv origin using log2 integer values for compactness and efficiency. It is important to note that this uv parameterization is the intrinsic parameterization within a given patch or coarse face and is distinct from any client specified face-varying channel data. Patches which result from irregular coarse faces non-quad faces in the Catmark scheme are offset by the one additional level needed to quadrangulate the irregular face. It is the indices of these offset faces that are stored in the PatchParam and used in other classes such as the FarPatchMap. These offset indices can be identified from the coarse face using the FarPtexIndices class when needed. A patch along an interpolated boundary edge is supported by an incomplete sets of control vertices. For consistency, patches in the PatchTable always have a full set of control vertex indices and the PatchParam Boundary bitmask identifies which control vertices are incomplete the incomplete control vertex indices are assigned values which duplicate the first valid index. Each bit in the boundary bitmask corresponds to one edge of the patch starting from the edge from the first vertex and continuing around the patch. With feature adaptive refinement, regular B-spline basis patches along interpolated boundaries will fall into one of the eight cases four boundary and four corner illustrated below Transition edges occur during feature adaptive refinement where a patch at one level of refinement is adjacent to pairs of patches at the next level of refinement. These T-junctions do not pose a problem when evaluating primvar data on patches, but they must be taken into consideration when tessellating patches e.g. while drawing in order to avoid cracks. The PatchParam Transition bitmask identifies the transition edges of a patch. Each bit in the bitmask corresponds to one edge of the patch just like the encoding of boundary edges. After refining an arbitrary mesh, any of the 16 possible transition edge configurations might occur. The method of handling transition edges is delegated to patch drawing code. Single-Crease Patches Using single-crease patches allows a mesh with creases to be represented with many fewer patches than would be needed otherwise. A single-crease patch is a variation of a regular BSpline patch with one additional crease sharpness parameter. Release Notes 3.x Evaluation of single-crease patches is currently only implemented for OSD patch drawing, but we expect to implement support in all of the evaluation code paths for future releases. Local Points The control vertices represented by a PatchTable are primarily refined points, i.e. points which result from applying the subdivision scheme uniformly or adaptively to the points of the coarse mesh. However, the final patches generated from irregular faces, e.g. patches incident on an extraordinary vertex might have a representation which requires additional local points. Legacy Gregory Patches Using Gregory patches to approximate the surface at the final patches generated from irregular faces is an alternative representation which does not require any additional local points to be computed. Instead, when Legacy Gregory patches are used, the PatchTable must also have an alternative representation of the mesh topology encoded as a vertex valence table and a quad offsets table. FarStencilTable The base container for stencil data is the StencilTable class. As with most other Far entities, it has an associated StencilTableFactory that requires a TopologyRefiner Advantages Stencils are used to factorize the interpolation calculations that subdivision schema apply to vertices of smooth surfaces. If the topology being subdivided remains constant, factorizing the subdivision weights into stencils during a pre-compute pass yields substantial amortizations at run-time when re-posing the control cage. Factorizing the subdivision weights also allows to express each subdivided vertex as a weighted sum of vertices from the control cage. This step effectively removes any data inter-dependency between subdivided vertices the computations of subdivision interpolation can be applied to each vertex in parallel without any barriers or constraint. The Osd classes leverage these properties by exploiting CPU and GPU parallelism. Principles Iterative subdivision algorithms converge towards the limit surface by successively refining the vertices of the coarse control cage. Each successive iteration interpolates the new vertices by applying polynomial weights to a basis of supporting vertices . The interpolation calculations for any given vertex can be broken down into sequences of multiply-add operations applied to the supporting vertices. Stencil table encodes a factorization of these weighted sums each stencils is created by combining the list of control vertices from the 1-ring. With iterative subdivision, each refinement step is dependent upon the previous subdivision step being completed, and a substantial number of steps may be required in order approximate the limit each subdivision step incurs an O4 n growing amount of computations. Instead, once the weights of the contributing coarse control vertices for a given refined vertex have been factorized, it is possible to apply the stencil and directly obtain the interpolated vertex data without having to process the data for the intermediate refinement levels. Cascading Stencils Client-code can control the amount of factorization of the stencils the tables can be generated with contributions all the way from a basis of coarse vertices, or reduced only to contributions from vertices from the previous level of refinement. The latter mode allows client-code to access and insert modifications to the vertex data at set refinement levels -- creating what are often referred to as hierarchical edits . Once the edits have been applied by the client-code, another set of stencils can be used to smooth the vertex data to a higher level of refinement. See implementation details, see the Far cascading stencil tutorial Limit Stencils Stencil tables can be trivially extended from discrete subdivided vertices to arbitrary locations on the limit surface. Aside from extraordinary points, every location on the limit surface can be expressed as a closed-form weighted average of a set of coarse control vertices from the 1-ring surrounding the face. The weight accumulation process is similar the control cage is adaptively subdivided around extraordinary locations. A stencil is then generated for each limit location simply by factorizing the bi-cubic Bspline patch weights over those of the contributing basis of control-vertices. The use of bi-cubic patches also allows the accumulation of analytical derivatives, so limit stencils carry a set of weights for tangent vectors. Once the stencil table has been generated, limit stencils are the most direct and efficient method of evaluation of specific locations on the limit of a subdivision surface, starting from the coarse vertices of the control cage. Also just as discrete stencils, limit stencils that are factorized from coarse control vertices do not have inter-dependencies and can be evaluated in parallel. For implementation details, see the glStencilViewer code example. Sample Location On Extraordinary Faces Each stencil is associated with a singular parametric location on the coarse mesh. The parametric location is defined as face location and local 0.0 - 1.0 u,v triplet In the case of face that are not quads, a parametric sub-face quadrant needs to be identified. This can be done either explicitly or implicitly by using the unique ptex face indices for instance. Code example When the control vertices controlPoints move in space, the limit locations can be very efficiently recomputed simply by applying the blending weights to the series of coarse control vertices class StencilType public void Clear memset x , 0 , sizeof StencilType void AddWithWeight StencilType const cv , float weight x cv . x weight y cv . y weight z cv . z weight float x , y , z std vector StencilType controlPoints , points , utan , vtan Update points by applying stencils controlStencils . UpdateValues StencilType controlPoints 0 , points 0 Update tangents by applying derivative stencils controlStencils . UpdateDerivs StencilType controlPoints 0 , utan 0 , vtan 0 Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"contributing.html\"\n+ \"loc\": \"far_overview.html\"\n }, {\n- \"title\": \"Getting Started\",\n- \"text\": \"Getting Started Getting Started 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Downloading the code Branches Git Flow Checking out branches API Versions Making Changes Code Overview Downloading the code The code is hosted on a Github public repository. Download and setup information for Git tools can be found here . You can access the OpenSubdiv Git repository at httpsgithub.comPixarAnimationStudiosOpenSubdiv From there, there are several ways of downloading the OpenSubdiv source code. Zip archive downloaded from here Using a GUI client you can find a list here Please refer to the documentation of your preferred application. From the GitShell, Cygwin or the CLI assuming that you have the Git tools installed, you can clone the OpenSubdiv repository directly with the following command git clone https github.comPixarAnimationStudiosOpenSubdiv.git These methods only pull static archives, which is are not under the version control system and therefore cannot pull updates or push changes back. If you intend on contributing features or fixes to the main trunk of the code, you will need to create a free Github account and clone a fork of the OpenSubdiv repository. Submissions to the main code trunk can be sent using Gits pull-request mechanisms. Please note that we are using the git flow tools so all changes should be made to our dev branch. Before we can accept submissions however, we will need a signed Contributors License Agreement . Branches Git Flow Since version 1.1.0, OpenSubdiv has adopted the Git Flow branching model . Our active development branch is named dev all new features and bug fixes should be submitted to this branch. The changes submitted to the dev branch are periodically patched to the release branch as new versions are released. Checking out branches The Git Flow tools are not a requisite for working with the OpenSubdiv code base, but new work should always be performed in the dev branch, or dedicated feature-branches. By default, a cloned repository will be pointing to the release branch. You can switch to the dev branch using the following command git checkout dev You can check that the branch has now been switched simply with git branch Which should return dev release API Versions OpenSubdiv maintains an internal API versioning system. The version number can be read from the file .opensubdivversion.h . Following the Git-Flow pattern, our releases are indexed using Gits tagging system. List of the existing tags git tag -- list Checking out version 1.2.0 git checkout v1_2_0 Making Changes Direct push access to the OpenSubdiv repository is currently limited to a small internal development team. External code should be submitted by sending Git pull-requests from forks of our dev branch. Code Overview The OpenSubdiv code base contains the following main areas Directory Contents .opensubdiv The main subdivision APIs Sdc, Vtr, Far and Osd. .examples A small collection of standalone applications that illustrate how to deploy the various features and optimizations of the OpenSubdiv APIs. The GL-based examples rely on the cross-platform GLFW API for interactive window management, while the DirectX ones are OS-native. .tutorials Tutorials showing how to manipulate the APIs of OpenSubdiv. .documentation The reStructuredText source files along with python scripts that generate the HTML documentation site. .regression Standalone regression tests and baseline data to help maintain the integrity of our APIs. If GPU SDKs are detected, some tests will attempt to run computations on those GPUs. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"dxPtexViewer\",\n+ \"text\": \"dxPtexViewer dxPtexViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS KEYBOARD CONTROLS SEE ALSO SYNOPSIS dxPtexViewer -f -yup -u -a -l isolation level -c animation loops -e environment map -d HDR diffuse map -s HDR specular map --disp displacement scale ptex color file ptex displacement file ptex occlusion file ptex specular file objfiles DESCRIPTION dxPtexViewer is a stand-alone application that showcases advanced HDR shading with color, displacement, occlusion and specular ptex maps. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. -y Swap Z-up geometry to Y-UP. -e environment map A low dynamic range spherical environment map used as a background. Ideally, a color-normalized version of the HDR light probe. -d HDR diffuse map An HDR file containing a diffuse environment map typically they are low resolution blurry hemispherical convolutions of the environment light probe. -s environment map An HDR file containing a specular environment map. --disp displacement scale A scalar multiplier for the shader displacement values. ptex color file A ptex file containing RGB channels read as material albedo color. ptex displacement file A single-channel ptex file preferrably float precision containing the displacement values. ptex occlusion file A single-channel ptex file preferrably 8 bits precision containing a pre-computed ambient occlusion signal. ptex specular file A single-channel ptex file preferrably 8 bits precision applied to modulate the specular reflectance of the material objfiles A sequence of obj files used as an animation loop the topology has to match the data contained in all the ptex files KEYBOARD CONTROLS q quit esc hide GUI f fit frame - increase decrese tessellation rate SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"getting_started.html\"\n+ \"loc\": \"dxptexviewer.html\"\n }, {\n \"title\": \"dxViewer\",\n- \"text\": \"dxViewer dxViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS dxViewer -f -yup -u -a -l refinement level -c animation loops objfiles -catmark -loop -bilinear DESCRIPTION dxViewer is a stand-alone application that showcases the application of uniform and feature adaptive subdivision schemes to a collection of geometric shapes. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"dxViewer dxViewer 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE SYNOPSIS DESCRIPTION OPTIONS SEE ALSO SYNOPSIS dxViewer -f -yup -u -a -l refinement level -c animation loops objfiles -catmark -loop -bilinear DESCRIPTION dxViewer is a stand-alone application that showcases the application of uniform and feature adaptive subdivision schemes to a collection of geometric shapes. Multiple controls are available to experiment with the algorithms. OPTIONS See the description of the common comand line options for the subset of common options supported here. SEE ALSO Other examples glViewer , glFVarViewer , glEvalLimit , glStencilViewer , glPtexViewer , glPaintTest , glShareTopology , dxViewer , dxPtexViewer , Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"dxviewer.html\"\n }, {\n- \"title\": \"Subdivision Compatibility\",\n- \"text\": \"Subdivision Compatibility Subdivision Compatibility 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Subdivision Compatibility Compatibility with OpenSubdiv 2.x Compatibility with RenderMan Incompatibilities Other Differences Subdivision Compatibility This document highlights areas of compatibility with other software that makes use of subdivision surfaces, including previous versions of OpenSubdiv. The compatibility here refers to the choice of subdivision rules that define the shape of the resulting surfaces. Different subdivision rules will lead to different shapes. Choices affecting shape include the types of subdivision schemes supported e.g. Catmull-Clark, Loop, etc. the basic rules applied for these schemes any extended rules to affect sharpness or creasing rules applied separately to face-varying data Ensuring all of these rules are consistent provides the basis for consistent shapes, but further approximations to the limit surface create the potential for subtle deviations. Even within OpenSubdiv, multiple approximations are possible and vary. For now we focus on the compatibility of subdivision rules and deal with the limit approximations only when noteworthy. Compatibility with OpenSubdiv 2.x The refactoring of OpenSubdiv 3.0 data representations presented a unique opportunity to revisit some corners of the subdivision specification and remove or update some legacy features. Face-varying Interpolation Options Face-varying interpolation options have been consolidated into a single enum with one additional choice new to 3.0. No functionality from 2.x has been removed -- just re-expressed in a simpler and more comprehensible form. Face-varying interpolation was previously defined by a boundary interpolation enum with four modes and an additional boolean propagate corners option, which was little understood, i.e. void HbrMeshSetFVarInterpolateBoundarMethodInterpolateBoundaryMethod const void HbrMeshSetFVarPropagateCornersbool const The latter was only used in conjunction with one of the four modes edge and corner, so it was effectively a unique fifth choice. Closer inspection of all of these modes also revealed some unexpected and undesirable behavior in some common cases -- to an extent that could not simply be changed -- and so an additional mode was added to avoid such behavior. All choices are now provided through a single linear interpolation enum, described and illustrated in more detail in the overview of Face-Varying Interpolation . The use of boundary in the name of the enum was intentionally removed as the choice also affects interior interpolation. The new use of linear is now intended to reflect the fact that interpolation is constrained to be linear where specified by the choice applied. All five of Hbrs original modes of face-varying interpolation are supported with minor modifications where Hbr was found to be incorrect in the presence of semi-sharp creasing. An additional mode corners only has also been added to avoid some of the undesired side-effects of some existing modes illustrated below. The new values for the SdcOptionsFVarLinearInterpolation enum and its equivalents for HbrMeshs InterpolateBoundaryMethod and PropagateCorners flag are as follows ordered such that the set of linear constraints applied is always increasing -- from completely smooth to completely linear Sdc FVarLinearInterpolation Hbr FVarInterpolateBoundaryMethod Hbr FVarPropogateCorners FVAR_LINEAR_NONE k_InterpolateBoundaryEdgeOnly NA ignored FVAR_LINEAR_CORNERS_ONLY NA NA FVAR_LINEAR_CORNERS_PLUS1 k_InterpolateBoundaryEdgeAndCorner false FVAR_LINEAR_CORNERS_PLUS2 k_InterpolateBoundaryEdgeAndCorner true FVAR_LINEAR_BOUNDARIES k_InterpolateBoundaryAlwaysSharp NA ignored FVAR_LINEAR_ALL k_InterpolateBoundaryNone NA ignored Aside from the two corners plus modes that preserve Hbr behavior, all other modes are designed so that the interpolation of a disjoint face-varying region is not affected by changes to other regions that may share the same vertex. So the behavior of a disjoint region should be well understood and predictable when looking at it in isolation e.g. with corners only one would expect to see linear constraints applied where there are topological corners or infinitely sharp creasing applied within the region, and nowhere else. This is not true of the plus modes, and they are named to reflect the fact that more is taken into account where disjoint regions meet. The following example illustrates some undesired effects of the plus modes, which in part motivated the addition of the new corners only mode. The example uses the catmark_fvar_bound0 and catmark_fvar_bound1 shapes from the suite of regression shapes. Both shapes are a simple regular 4x4 grid of quads with face-varying UV data partitioned into multiple disjoint regions. The bound0 shape has two disjoint UV regions -- an upper and lower region -- while the bound1 shape further splits the lower region in two. This first figure illustrates the effect of the original plus1 mode which is also the same for plus2 Note that the effect of splitting the lower UV region in two has the undesired side effect of sharpening the boundary of the upper region. This is the result of the plus1 mode making collective decisions about the sharpness of all face-varying boundaries at the vertex rather than decisions local to each region. In both the plus1 and plus2 cases, all face-varying boundaries sharing a vertex will be sharpened if there are more than two regions meeting at that vertex. The second figure illustrates the effect of the new corners only mode As expected, the splitting of the lower region does not impact the upper region. In this case the decision to sharpen a face-varying boundary is made based on the local topology of each region. Vertex Interpolation Options Since the various options are now presented through a new API Sdc rather than Hbr, based on the history of some of these options and input from interested parties, the following changes have been implemented The naming of the standard creasing method has been changed from Normal to Uniform . Values for SdcOptionsCreasingMethod are now CREASE_UNIFORM standard integer subtraction per level default CREASE_CHAIKIN Chaikin non-uniform averaging around vertices Legacy modes of the smoothtriangle rule have been removed as they were never actually enabled in the code. Values for SdcOptionsTriangleSubdivision are now TRI_SUB_CATMARK Catmull-Clark weights default TRI_SUB_SMOOTH smooth triangle weights These should have little impact since one is a simple change in terminology as part of a new API while the other was removal of an option that was never used. Change to Chaikin creasing method In the process of re-implementing the Chaikin creasing method, observations lead to a conscious choice to change the behavior of Chaikin creasing in the presence of infinitely sharp edges most noticeable at boundaries. Previously, the inclusion of infinite sharpness values in the Chaikin methods computation of edge sharpness around a vertex would prevent a semi-sharp edge from decaying to zero. Infinitely sharp edges are now excluded from the Chaikin non-uniform averaging yielding a much more predictable and desirable result. For example, where the sharpness assignment is actually uniform at such a vertex, the result will now behave the same as the Uniform method. Since this feature has received little use only recently activated in RenderMan, now seemed the best time to make the change before more widespread adoption. Hierarchical Edits While extremely powerful, Hierarchical Edits come with additional maintenance and implementation complexity. Support for them in popular interchange formats and major DCC applications has either been dropped or was never implemented. As a result, the need for Hierarchical Edits is too limited to justify the cost and support for them, and they have therefore been removed from the 3.0 release of OpenSubdiv. Dropping support for Hierarchical Edits allows for significant simplifications of many areas of the subdivision algorithms. While the 3.0 release does not offer direct support for Hierarchical Edits, the architectural changes and direction of 3.0 still facilitate the application of the most common value edits for those wishing to use them -- though not always in the same optimized context. Of course, support for Hierarchical Edits in the future will be considered based on demand and resources. Non-Manifold Topology OpenSubdiv 2.x and earlier was limited to dealing with meshes whose topology was manifold -- a limitation imposed by the use of Hbr. With 3.0 no longer using Hbr, the manifold restriction has also been removed. OpenSubdiv 3.0, therefore, supports a superset of the meshes supported by 2.x and earlier versions with one known exception noted below. Non-manifold meshes that are acceptable to 3.0 however will likely not work with 2.x or earlier. The one known case that 3.0 will not represent the same as 2.x is ironically a case that is non-manifold, and for which Hbr did make special accommodation. That case occurs at a non-manifold vertex where two or more faces meet at a common vertex, but do not share a common edge, and when the boundary interpolation mode is set for smooth corners i.e. edge only, as illustrated below The cage is on the left and is refined to level 2 on the right. On the immediate right, boundary interpolation is set to sharp corners and the results appear the same for 2.x and 3.0. The center and far right illustrate the affects of setting boundary interpolation to smooth corners with 2.x and 3.0 respectively. Note that the 2.x result allows the refined mesh and so the limit surface to split into two while the 3.0 result keeps it connected. When Hbr encounters such vertices, regardless of the boundary mode it splits the vertex -- creating a separate instance of it for each face. So when building an HbrMesh, after finalizing the mesh, it will result in having more vertices than were originally defined termed split vertices. OpenSubdiv 2.x and earlier successfully hid the presence of these extra vertices from users. This case behaves in such a way that violates certain properties of the surface that 3.0 has attempted to emphasize. One of these relates to the nature of the limit surface and becomes more significant in the context of face varying if the cage is connected then so too is its limit surface, or similarly, if the cage consists of N connected regions then the limit surface similarly consists of N connected regions. Another undesirable property here is that the vertex V at which these faces meet must have more than one child vertex V . This makes it difficult to hide split vertices -- OpenSubdiv 2.x tables had an extra level of indirection that made it possible to do this relatively easily, but 3.0 has dispensed with such indirection where possible to streamline performance. Compatibility with RenderMan Since RenderMan and OpenSubdiv versions prior to 3.0 share a common library Hbr, most differences between RenderMan and OpenSubdiv 3.0 are covered in the preceding section of compatibility with OpenSubdiv 2.x. In addition to some features between RenderMan and OpenSubdiv that are not compatible, there are also other differences that may be present due to differences in the implementations of similar features. For most use cases, OpenSubdiv 3.0 is largely compatible with RenderMan. There are however some cases where some differences can be expected. These are highlighted below for completeness. Incompatibilities OpenSubdiv and RenderMan will be incompatible when certain features are used that are not common to both. They are fully described in the 2.x compatibility section and are listed briefly here. OpenSubdiv 3.0 Features Not Supported by RenderMan Non-manifold meshes Choice of the corners only face varying interpolation option RenderMan Features Not Supported by OpenSubdiv 3.0 Hierarchical Edits Other Differences Some differences can occur due to the differing implementations of the feature sets. Additionally, OpenSubdiv 3.0s implementation fixes some issues discovered in Hbr. Smooth Face-Varying Interpolation with Creasing There have been two discrepancies noted in the way that face-varying data is interpolated smoothly in the presence of creases. Smooth face-varying interpolation is expected to match vertex interpolation in the interior and only differ along the boundaries or discontinuities where the face-varying topology is intentionally made to differ from the vertex topology. A simple and effective way to identify discrepancies is to use the X and Y coordinates of vertex positions as the U and V of texture coordinates. If these U and V coordinates are assigned to a face-varying channel, smooth interpolation of U and V is expected to exactly match interpolation of X and Y, regardless of the presence of any sharpness and creasing. Two discrepancies can be seen with Hbr when superimposing the XY vertex interpolation with the projected UV face-varying interpolation. The first discrepancy occurs with interpolation around dart vertices This example shows a simple regular XY grid on the left with an interior sharp edge creating a dart vertex in the center. With no asymmetry in the vertices, the sharpness has no asymmetric affect and the XY vertex interpolation on the immediate right shows the regular grid expected from refinement. On the far right is the UV interpolation from Hbr, which exhibits distortion around the center dart vertex. The second discrepancy occurs with interpolation involving any fractional sharpness values. Hbr effectively ignores any fractional sharpness value in its face-varying interpolation. So edges of vertices with sharpness of say 2.5, will be treated as though their sharpness is 2.0 when face-varying values are interpolated. Similarly, any non-zero sharpness value less than 1.0 is treated as zero by truncation and so is essentially ignored. This example shows an asymmetric 2x2 grid of quads on the left with the center vertex progressively sharpened from 0.5 to 1.0. The three cases of the vertex smooth and sharpened are superimposed on the immediate right to display the three distinct interpolation results. On the far right the interpolation from Hbr displays the same three cases, but only two are visibly distinct -- the sharpness of 0.5 being treated the same as if it were 0.0. Both of these cases are corrected in OpenSubdiv 3.0. Smooth face-varying interpolation in the presence of creasing should match the expected behavior of the vertex interpolation, except where the face-varying topology is explicitly made to differ. The Chaikin Creasing Method At least two discrepancies are know to exist between the implementations of Hbr in RenderMan and OpenSubdiv 3.0 Use of Chaikin creasing with boundaries or infinitely sharp edges Subtle shape differences due to Hbrs use of predictive sharpness Fortunately, this feature was only recently added to Hbr and RenderMan and is little used, so it is expected these differences will have little impact. The first discrepancy is mentioned briefly in the previous section on compatibility between OpenSubdiv 2.x and 3.0. A conscious decision was made to change the averaging of sharpness values involving infinitely sharp edges in order to make results more predictable and favorable. The effects can be seen comparing the regression shape catmark_chaikin2. The second is more subtle and results from an oversight within Hbrs implementation that is not easily corrected. When determining what subdivision rule to apply from one level to the next, the sharpness values at the next level must be known in order to determine whether or not a transition between differing rules is required. If the rule at the next level differs from the previous, a combination of the two is applied. Such a change results from the sharpness values of one or more edges or the vertex itself decaying to zero. Rather than compute the sharpness values at the next level accurately, Hbr predicts it by simply subtracting 1.0 from it, as is done with the uniform creasing method, and it bases decisions on that predicted result. This does not work for Chaikin though. A sharpness value less than 1.0 may not decay to 0 if it is averaged with neighboring sharpness values greater than 1.0, so this sharpness prediction can result in the wrong rule being chosen for the next level. A typical case would have the subdivision rules for Chaikin creasing transition from Corner to Crease at one level, then from Crease to Smooth at the next. Hbrs predictive creasing might mistakenly detect the transition as Corner to Smooth at one level, then after properly computing the sharpness values for the next level later, from Crease to Smooth for the next. One of the regression shapes catmark_chakin1 was disabled from the regression suite because of this effect. The differences in shape that trigger its regression failure were investigated and determined to be the result of this issue. From observations thus far these differences are subtle but can be noticeable. Numerical Precision Since its inception, OpenSubdiv has sought to produce results that were numerically consistent to RenderMan. A regression suite to ensure a certain level of accuracy was provided to detect any substantial deviation. At some point in the development of OpenSubdiv, the point was made that numerical accuracy of Hbr could be improved by changing the order of operations and combining the vertex with the lowest coefficient first in one of the subdivision rules. This was applied more thoroughly in the independent implementation of 3.0 there seemed no reason not to. In most cases the relative magnitudes of the coefficients of subdivision and limit masks is clear so no overhead was necessary to detect them. At a certain point though, this greater accuracy came in conflict with the regression suite. It turned out that high-valence vertices could not be computed to within the desired tolerances set within the suite. The summation of many small coefficients for the adjacent vertices first, before the addition of the much larger coefficient for the primary vertex, allowed for the accumulation of precision that was being truncated by adding the much larger coefficient first in the Hbr implementation. With extremely high valence vertices, a difference in magnitude between the most and least significant coefficients of several orders of magnitude is likely, and that has a significant impact on the single-precision floating point computations. The improved accuracy of OpenSubdiv 3.0 can reach a magnitude that will not go undetected. Whether or not this can lead to visual artifacts is unclear. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n- \"tags\": \"\",\n- \"loc\": \"compatibility.html\"\n- }, {\n \"title\": \"Downloads\",\n- \"text\": \"Downloads Downloads 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Coupe Car Coupe Car Note we can use javascripts buttons to trigger download see button-download or we can just paste regular ReST hyperlinks to the files here. Direct links probably more reliable though. Coupe car maya Download Coupe car obj Download EULA OpenSubdiv is covered by a modified Apache 2.0 license included below, and is free to use for commercial or non-commercial use. All Pixar patents in the area of subdivision surface algorithms have also been released for public use. We welcome any involvement in the development or extension of this code in fact, we would love it. Please contact us if you are interested. Copyright 2013 Pixar Licensed under the Apache License, Version 2.0 the Apache License with the following modification you may not use this file except in compliance with the Apache License and the following modification to it Section 6. Trademarks. is deleted and replaced with 6. Trademarks. This License does not grant permission to use the trade names, trademarks, service marks, or product names of the Licensor and its affiliates, except as required to comply with Section 4c of the License and to reproduce the content of the NOTICE file. You may obtain a copy of the Apache License at httpwww.apache.orglicensesLICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the Apache License with the above modification is distributed on an AS IS BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the Apache License for the specific language governing permissions and limitations under the Apache License. Accept Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"Downloads Downloads 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Coupe Car Coupe Car Note we can use javascripts buttons to trigger download see button-download or we can just paste regular ReST hyperlinks to the files here. Direct links probably more reliable though. Coupe car maya Download Coupe car obj Download EULA OpenSubdiv is covered by a modified Apache 2.0 license included below, and is free to use for commercial or non-commercial use. All Pixar patents in the area of subdivision surface algorithms have also been released for public use. We welcome any involvement in the development or extension of this code in fact, we would love it. Please contact us if you are interested. Copyright 2013 Pixar Licensed under the Apache License, Version 2.0 the Apache License with the following modification you may not use this file except in compliance with the Apache License and the following modification to it Section 6. Trademarks. is deleted and replaced with 6. Trademarks. This License does not grant permission to use the trade names, trademarks, service marks, or product names of the Licensor and its affiliates, except as required to comply with Section 4c of the License and to reproduce the content of the NOTICE file. You may obtain a copy of the Apache License at httpwww.apache.orglicensesLICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the Apache License with the above modification is distributed on an AS IS BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the Apache License for the specific language governing permissions and limitations under the Apache License. Accept Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"downloads.html\"\n }, {\n- \"title\": \"BFR Overview\",\n- \"text\": \"BFR Overview BFR Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Base Face Representation Bfr Evaluation BfrSurfaceFactory BfrSurface Parameterization BfrParameterization Discontinuous Parameterizations Tessellation BfrTessellation Tessellation Rates Differences from Hardware Tessellation More on BfrSurfaceFactory BfrSurfaceFactoryCache Defining a Thread-Safe SurfaceFactory Internal versus External SurfaceFactoryCache Customizing a BfrSurfaceFactory The BfrSurfaceFactoryMeshAdapter Interface Customizing the Subclass Interface Base Face Representation Bfr Bfr is an alternate API layer that treats a subdivision mesh provided by a client as a piecewise parameteric surface primitive . The name Bfr derives from the fact that the concepts and classes of this interface all relate to the base faces of a mesh. Concepts such as parameterization , evaluation and tessellation all refer to and are embodied by classes that deal with a specific face of the original unrefined mesh. The Bfr interfaces allow the limit surface for a single face to be identified and evaluated independently of all other faces without any global pre-processing. While concepts and utilities from the Far interface are used internally, the details of their usage is hidden. There is no need to coordinate adaptive refinement with tables of patches, stencils, Ptex indices, patch maps, etc. The resulting evaluation interface is much simpler, more flexible and more scalable than those assembled with the Far classes -- providing a preferable alternative for many CPU-based use cases. The main classes in Bfr include SurfaceFactory A light-weight interface to a mesh that constructs pieces of limit surface for specified faces of a mesh in the form of Surfaces. Surface A class encapsulating the limit surface of a face with methods for complete parametric evaluation. Parameterization A simple class defining the available parameterizations of faces and for identifying that of a particular face. Tessellation A simple class providing information about a specified tessellation pattern for a given Parameterization. Bfr is well suited to cases where evaluation of the mesh may be sparse, dynamically determined or iterative Newton, gradient descent, etc. It is not intended to replace the cases for which Far has been designed i.e. repeated evaluation of a fixed set of points but is intended to complement them. While simplicity, flexibility and reasonable performance were the main goals of Bfr , its current implementation often outperforms the table-based solutions of Far for many common use cases -- both in terms of execution time and memory use. An area that Bfr does not address, and where Far remains more suited, is capturing a specific representation of the limit surface for external use. Bfr intentionally keeps internal implementation details private to allow future improvements or extensions. Those representation details may be publicly exposed in future releases, but until then, use of Far is required for such purposes. Evaluation Since subdivision surfaces are piecewise parametric surfaces, the main operation of interest is evaluation. Bfr deals with the limit surface of a mesh as a whole by associating pieces of surface with each face of the mesh. These pieces of surface are referred to in the context of Bfr simply as surfaces and represented by BfrSurface. Each face of the mesh has an implicit local 2D parameterization and individual coordinates of that parameterization are used to evaluate its corresponding Surface. In general, 3- and 4-sided faces use the same parameterizations for quad and triangular patches used elsewhere in OpenSubdiv Parameterizations are defined for other faces more details to follow, so Surfaces for all faces can be evaluated given any 2D parametric coordinate of its face. Given an instance of a mesh, usage first requires the creation of a BfrSurfaceFactory corresponding to that mesh -- from which Surfaces can then be created for evaluation. Construction of the SurfaceFactory involves no pre-processing and Surfaces can be created and discarded as needed. The processes of constructing and evaluating Surfaces are described in more detail below. BfrSurfaceFactory Construction of BfrSurfaces requires an instance of BfrSurfaceFactory. An instance of SurfaceFactory is a light-weight interface to an instance of a mesh that requires little to no construction cost or memory. The SurfaceFactory does no work until a Surface is requested for a particular face -- at which point the factory inspects the mesh topology around that face to assemble the Surface. SurfaceFactory is actually a base class that is inherited to provide a consistent construction interface for Surfaces. Subclasses are derived to support a particular class of connected mesh -- to implement the topology inspection around each face required to construct the Surface. Use of these subclasses is very simple given the public interface of SurfaceFactory, but defining such a subclass is not. That more complex use case of SurfaceFactory will be described in detail later with other more advanced topics. In many cases, it is not necessary to explicitly define a subclass of SurfaceFactory, as the tutorials for Bfr illustrate. If already using OpenSubdiv for other reasons, a FarTopologyRefiner will have been constructed to represent the initial base mesh before refinement. Bfr provides a subclass of SurfaceFactory using FarTopologyRefiner as the base mesh ignoring any levels of refinement for immediate use in such cases. For those cases when no connected mesh representation is available at all i.e. only raw, unconnected mesh data exists construction of a FarTopologyRefiner provides a reasonably efficient connected mesh representation see the Far tutorials for construction details, whose provided subclass for SurfaceFactory is then readily available. Given the different interpolation types for mesh data i.e. vertex, varying and face-varying, the common interface for SurfaceFactory provides methods to construct Surfaces explicitly for all data types. So for positions, the methods for vertex data must be used to obtain the desired Surface, while for texture coordinates the methods for face-varying are usually required, e.g. Surface CreateVertexSurface Index faceIndex const Surface CreateVaryingSurface Index faceIndex const Surface CreateFaceVaryingSurface Index faceIndex const The Surfaces created by these construction methods may all be distinct as the underlying representations of the Surfaces and the indices of the data that define them will often differ. For example, the position data may require a bicubic patch while the face-varying texture data may be linear or a different type of bicubic patch given the different interpolation rules for face-varying and the possibility of seams. While the internal representations of the Surfaces constructed for different data interpolation types may differ, since they are all constructed as Surfaces, the functionality used to evaluate them is identical. BfrSurface The Surface class encapsulates the piece of limit surface associated with a particular face of the mesh. The term surface is used rather than patch to emphasize that the Surface may itself be a piecewise parametric surface composed of more than one patch potentially even a complex set of patches. Surface is also a class template selected by floating point precision, and so typically declared as BfrSurfacefloat. Just as a simpler type name is likely to be declared when used, the simple name Surface will be used to refer to it here. And where code fragments may be warranted, float will be substituted for the template parameter for clarity. Once created, there are two steps required to evaluate a Surface preparation of associated data points from the mesh the actual calls to evaluation methods using these data points The latter is straight-forward, but the former warrants a little more explanation. The shape of a Surface for a base face is influenced by the set of data points associated with both the vertices of the face and a subset of those in its immediate neighborhood. These control points are identified when the Surface is initialized and are publicly available for inspection if desired. The control points are sufficient to define the Surface if the face and its neighborhood are regular, but any irregularity an extra-ordinary vertex, crease, etc. usually requires additional, intermediate points to be computed from those control points in order to evaluate the Surface efficiently. Having previously avoided use of the term patch in favor of surface, the term patch points is now used to refer to these intermediate points. Patch points always include the control points as a subset and may be followed by points needed for any additional patches required to represent a more complex Surface. While the patch points are assembled in a local array for direct use by the Surface, the control points can either be gathered and accessed locally or indexed from buffers associated with the mesh for other purposes e.g. computing a bounding box of the Surface Once the patch points for a Surface are prepared, they can be passed to the main evaluation methods with the desired parametric coordinates. As previously noted, since the Surface class is a template for floating point precision, evaluation is supported in single or double precision by constructing a Surface for the desired precision. Evaluation methods are overloaded to obtain simply position or including all first or second derivatives. So preparation and evaluation can be achieved with the following Preparing patch points void PreparePatchPoints float const meshPoints , PointDescriptor meshPointDescriptor , float patchPoints , PointDescriptor patchPointDescriptor const Evaluating position and 1st derivatives void Evaluate float const uv 2 , float const patchPoints , PointDescriptor patchPointDescriptor , float P , float dPdu , float dPdv const The PointDescriptor class here is a simple struct defining the size and stride of the associated array of points. Any use of mesh points, control points or patch points generally requires an accompanying descriptor. Depending on the complexity of the limit surface, this preparation of patch points can be costly -- especially if only evaluating the Surface once or twice. In such cases, it is worth considering evaluating limit stencils, i.e. sets of coefficients that combine the original control vertices of the mesh without requiring the computation of intermediate values. The cost of evaluating stencils is considerably higher than direct evaluation, but that added overhead is often offset by avoiding the use of patch points. Surfaces should be considered a class for transient use as retaining them for longer term usage can reduce their benefits. The relatively high cost of initialization of irregular Surfaces can be a deterrent and often motivates their retention despite increased memory costs. Retaining all Surfaces of a mesh for random sampling is a situation that should be undertaken with caution and will be discussed in more detail later with other advanced topics. Parameterization Each face of a mesh has an implicit local 2D parameterization whose 2D coordinates are used to evaluate the Surface for that face. Bfr adopts the parameterizations defined elsewhere in OpenSubdiv for quadrilateral and triangular patches, for use quadrilateral and triangular faces But the parameterization of a face is also dependent on the subdivision scheme applied to it. Subdivision schemes that divide faces into quads are ultimately represented by quadrilateral patches. So a face that is a quad can be parameterized as a single quad, but other non-quad faces are parameterized as a set of quad sub-faces, i.e. faces resulting from subdivision A triangle subdivided with a quad-based scheme e.g. Catmull-Clark will therefore not have the parameterization of the triangular patch indicated previously, but another defined by its quad sub-faces illustrated above to be described in more detail below. Subdivision schemes that divide faces into triangles are currently restricted to triangles only, so all faces are parameterized as single triangles. If Loop subdivision is extended to non-triangles in future, a parameterization involving triangular sub-faces will be necessary. Note that triangles are often parameterized elsewhere in terms of barycentric coordinates u,v,w where w 1 - u - v . As is the case elsewhere in OpenSubdiv, Bfr considers parametric coordinates as 2D u,v pairs for all purposes. All faces have an implicit 2D local parameterization and all interfaces requiring parametric coordinates consider only the u,v pair. If interaction with some other tool set requiring barycentric coordinates for triangles is necessary, it is left to users to compute the implicit w accordingly. BfrParameterization BfrParameterization is a simple class that fully defines the parameterization for a particular face. An instance of Parameterization is fully defined on construction given the size of a face and the subdivision scheme applied to it where the face size is its number of verticesedges. Since any parameterization of N -sided faces requires N in some form, the face size is stored as a member and made publicly available. Each Surface has the Parameterization of its face assigned internally as part of its construction, and that is used internally by the Surface in many of its methods. The need to deal directly with the explicit details of the Parameterization class is not generally necessary. Often it is sufficient to retrieve the Parameterization from a Surface for use in some other context e.g. passed to BfrTessellation. The enumerated type ParameterizationType currently defines three kinds of parameterizations -- one of which is assigned to each instance on construction according to the properties of the face QUAD Applied to quadrilateral faces with a quad-based subdivision scheme e.g. Catmark or Bilinear. TRI Applied to triangular faces with a triangle-based subdivision scheme e.g. Loop. QUAD_SUBFACES Applied to non-quad faces with a quad-based subdivision scheme -- dividing the face into quadrilateral sub-faces. Parameterizations that involve subdivision into sub-faces, e.g. QUAD_SUBFACES, may warrant some care as they are not continuous. Depending on how they are defined, the sub-faces may be disjoint e.g. Bfr or overlap in parametric space e.g. Ptex. To help these situations, methods to detect the presence of sub-faces and deal with their local parameterizations are made available. Discontinuous Parameterizations When a face does not have a regular parameterization, the division of the parameterization into sub-faces can create complications -- as noted and addressed elsewhere in OpenSubdiv. BfrParameterization defines a quadrangulated sub-face parameterization differently from the Far and Osd interfaces. For an N -sided face, Far uses a parameterization adopted by Ptex. In this case, all quad sub-faces are parameterized over the unit square and require an additional index of the sub-face to identify them. So Ptex coordinates require three values the index and u,v of the sub-face. To embed sub-face coordinates in a single u,v pair, Bfr tiles the sub-faces in disjoint regions in parameter space. This tiling is similar to the Udim convention for textures, where a UDim on the order of sqrtN is used to preserve accuracy for increasing N Note also that the edges of each sub-face are of parametric length 0.5, which results in a total parametric length of 1.0 for all base edges. This differs again from Ptex, which parameterizes sub-faces with edge lengths of 1.0, and so can lead to inconsistencies in parametric scale typically with derivatives across edges of the mesh if not careful. As previously mentioned, care may be necessary when dealing with the discontinuities that exist in parameterizations with sub-faces. This is particularly true if evaluating data at sampled locations of the face and needing to evaluate at other locations interpolated from these. Interpolation between parametric locations, e.g. A, B and C, should be avoided when discontinuous. In many cases, dealing directly with coordinates of the sub-faces is unavoidable, e.g. interpolating Ptex coordinates for sampling of textures assigned explicitly to the sub-faces. Methods are provided to convert from Bfr s tiled parameterization to and from other representations that use a local parameterization for each sub-face. Tessellation Once a Surface can be evaluated it can be tessellated. Given a 2D parameterization, a tessellation consists of two parts a set of parametric coordinates sampling the Parameterization a set of faces connecting these coordinates that covers the entire Parameterization Once evaluated, the resulting set of sample points and the faces connecting them effectively define a mesh for that parameterization. For the sake of brevity both here and in the programming interface, the parametric coordinates or sample points are referred to simply as coords or Coords -- avoiding the term points, which is already a heavily overloaded term. Similarly the faces connecting the coords are referred to as facets or Facets -- avoiding the term face to avoid confusion with the base face of the mesh being tessellated. Bfr provides a simple class to support a variety of tessellation patterns for the different Parameterization types and methods for retrieving its associated coords and facets. In many cases the patterns they define are similar to those of GPU hardware tessellation -- which may be more familiar to many -- but they do differ in several ways, as noted below. BfrTessellation In Bfr a Tessellation is a simple class defined by a Parameterization and a given set of tessellation rates and a few additional options. These two elements define a specific tessellation pattern for all faces sharing that Parameterization. An instance of Tessellation can then be inspected to identify all or subsets of its coords or facets. The process of tessellation in other contexts usually generates triangular facets, but that is not the case with Bfr . While producing triangular facets is the default, options are available to have Tessellation include quads in patterns for parameterizations associated with quad-based subdivision schemes. For simple uniform patterns, these produce patterns that are similar in topology to those resulting from subdivision Tessellation of 4- and 5-sided faces of a quad-based scheme using quadrilateral facets left and triangular right The name Tessellation was chosen rather than Tessellator as it is a passive class that simply holds information define its pattern. It doesnt do much other than providing information about the pattern when requested. A few general properties about the pattern are determined and retained on construction, after which an instance is immutable. So it does not maintain any additional state between queries. In order to provide flexibility when dealing with tessellations of adjacent faces, the coords arising from an instance of Tessellation are ordered and are retrievable in ways to help identify points along edges that may be shared between two or more faces. The coords of a Tessellation are generated in concentric rings, beginning with the outer ring and starting with the first vertex Ordering of coords around boundary for quad and tri parameterizations. Methods of the Tessellation class allow the coords associated with specific vertices or edges to be identified, as well as providing the coords for the entire ring around the boundary separately from those of the interior if desired. While the ordering of coords in the interior is not defined and so not to be relied upon, the ordering of the boundary coords is specifically fixed to support the correlation of potentially shared coords between faces. The Tessellation class is completely independent of the Surface class. Tessellation simply takes a Parameterization and tessellation rates and provides the coords and facets that define its pattern. So Tessellation can be used in any other evaluation context where the Parameterizations are appropriate. Tessellation Rates For a particular Parameterization, the various tessellation patterns are determined by one or more tessellation rates. The simplest set of patterns uses a single tessellation rate and is said to be uniform, i.e. all edges and the interior of the face are split to a similar degree Uniform tessellation of a quadrilateral and triangle with rates of 5 and 8. More complex non-uniform patterns allow the edges of the face to be split independently from the interior of the face. Given rates for each edge, a suitable uniform rate for the interior can be either inferred or specified explicitly. These are typically referred to as the outer rates and the inner rate. The single rate specified for a simple uniform tessellation is essentially the specification of a single inner rate while the outer rates for all edges are inferred as the same. Non-uniform tessellation of a quadrilateral, triangle and 5-sided face with various outer and inner rates. In the case of Parameterizations for quads, it is common elsewhere to associate two inner rates with the opposing edges. So two separate inner rates are available for quad parameterizations -- to be specified or otherwise inferred Quad tessellations with differing inner rates with matching left and varying outer rates right. Differences from Hardware Tessellation Since the specifications for hardware tessellation often leave some details of the patterns as implementation dependent, no two hardware implementations are necessarily the same. Typically there may be subtle differences in the non-uniform tessellation patterns along boundaries, and that is to be executed here. Bfr does provide some obvious additional functionality not present in hardware tessellation and vice versa, e.g. Bfr provides the following not supported by hardware tessellation patterns for parameterizations other than quads and tris e.g. N-sided preservation of quad facets of quad-based parameterizations while hardware tessellation provides the following not supported by Bfr patterns for so-called fractional tessellation non-integer rates The lack of fractional tessellation in Bfr is something that may be addressed in a future release. Where the functionality of Bfr and hardware tessellation overlap, a few other differences are worth noting indexing of edges and their associated outer tessellation rates uniform tessellation patterns for triangles differ significantly For the indexing of edges and rates, when specifying an outer rate associated with an edge, the array index for rate i is expected to correspond to edge i . Bfr follows the convention established elsewhere in OpenSubdiv of labelingindexing edges 0, 1, etc. between vertex pairs 0,1, 1,2, etc. So outer rate 0 corresponds to the edge between vertices 0,1. In contrast, hardware tessellation associates the rate for the edge between vertices 0,1 as outer rate 1 -- its outer rate 0 is between vertices N-1,0. So an offset of 1 is warranted when comparing the two. Outer edge tessellation rates of 1,3,5,7 applied to a quad with Bfr left and GPU tessellation right. For the uniform tessellation of triangles, its well known that the needs of hardware implementation led designers to factor the patterns for triangles to make use of the same hardware necessary for quads. As a result, many edges are introduced into a simple tessellation of a triangle that are not parallel to one of its three edges. Bfr uses patterns more consistent with those resulting from the subdivision of triangles. Only edges parallel to the edges of the triangle are introduced, which creates more uniform facets both edge lengths and area and reduces their number by one third. This can reduce artifacts that sometimes arise with use of the hardware patterns at lower tessellation rates Uniform tessellation of a triangle with Bfr left and GPU tessellation right. These triangular patterns have been referred to as integer spacing for triangular patches in early work on hardware tessellation. But use of these patterns was generally discarded in favor of techniques that split the triangle into three quads -- allowing the hardware solution for quad tessellation to be reused. More on BfrSurfaceFactory The primary function of BfrSurfaceFactory is to identify and construct a representation of the limit surface for a given face of a mesh. It achieves this by inspecting the topology around the given face and constructing a suitable representation encapsulated in a Surface. The regions around a face can be divided into two categories based on their topology those that are regular and those that are not, i.e. those that are irregular. Recalling the illustration from Irregular versus Irregular Features Patches of regular Surfaces Potential patches of irregular Surfaces The representation of the limit surface for regular regions is trivial -- it is a single parametric patch whose basis is determined by the subdivision scheme e.g. uniform bicubic B-spline for Catmull-Clark. In contrast, the representation of the limit surface for an irregular region cannot be accurately represented so simply. It can be far more complex depending on the features present extra-ordinary vertices, creasing of edges, etc.. It may be as simple as a different kind of parametric patch whose points are derived from those of the mesh, or it may often be a set of patches in a hierarchy resulting from local subdivision. Bfr intentionally hides the details of these representations to allow future improvement. The cost of determining and assembling the representations of irregular Surfaces is therefore often significant. Some of the performance benefits of the SurfaceFactory are achieved by having it cache the complexities of the irregular surfaces that it encounters. In many common use cases, awareness and management of this caching is not necessary as illustrated by the tutorials. But the thread-safe construction of Surfaces is one area where some awareness is required. Other use cases that share the cache between meshes are also worth exploring as they can further reduce potentially significant costs. BfrSurfaceFactoryCache The SurfaceFactoryCache is the class used by SurfaceFactory to cache the topological information that it can reuse for other similarly irregular faces of the mesh. Though it is a publicly accessible class, the SurfaceFactoryCache has little to no public interface other than construction made available to support more advanced cases covered later and in most cases it can be completely ignored. Typically an instance of SurfaceFactory has an internal SurfaceFactoryCache member which is used by that factory for its lifetime. Since that cache member is mutable -- potentially updated when an irregular Surface is created -- it does need to be thread-safe if the SurfaceFactory is to be used in a threaded context. To accommodate this need, SurfaceFactoryCache is defined as a base class with an accompanying class template to allow the trivial declaration of thread-safe subclasses template typename MUTEX_TYPE , typename READ_LOCK_GUARD_TYPE , typename WRITE_LOCK_GUARD_TYPE class SurfaceFactoryCacheThreaded public SurfaceFactoryCache ... For example, a local type for a thread-safe cache using stdshared_mutex from C17 could be simply declared as follows include shared_mutex typedef Bfr SurfaceFactoryCacheThreaded std shared_mutex , std shared_lock std shared_mutex , std unique_lock std shared_mutex ThreadSafeCache Such thread-safe cache types are essential when distributing the work of a single SurfaceFactory across multiple threads. They can be encapsulated in the definitions of subclasses of SurfaceFactory or used to define external cache instances for use with any subclass of SurfaceFactory. Defining a Thread-Safe SurfaceFactory The thread-safety of a SurfaceFactory is purely dependent on the thread-safety of the SurfaceFactoryCache that it uses. With caching disabled, any SurfaceFactory is thread-safe but will be far less efficient in dealing with irregular Surfaces. When a subclass of SurfaceFactory is defined discussed in more detail later, one of its responsibilities is to identify and manage an instance of SurfaceFactoryCache for its internal use. Defining such a subclass is a simple matter of declaring a thread-safe SurfaceFactoryCache type as noted above along with a local member of that type to be used by each instance. Given the widespread use of the FarTopologyRefiner in OpenSubdiv, and the lack of a connected mesh representation in many contexts, a subclass of SurfaceFactory is made available to use a TopologyRefiner as a mesh, i.e. the BfrRefinerSurfaceFactory subclass. Since many OpenSubdiv users may make use of the RefinerSurfaceFactory subclass, and they may have different preferences of threading model, the RefinerSurfaceFactory subclass is similarly defined as a class template to enable threading flexibility. In this case, the template is parameterized by the desired type of SurfaceFactoryCache, which embodies the threading specifications as noted above, i.e. template class CACHE_TYPE SurfaceFactoryCache class RefinerSurfaceFactory public ... ... The default template is the base SurfaceFactoryCache which is not thread-safe, but a simple declaration of a thread-safe cache type is sufficient to declare a similarly thread-safe RefinerSurfaceFactory type include opensubdivbfrsurfaceFactoryCache.h Declare thread-safe cache type see stdshared_mutex example above typedef Bfr SurfaceFactoryCacheThreaded ... ThreadSafeCache Declare thread-safe factory type typedef Bfr RefinerSurfaceFactory ThreadSafeCache ThreadSafeFactory The resulting factory type safely allows the construction of Surfaces and their subsequent evaluation and tessellation to be distributed over multiple threads. Internal versus External SurfaceFactoryCache Typical usage of the SurfaceFactoryCache by the SurfaceFactory is to have the factory create an internal cache member to be used for the lifetime of the factory associated with a mesh. But the data stored in the cache is not in any way dependent on the factory or mesh used to create it. So a cache can potentially be shared by multiple factories. While such sharing is possible -- and the Bfr interfaces intentionally permit it -- any exploration should proceed with caution. Greater public knowledge and control of the cache is ultimately necessary to manage its potentially unbounded memory increase, and support in the public interface is currently limited. A cache stored as a member varialbe and managed exclusively by the factory is said to be internal while one managed exclusively by its client is said to be external. In both cases, the factory deals with retrieving data from or adding data to the cache -- only management of the caches ownership differs, and that ownership is never transferred. A subset of the methods of SurfaceFactoryOptions provide the means of specifying the use of an internal or external cache, or no caching at all Assign an external cache to override the internal Options SetExternalCache SurfaceFactoryCache cache Enable or disable caching default is true Options EnableCaching bool on As noted here, specifying an external cache will override use of a factorys internal cache. Disabling caching takes precedence over both, but is generally not practical and exists mainly to aide debugging. The common use of the internal cache is to create a SurfaceFactory and distribute processing of the Surfaces of its faces over multiple threads, or to construct Surfaces for the mesh for any other purpose while the mesh remains in scope. There is no need to deal explicitly with the SurfaceFactoryCache in these cases. Use cases for an external cache are more varied and explicit, including creating a single external cache to process a sequence of meshes on a single thread cache thread-safety not required creating a separate external cache on each thread to process a set of meshes distributed over multiple threads cache thread-safety not required creating a single external cache for multiple meshes distributed over multiple threads cache thread-safety required, and beware of unbounded memory growth here Future extensions to the public interface of SurfaceFactoryCache may be made to support common use cases as their common needs are made clearer. Customizing a BfrSurfaceFactory One of the goals of Bfr is to provide a lightweight interface for the evaluation of Surfaces from any connected mesh representation. In order to do so, the factory needs to gather topological information from that mesh representation. That information is provide to the factory through inheritance a subclass of SurfaceFactory is defined that fulfills all requirements of the factory. It must be made clear that a subclass can only be created from a connected mesh representation, i.e. a representation that includes connectivity or adjacency relationships between its components vertices, faces and edges. Classes for simple containers of mesh topology used for external formats e.g. USD, Alembic, etc. are generally not connected . Many applications construct a connected mesh representation for internal use when loading such mesh data -- using a variety of techniques including half-edges, winged-edges or table-based relationships. There are many choices here that offer a variety of trade-offs depending on usage e.g. fixed vs dynamic topology and so no best solution. Once constructed and available within an application, Bfr strives to take advantage of that representation. As a minimum requirement for supporting a subclass of SurfaceFactory, a connected mesh representation must be able to efficiently identify the incident faces of any given vertex. As noted earlier, when no such representation is available, users can construct a FarTopologyRefiner for their connected mesh and use BfrRefinerSurfaceFactory. There are three requirements of a subclass of SurfaceFactory fulfill the interface required to adapt the connected mesh to the factory provide an internal cache for the factory of the preferred type extend the existing SurfaceFactory interface for the connected mesh type The first of these is the most significant and is the focus here. The second was mentioned previously with the SurfaceFactoryCache and is trivial. The last should also be trivial and is generally optional at minimum the subclass will need a constructor to create an instance of the factory from a given mesh, but anything more is not strictly essential. It is important to note that anyone attempting to write such a subclass must have an intimate understanding of the topological capabilities and limitations of the mesh representation involved. The SurfaceFactory is topologically robust in that it will support meshes with a wide range of degenerate or non-manifold features, but in order to process topology efficiently, a subclass needs to indicate when and where those degeneracies may occur. A simplified implementation of the BfrRefinerSurfaceFactory is provided in the tutorials for illustration purposes. The BfrSurfaceFactoryMeshAdapter Interface The SurfaceFactoryMeshAdapter class defines the interface used to satisfy the topological requirements of the SurfaceFactory. An implementation for a particular mesh class provides the base factory with everything needed to identify the limit surface of a given face from its surrounding topology. The SurfaceFactory actually inherits the SurfaceFactoryMeshAdapter interface but does not implement it -- deferring that to its subclasses -- since separate subclasses of SurfaceFactoryMeshAdapter serve no other purpose. The limit surface for a face is fully defined by the complete set of incident vertices, faces and edges surrounding the face. But it is difficult to accurately and efficiently assemble and represent all of that required information in a single class or query for all possible cases. So the mesh adapter interface provides a suite of methods to allow the factory to gather only what it needs for the Surface required -- which may differ considerably according to whether the Surface is for vertex or face-varying data, linear or non-linear, etc. The virtual methods required can be organized into small groups devoted to particular aspects of construction. A description of the methods and purposes for each group follows, with more details and exact signatures available in the accompanying Doxygen for the SurfaceFactoryMeshAdapter class. Basic Properties of a Face A small set of simple methods indicate whether the SurfaceFactory needs to create a Surface for a face, and if so, how virtual bool isFaceHole Index faceIndex const 0 virtual int getFaceSize Index faceIndex const 0 These are trivial and self-explanatory. Identifying Indices for an Entire Face If the Surface requested turns out to be linearly interpolated e.g. for varying or linear face-varying data indices for the control point data are all assigned to the face and can be trivially identified virtual int getFaceVertexIndices Index faceIndex , Index vertexIndices const 0 virtual int getFaceFVarValueIndices Index faceIndex , FVarID faceVaryingID , Index faceVaryingIndices const 0 Since multiple sets of face-varying data with different topology may be assigned to the mesh, an identifier needs to be specified both in the public interface when requesting a Surface and here when the factory assembles it. How a face-varying identifier is interpreted is completely determined by the subclass through the implementation of the methods that require it. Specifying the Neighborhood Around a Vertex When the Surface requested is not linear, the entire neighborhood around the face must be determined. This is achieved by specifying the neighborhoods around each of the vertices of the face, which the factory then assembles. For the neighborhood of each face-vertex, the factory obtains a complete specification in a simple VertexDescriptor class. An instance of VertexDescriptor is provided and populated with the following method virtual int populateFaceVertexDescriptor Index faceIndex , int faceVertex , VertexDescriptor vertexDescriptor const 0 Within this method, the given VertexDescriptor instance is initialized using a small suite of VertexDescriptor methods that specify the following information about the vertex and its neighborhood whether the neighborhood is manifold ordered counter-clockwise whether the vertex is on a boundary the sizes of all or each incident face the sharpness of the vertex the sharpness of edges of incident faces These methods are specified between Initialize and Finalize methods, so an interior vertex of valence 4 with three incident quads and one incident triangle might be specified as follows int vertexValence 4 vertexDescriptor . Initialize vertexValence vertexDescriptor . SetManifold true vertexDescriptor . SetBoundary false vertexDescriptor . SetIncidentFaceSize 0 , 4 vertexDescriptor . SetIncidentFaceSize 1 , 4 vertexDescriptor . SetIncidentFaceSize 2 , 3 vertexDescriptor . SetIncidentFaceSize 3 , 4 vertexDescriptor . Finalize Specifying the vertex neighborhood as manifold is critical to allowing the factory to inspect the neighborhood efficiently. A manifold vertex has its incident faces and edges ordered in a counter-clockwise orientation and is free of degeneracies. If it is not clear that a vertex is manifold, it should not be set as such or the factorys inspection of associated data will not be correct. Identifying Indices Around a Vertex When the Surface requested is not linear, the indices of control point data for the entire neighborhood of the face are ultimately required, and that entire set is similarly determined by identifying the indices for each of the neighborhoods of the face-vertices virtual int getFaceVertexIncidentFaceVertexIndices Index faceIndex , int faceVertex , Index vertexIndices const 0 virtual int getFaceVertexIncidentFaceFVarValueIndices Index faceIndex , int faceVertex , FVarID faceVaryingID , Index faceVaryingIndices const 0 As was the case with the methods retrieving indices for the entire face, one exists for identifying indices vertex data while another exists to identify indices for a specified set of face-varying data. Customizing the Subclass Interface Once the topological requirements of a subclass have been satisfied for its mesh representation, minor customizations of the inherited interface of SurfaceFactory may be useful. Consider a class called Mesh and its associated subclass of SurfaceFactory called MeshSurfaceFactory. At minimum, a constructor of MeshSurfaceFactory is necessary to construct an instance for a particular instance of mesh. This is typically achieved as follows MeshSurfaceFactory Mesh const mesh , Options const options In addition to the Mesh instance, such a constructor passes a set of Options i.e. SurfaceFactoryOptions to the base SurfaceFactory. Any additional arguments are possible here, e.g. perhaps only a single face-varying UV set is supported, and that might be specified by identifying it on construction. Given that mesh representations often have their own associated classes that internally contain the actual data, it may be useful to provide a few other conveniences to simplify working with a Mesh. For example, if mesh data is stored in a class called MeshPrimvar, a method to construct a Surface from a given MeshPrimvar may be useful, e.g. bool InitPrimvarSurface int faceIndex , MeshPrimvar const meshPrimvar , Surface float surface which would then determine the nature of the MeshPrimvar data interpolated as vertex, varying or face-varying and act accordingly. It may also be worth simplifying the template complexity here if only one precision is ever required. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"title\": \"Contributing to OpenSubdiv\",\n+ \"text\": \"Contributing to OpenSubdiv Contributing to OpenSubdiv 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Contributor License Agreement Understand Git Recommended Git Workflow Sending a Pull Request Contributor License Agreement In order for us to accept code submissions merge git pull-requests, contributors need to sign the Contributor License Agreement CLA. There are two CLAs, one for individuals and one for corporations. As for the end-user license, both are based on Apache. They are found in the code repository individual form , corporate form . Please email the signed CLA to opensubdiv-clapixar.com . Understand Git First, you should familiarize yourself with the Git data model and commands. For small changes you may not need to understand Git deeply, but for larger changes or working with the codebase over a long period of time, it becomes critical to understand more of whats going on under the hood. There are many free resources on the internet, one which weve found useful is the following e-book httpsgithub.compluralsightgit-internals-pdfreleases Recommended Git Workflow Once you have a local development tree cloned and working, you can start making changes. You will need to integrate changes from the source tree as you work the following outlines the workflow used by core OpenSubdiv engineers at Pixar and DreamWorks Fork the repository into your own local copy. This can be done via the GitHub website using the fork button. Clone your fork locally git clone your_fork_url OpenSubdiv.your_name e.g. git clone httpsgithub.comyourusernameOpenSubdiv.git OpenSubdiv.yourusername Setup two remotes, origin and upstream . Origin will be setup as a result of cloning your remote repository, but upstream must be setup manually git remote add upstream httpsgithub.comPixarAnimationStudiosOpenSubdiv.git Verify your remotes are setup correctly git remote -v Which should look something like origin httpsgithub.comyourusernameOpenSubdiv.git fetch origin httpsgithub.comyourusernameOpenSubdiv.git push upstream httpsgithub.comPixarAnimationStudiosOpenSubdiv.git fetch upstream httpsgithub.comPixarAnimationStudiosOpenSubdiv.git push Finally, fetch the upstream content this is required for the next step git fetch upstream Setup a new branch for each change. Working with branches in Git is its greatest pleasure, we strongly suggest setting up a new branch for each change which you plan to pull-request. All work is done in the dev branch, so be sure to keep your change in sync with this upstream branch. To begin, start your new branch from the dev branch git checkout -b dev-feature upstreamdev As you are working on your feature, new changes will be merged into the upstream repository, to sync these changes down and preserve your local edits, you can continually rebase your local work git pull --rebase upstream dev Notice the --rebase option here. It updates the current branch to the upstreamdev branch and rebases all edits so they are at the head of your local feature branch. Alternatively, you can rebase all your work at once when your feature is complete. Sending a Pull Request First, rebase and squash your changes appropriately to produce a clean set of changes at the head of your tree. We require changes to be grouped locally to ensure that rolling back changes can be done easily. If youve followed the steps above, your pending change should already be queued up as required. If you have not, you may need to rebase and squash changes at this point. Once the change is clean, push your changes to origin and go to the GitHub website to submit your pull request. Be sure to submit your request against the dev branch. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n- \"loc\": \"bfr_overview.html\"\n+ \"loc\": \"contributing.html\"\n+ }, {\n+ \"title\": \"Subdivision Compatibility\",\n+ \"text\": \"Subdivision Compatibility Subdivision Compatibility 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Subdivision Compatibility Compatibility with OpenSubdiv 2.x Compatibility with RenderMan Incompatibilities Other Differences Subdivision Compatibility This document highlights areas of compatibility with other software that makes use of subdivision surfaces, including previous versions of OpenSubdiv. The compatibility here refers to the choice of subdivision rules that define the shape of the resulting surfaces. Different subdivision rules will lead to different shapes. Choices affecting shape include the types of subdivision schemes supported e.g. Catmull-Clark, Loop, etc. the basic rules applied for these schemes any extended rules to affect sharpness or creasing rules applied separately to face-varying data Ensuring all of these rules are consistent provides the basis for consistent shapes, but further approximations to the limit surface create the potential for subtle deviations. Even within OpenSubdiv, multiple approximations are possible and vary. For now we focus on the compatibility of subdivision rules and deal with the limit approximations only when noteworthy. Compatibility with OpenSubdiv 2.x The refactoring of OpenSubdiv 3.0 data representations presented a unique opportunity to revisit some corners of the subdivision specification and remove or update some legacy features. Face-varying Interpolation Options Face-varying interpolation options have been consolidated into a single enum with one additional choice new to 3.0. No functionality from 2.x has been removed -- just re-expressed in a simpler and more comprehensible form. Face-varying interpolation was previously defined by a boundary interpolation enum with four modes and an additional boolean propagate corners option, which was little understood, i.e. void HbrMeshSetFVarInterpolateBoundarMethodInterpolateBoundaryMethod const void HbrMeshSetFVarPropagateCornersbool const The latter was only used in conjunction with one of the four modes edge and corner, so it was effectively a unique fifth choice. Closer inspection of all of these modes also revealed some unexpected and undesirable behavior in some common cases -- to an extent that could not simply be changed -- and so an additional mode was added to avoid such behavior. All choices are now provided through a single linear interpolation enum, described and illustrated in more detail in the overview of Face-Varying Interpolation . The use of boundary in the name of the enum was intentionally removed as the choice also affects interior interpolation. The new use of linear is now intended to reflect the fact that interpolation is constrained to be linear where specified by the choice applied. All five of Hbrs original modes of face-varying interpolation are supported with minor modifications where Hbr was found to be incorrect in the presence of semi-sharp creasing. An additional mode corners only has also been added to avoid some of the undesired side-effects of some existing modes illustrated below. The new values for the SdcOptionsFVarLinearInterpolation enum and its equivalents for HbrMeshs InterpolateBoundaryMethod and PropagateCorners flag are as follows ordered such that the set of linear constraints applied is always increasing -- from completely smooth to completely linear Sdc FVarLinearInterpolation Hbr FVarInterpolateBoundaryMethod Hbr FVarPropogateCorners FVAR_LINEAR_NONE k_InterpolateBoundaryEdgeOnly NA ignored FVAR_LINEAR_CORNERS_ONLY NA NA FVAR_LINEAR_CORNERS_PLUS1 k_InterpolateBoundaryEdgeAndCorner false FVAR_LINEAR_CORNERS_PLUS2 k_InterpolateBoundaryEdgeAndCorner true FVAR_LINEAR_BOUNDARIES k_InterpolateBoundaryAlwaysSharp NA ignored FVAR_LINEAR_ALL k_InterpolateBoundaryNone NA ignored Aside from the two corners plus modes that preserve Hbr behavior, all other modes are designed so that the interpolation of a disjoint face-varying region is not affected by changes to other regions that may share the same vertex. So the behavior of a disjoint region should be well understood and predictable when looking at it in isolation e.g. with corners only one would expect to see linear constraints applied where there are topological corners or infinitely sharp creasing applied within the region, and nowhere else. This is not true of the plus modes, and they are named to reflect the fact that more is taken into account where disjoint regions meet. The following example illustrates some undesired effects of the plus modes, which in part motivated the addition of the new corners only mode. The example uses the catmark_fvar_bound0 and catmark_fvar_bound1 shapes from the suite of regression shapes. Both shapes are a simple regular 4x4 grid of quads with face-varying UV data partitioned into multiple disjoint regions. The bound0 shape has two disjoint UV regions -- an upper and lower region -- while the bound1 shape further splits the lower region in two. This first figure illustrates the effect of the original plus1 mode which is also the same for plus2 Note that the effect of splitting the lower UV region in two has the undesired side effect of sharpening the boundary of the upper region. This is the result of the plus1 mode making collective decisions about the sharpness of all face-varying boundaries at the vertex rather than decisions local to each region. In both the plus1 and plus2 cases, all face-varying boundaries sharing a vertex will be sharpened if there are more than two regions meeting at that vertex. The second figure illustrates the effect of the new corners only mode As expected, the splitting of the lower region does not impact the upper region. In this case the decision to sharpen a face-varying boundary is made based on the local topology of each region. Vertex Interpolation Options Since the various options are now presented through a new API Sdc rather than Hbr, based on the history of some of these options and input from interested parties, the following changes have been implemented The naming of the standard creasing method has been changed from Normal to Uniform . Values for SdcOptionsCreasingMethod are now CREASE_UNIFORM standard integer subtraction per level default CREASE_CHAIKIN Chaikin non-uniform averaging around vertices Legacy modes of the smoothtriangle rule have been removed as they were never actually enabled in the code. Values for SdcOptionsTriangleSubdivision are now TRI_SUB_CATMARK Catmull-Clark weights default TRI_SUB_SMOOTH smooth triangle weights These should have little impact since one is a simple change in terminology as part of a new API while the other was removal of an option that was never used. Change to Chaikin creasing method In the process of re-implementing the Chaikin creasing method, observations lead to a conscious choice to change the behavior of Chaikin creasing in the presence of infinitely sharp edges most noticeable at boundaries. Previously, the inclusion of infinite sharpness values in the Chaikin methods computation of edge sharpness around a vertex would prevent a semi-sharp edge from decaying to zero. Infinitely sharp edges are now excluded from the Chaikin non-uniform averaging yielding a much more predictable and desirable result. For example, where the sharpness assignment is actually uniform at such a vertex, the result will now behave the same as the Uniform method. Since this feature has received little use only recently activated in RenderMan, now seemed the best time to make the change before more widespread adoption. Hierarchical Edits While extremely powerful, Hierarchical Edits come with additional maintenance and implementation complexity. Support for them in popular interchange formats and major DCC applications has either been dropped or was never implemented. As a result, the need for Hierarchical Edits is too limited to justify the cost and support for them, and they have therefore been removed from the 3.0 release of OpenSubdiv. Dropping support for Hierarchical Edits allows for significant simplifications of many areas of the subdivision algorithms. While the 3.0 release does not offer direct support for Hierarchical Edits, the architectural changes and direction of 3.0 still facilitate the application of the most common value edits for those wishing to use them -- though not always in the same optimized context. Of course, support for Hierarchical Edits in the future will be considered based on demand and resources. Non-Manifold Topology OpenSubdiv 2.x and earlier was limited to dealing with meshes whose topology was manifold -- a limitation imposed by the use of Hbr. With 3.0 no longer using Hbr, the manifold restriction has also been removed. OpenSubdiv 3.0, therefore, supports a superset of the meshes supported by 2.x and earlier versions with one known exception noted below. Non-manifold meshes that are acceptable to 3.0 however will likely not work with 2.x or earlier. The one known case that 3.0 will not represent the same as 2.x is ironically a case that is non-manifold, and for which Hbr did make special accommodation. That case occurs at a non-manifold vertex where two or more faces meet at a common vertex, but do not share a common edge, and when the boundary interpolation mode is set for smooth corners i.e. edge only, as illustrated below The cage is on the left and is refined to level 2 on the right. On the immediate right, boundary interpolation is set to sharp corners and the results appear the same for 2.x and 3.0. The center and far right illustrate the affects of setting boundary interpolation to smooth corners with 2.x and 3.0 respectively. Note that the 2.x result allows the refined mesh and so the limit surface to split into two while the 3.0 result keeps it connected. When Hbr encounters such vertices, regardless of the boundary mode it splits the vertex -- creating a separate instance of it for each face. So when building an HbrMesh, after finalizing the mesh, it will result in having more vertices than were originally defined termed split vertices. OpenSubdiv 2.x and earlier successfully hid the presence of these extra vertices from users. This case behaves in such a way that violates certain properties of the surface that 3.0 has attempted to emphasize. One of these relates to the nature of the limit surface and becomes more significant in the context of face varying if the cage is connected then so too is its limit surface, or similarly, if the cage consists of N connected regions then the limit surface similarly consists of N connected regions. Another undesirable property here is that the vertex V at which these faces meet must have more than one child vertex V . This makes it difficult to hide split vertices -- OpenSubdiv 2.x tables had an extra level of indirection that made it possible to do this relatively easily, but 3.0 has dispensed with such indirection where possible to streamline performance. Compatibility with RenderMan Since RenderMan and OpenSubdiv versions prior to 3.0 share a common library Hbr, most differences between RenderMan and OpenSubdiv 3.0 are covered in the preceding section of compatibility with OpenSubdiv 2.x. In addition to some features between RenderMan and OpenSubdiv that are not compatible, there are also other differences that may be present due to differences in the implementations of similar features. For most use cases, OpenSubdiv 3.0 is largely compatible with RenderMan. There are however some cases where some differences can be expected. These are highlighted below for completeness. Incompatibilities OpenSubdiv and RenderMan will be incompatible when certain features are used that are not common to both. They are fully described in the 2.x compatibility section and are listed briefly here. OpenSubdiv 3.0 Features Not Supported by RenderMan Non-manifold meshes Choice of the corners only face varying interpolation option RenderMan Features Not Supported by OpenSubdiv 3.0 Hierarchical Edits Other Differences Some differences can occur due to the differing implementations of the feature sets. Additionally, OpenSubdiv 3.0s implementation fixes some issues discovered in Hbr. Smooth Face-Varying Interpolation with Creasing There have been two discrepancies noted in the way that face-varying data is interpolated smoothly in the presence of creases. Smooth face-varying interpolation is expected to match vertex interpolation in the interior and only differ along the boundaries or discontinuities where the face-varying topology is intentionally made to differ from the vertex topology. A simple and effective way to identify discrepancies is to use the X and Y coordinates of vertex positions as the U and V of texture coordinates. If these U and V coordinates are assigned to a face-varying channel, smooth interpolation of U and V is expected to exactly match interpolation of X and Y, regardless of the presence of any sharpness and creasing. Two discrepancies can be seen with Hbr when superimposing the XY vertex interpolation with the projected UV face-varying interpolation. The first discrepancy occurs with interpolation around dart vertices This example shows a simple regular XY grid on the left with an interior sharp edge creating a dart vertex in the center. With no asymmetry in the vertices, the sharpness has no asymmetric affect and the XY vertex interpolation on the immediate right shows the regular grid expected from refinement. On the far right is the UV interpolation from Hbr, which exhibits distortion around the center dart vertex. The second discrepancy occurs with interpolation involving any fractional sharpness values. Hbr effectively ignores any fractional sharpness value in its face-varying interpolation. So edges of vertices with sharpness of say 2.5, will be treated as though their sharpness is 2.0 when face-varying values are interpolated. Similarly, any non-zero sharpness value less than 1.0 is treated as zero by truncation and so is essentially ignored. This example shows an asymmetric 2x2 grid of quads on the left with the center vertex progressively sharpened from 0.5 to 1.0. The three cases of the vertex smooth and sharpened are superimposed on the immediate right to display the three distinct interpolation results. On the far right the interpolation from Hbr displays the same three cases, but only two are visibly distinct -- the sharpness of 0.5 being treated the same as if it were 0.0. Both of these cases are corrected in OpenSubdiv 3.0. Smooth face-varying interpolation in the presence of creasing should match the expected behavior of the vertex interpolation, except where the face-varying topology is explicitly made to differ. The Chaikin Creasing Method At least two discrepancies are know to exist between the implementations of Hbr in RenderMan and OpenSubdiv 3.0 Use of Chaikin creasing with boundaries or infinitely sharp edges Subtle shape differences due to Hbrs use of predictive sharpness Fortunately, this feature was only recently added to Hbr and RenderMan and is little used, so it is expected these differences will have little impact. The first discrepancy is mentioned briefly in the previous section on compatibility between OpenSubdiv 2.x and 3.0. A conscious decision was made to change the averaging of sharpness values involving infinitely sharp edges in order to make results more predictable and favorable. The effects can be seen comparing the regression shape catmark_chaikin2. The second is more subtle and results from an oversight within Hbrs implementation that is not easily corrected. When determining what subdivision rule to apply from one level to the next, the sharpness values at the next level must be known in order to determine whether or not a transition between differing rules is required. If the rule at the next level differs from the previous, a combination of the two is applied. Such a change results from the sharpness values of one or more edges or the vertex itself decaying to zero. Rather than compute the sharpness values at the next level accurately, Hbr predicts it by simply subtracting 1.0 from it, as is done with the uniform creasing method, and it bases decisions on that predicted result. This does not work for Chaikin though. A sharpness value less than 1.0 may not decay to 0 if it is averaged with neighboring sharpness values greater than 1.0, so this sharpness prediction can result in the wrong rule being chosen for the next level. A typical case would have the subdivision rules for Chaikin creasing transition from Corner to Crease at one level, then from Crease to Smooth at the next. Hbrs predictive creasing might mistakenly detect the transition as Corner to Smooth at one level, then after properly computing the sharpness values for the next level later, from Crease to Smooth for the next. One of the regression shapes catmark_chakin1 was disabled from the regression suite because of this effect. The differences in shape that trigger its regression failure were investigated and determined to be the result of this issue. From observations thus far these differences are subtle but can be noticeable. Numerical Precision Since its inception, OpenSubdiv has sought to produce results that were numerically consistent to RenderMan. A regression suite to ensure a certain level of accuracy was provided to detect any substantial deviation. At some point in the development of OpenSubdiv, the point was made that numerical accuracy of Hbr could be improved by changing the order of operations and combining the vertex with the lowest coefficient first in one of the subdivision rules. This was applied more thoroughly in the independent implementation of 3.0 there seemed no reason not to. In most cases the relative magnitudes of the coefficients of subdivision and limit masks is clear so no overhead was necessary to detect them. At a certain point though, this greater accuracy came in conflict with the regression suite. It turned out that high-valence vertices could not be computed to within the desired tolerances set within the suite. The summation of many small coefficients for the adjacent vertices first, before the addition of the much larger coefficient for the primary vertex, allowed for the accumulation of precision that was being truncated by adding the much larger coefficient first in the Hbr implementation. With extremely high valence vertices, a difference in magnitude between the most and least significant coefficients of several orders of magnitude is likely, and that has a significant impact on the single-precision floating point computations. The improved accuracy of OpenSubdiv 3.0 can reach a magnitude that will not go undetected. Whether or not this can lead to visual artifacts is unclear. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"tags\": \"\",\n+ \"loc\": \"compatibility.html\"\n }, {\n \"title\": \"Building with CMake\",\n- \"text\": \"Building with CMake Building with CMake 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Overview Step 1 Dependencies Required Optional Step 2 Configuring CMake Useful Build Options Environment Variables Automated Script Using Intels C Studio XE Using Clang Step 3 Building Build Targets Compiling Linking an OpenSubdiv Application Overview Assuming that you have cloned the source repository and selected an appropriate release branch, the following instructions will walk you through the CMake and configuration and build process. CMake is a cross-platform, open-source build system. CMake controls the compilation process using platform independent configuration files in order to generate Makefiles and workspaces that are native to the platform of choice. The process involves the following steps Locate build the requisite dependencies Configure run CMake to generate Makefiles MSVC solution XCode project Run the build from make MSVC XCode Step 1 Dependencies CMake will adapt the build based on which dependencies have been successfully discovered and will disable certain features and code examples accordingly. Please refer to the documentation of each of the dependency packages for specific build and installation instructions. Required CMake version 3.12 Optional Ptex support features for ptex textures and the ptexViewer example Zlib required for Ptex under Windows CUDA TBB OpenCL DX11 SDK GLFW required for standalone examples and some regression tests Docutils required for reST-based documentation Python Pygments required for Docutils reST styling Doxygen Step 2 Configuring CMake One way to configure CMake is to use the CMake GUI . In many cases CMake can fall back on default standard paths in order to find the packages that OpenSubdiv depends on. For non-standard installations however, a complete set of override variables is available. The following sub-section lists some of these variables. For more specific details, please consult the source of the custom CMake modules in the OpenSubdivcmake folder. Useful Build Options The following configuration arguments can be passed to the CMake command line. - DCMAKE_BUILD_TYPE Debug Release - DCMAKE_INSTALL_PREFIX base path to install OpenSubdiv default Current directory - DCMAKE_LIBDIR_BASE library directory basename default lib - DCUDA_TOOLKIT_ROOT_DIR path to CUDA - DPTEX_LOCATION path to Ptex - DGLFW_LOCATION path to GLFW - DTBB_LOCATION path to Intel s TBB - DICC_LOCATION path to Intel s C Studio XE - DNO_LIB 1 disable the opensubdiv libs build caveat emptor - DNO_EXAMPLES 1 disable examples build - DNO_TUTORIALS 1 disable tutorials build - DNO_REGRESSION 1 disable regression tests build - DNO_PTEX 1 disable PTex support - DNO_DOC 1 disable documentation build - DNO_OMP 1 disable OpenMP - DNO_TBB 1 disable TBB - DNO_CUDA 1 disable CUDA - DNO_OPENCL 1 disable OpenCL - DNO_OPENGL 1 disable OpenGL - DNO_CLEW 1 disable CLEW wrapper library Environment Variables The paths to Ptex, GLFW, other dependencies can also be specified through the following environment variables PTEX_LOCATION , GLFW_LOCATION Automated Script The GUI solution will probably become a burden for active developpers who tend to re-run the configuration step fairly often. A scripted solution can save a lot of time. Here is a typical workflow git clone https github.comPixarAnimationStudiosOpenSubdiv.git folder cd folder mkdir build cd build source .. .. cmake_setup Where cmake_setup is a configuration script. Here is an example CMake configuration script for a full typical windows-based build that can be run in GitShell bintcsh Replace the .. with a full path to the root of the OpenSubdiv source tree if necessary cProgram Files x86CMake 2.8bincmake.exe - G Visual Studio 15 2017 Win64 - D GLFW_LOCATIONstringcProgram Filesglfw-2.7.7.bin.WIN64 - D OPENCL_INCLUDE_DIRSstringcProgramDataNVIDIA CorporationNVIDIA GPU Computing SDK 4.2OpenCLcommoninc - D _OPENCL_CPP_INCLUDE_DIRSstringcProgramDataNVIDIA CorporationNVIDIA GPU Computing SDK 4.2OpenCLcommoninc - D OPENCL_LIBRARIESstringcProgramDataNVIDIA CorporationNVIDIA GPU Computing SDK 4.2OpenCLcommonlibx64OpenCL.lib - D PTEX_LOCATIONstringcUsersopensubdivdemosrcptexx64 .. copy Ptex dependencies Windows only mkdir - p bin Debug , Release cp - f c Users opensubdiv demo src zlib -1.2.7 contrib vstudio vc10 x64 ZlibDllRelease zlibwapi . dll bin Debug cp - f c Users opensubdiv demo src zlib -1.2.7 contrib vstudio vc10 x64 ZlibDllRelease zlibwapi . dll bin Release cp - f c Users opensubdiv demo src ptex x64 lib Ptex . dll bin Debug cp - f c Users opensubdiv demo src ptex x64 lib Ptex . dll bin Release Important Notice that the following scripts start by recursively removing the ..build and ..inst directories. Make sure you modify them to suit your build workflow. Here is a similar script for Nix-based platforms echo Removing build cd .. rm - rf build inst mkdir build cd build echo Running cmake cmake - DPTEX_LOCATION home opensubdiv dev opensource ptex install - DGLFW_LOCATION home opensubdiv dev opensource glfw build - DDOXYGEN_EXECUTABLE home opensubdiv dev opensource doxygen inst bin doxygen - DCMAKE_INSTALL_PREFIX .. inst - DCMAKE_BUILD_TYPE Debug .. Here is a similar script for macOS echo Removing build cd .. rm - rf build inst mkdir build cd build echo Running cmake cmake - DOPENGL_INCLUDE_DIR Applications Xcode . app Contents Developer Platforms MacOSX . platform Developer SDKs MacOSX10 .9 . sdk System Library Frameworks OpenGL . framework Headers - DGLFW_LOCATION Users opensubdiv dev opensource glfw inst - DNO_OMP 1 - DNO_REGRESSION 0 - DCMAKE_INSTALL_PREFIX .. inst - DCMAKE_BUILD_TYPE Debug .. Using Intels C Studio XE OpenSubdiv can be also be built with Intels C compiler icc. The default compiler can be overriden in CMake with the following configuration options - DCMAKE_CXX_COMPILER path to icc executable - DCMAKE_C_COMPILER path to icc executable The installation location of the C Studio XE can be overriden with - DICC_LOCATION path to Intel s C Studio XE Using Clang CMake can also be overriden to use the clang compilers by configuring the following options - DCMAKE_CXX_COMPILER clang - DCMAKE_C_COMPILER clang Step 3 Building CMake provides a cross-platform command-line build cmake -- build . -- target install -- config Release Alternatively, you can use native toolkits to launch the build. The steps differ for each OS Windows launch VC with the solution generated by CMake in your build directory. macOS launch Xcode with the xcodeproj generated by CMake in your build directory Nix run make in your build directory - use the clean target to remove previous build results - use VERBOSE1 for verbose build output Build Targets Makefile-based builds allow the use of named target. Here are some of the more useful target names osd_staticdynamic_CPUGPU The core components of the OpenSubdiv libraries example_name Builds specific code examples by name glViewer, ptexViewer... doc Builds ReST and doxygen documentation doc_html Builds ReST documentation doc_doxy Builds Doxygen documentation Compiling Linking an OpenSubdiv Application Here are example commands for building an OpenSubdiv application on several architectures Linux g -IOPENSUBDIVinclude -c myapp.cpp g myapp.o -LOPENSUBDIVlib -losdGPU -losdCPU -o myapp macOS g -IOPENSUBDIVinclude -c myapp.cpp g myapp.o -LOPENSUBDIVlib -losdGPU -losdCPU -o myapp install_name_tool -add_rpath OPENSUBDIVlib myapp On 64-bit OS-X add -m64 after each g . Windows cl nologo MT TP DWIN32 IOPENSUBDIVinclude -c myapp.cpp link nologo outmyapp.exe LIBPATHOPENSUBDIVlib libosdGPU.lib libosdCPU.lib myapp.obj Note HBR uses the offsetof macro on a templated struct, which appears to spurriously set off a warning in both gcc and Clang. It is recommended to turn the warning off with the -Wno-invalid-offsetof flag. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"Building with CMake Building with CMake 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Overview Step 1 Dependencies Required Optional Step 2 Configuring CMake Useful Build Options Environment Variables Automated Script Using Intels C Studio XE Using Clang Step 3 Building Build Targets Compiling Linking an OpenSubdiv Application Overview Assuming that you have cloned the source repository and selected an appropriate release branch, the following instructions will walk you through the CMake and configuration and build process. CMake is a cross-platform, open-source build system. CMake controls the compilation process using platform independent configuration files in order to generate Makefiles and workspaces that are native to the platform of choice. The process involves the following steps Locate build the requisite dependencies Configure run CMake to generate Makefiles MSVC solution XCode project Run the build from make MSVC XCode Step 1 Dependencies CMake will adapt the build based on which dependencies have been successfully discovered and will disable certain features and code examples accordingly. Please refer to the documentation of each of the dependency packages for specific build and installation instructions. Required CMake version 3.12 Optional Ptex support features for ptex textures and the ptexViewer example Zlib required for Ptex under Windows CUDA TBB OpenCL DX11 SDK GLFW required for standalone examples and some regression tests Docutils required for reST-based documentation Python Pygments required for Docutils reST styling Doxygen Step 2 Configuring CMake One way to configure CMake is to use the CMake GUI . In many cases CMake can fall back on default standard paths in order to find the packages that OpenSubdiv depends on. For non-standard installations however, a complete set of override variables is available. The following sub-section lists some of these variables. For more specific details, please consult the source of the custom CMake modules in the OpenSubdivcmake folder. Useful Build Options The following configuration arguments can be passed to the CMake command line. - DCMAKE_BUILD_TYPE Debug Release - DCMAKE_INSTALL_PREFIX base path to install OpenSubdiv default Current directory - DCMAKE_LIBDIR_BASE library directory basename default lib - DCUDA_TOOLKIT_ROOT_DIR path to CUDA - DPTEX_LOCATION path to Ptex - DGLFW_LOCATION path to GLFW - DTBB_LOCATION path to Intel s TBB - DICC_LOCATION path to Intel s C Studio XE - DNO_LIB 1 disable the opensubdiv libs build caveat emptor - DNO_EXAMPLES 1 disable examples build - DNO_TUTORIALS 1 disable tutorials build - DNO_REGRESSION 1 disable regression tests build - DNO_PTEX 1 disable PTex support - DNO_DOC 1 disable documentation build - DNO_OMP 1 disable OpenMP - DNO_TBB 1 disable TBB - DNO_CUDA 1 disable CUDA - DNO_OPENCL 1 disable OpenCL - DNO_OPENGL 1 disable OpenGL - DNO_CLEW 1 disable CLEW wrapper library Environment Variables The paths to Ptex, GLFW, other dependencies can also be specified through the following environment variables PTEX_LOCATION , GLFW_LOCATION Automated Script The GUI solution will probably become a burden for active developpers who tend to re-run the configuration step fairly often. A scripted solution can save a lot of time. Here is a typical workflow git clone https github.comPixarAnimationStudiosOpenSubdiv.git folder cd folder mkdir build cd build source .. .. cmake_setup Where cmake_setup is a configuration script. Here is an example CMake configuration script for a full typical windows-based build that can be run in GitShell bintcsh Replace the .. with a full path to the root of the OpenSubdiv source tree if necessary cProgram Files x86CMake 2.8bincmake.exe - G Visual Studio 15 2017 Win64 - D GLFW_LOCATIONstringcProgram Filesglfw-2.7.7.bin.WIN64 - D OPENCL_INCLUDE_DIRSstringcProgramDataNVIDIA CorporationNVIDIA GPU Computing SDK 4.2OpenCLcommoninc - D _OPENCL_CPP_INCLUDE_DIRSstringcProgramDataNVIDIA CorporationNVIDIA GPU Computing SDK 4.2OpenCLcommoninc - D OPENCL_LIBRARIESstringcProgramDataNVIDIA CorporationNVIDIA GPU Computing SDK 4.2OpenCLcommonlibx64OpenCL.lib - D PTEX_LOCATIONstringcUsersopensubdivdemosrcptexx64 .. copy Ptex dependencies Windows only mkdir - p bin Debug , Release cp - f c Users opensubdiv demo src zlib -1.2.7 contrib vstudio vc10 x64 ZlibDllRelease zlibwapi . dll bin Debug cp - f c Users opensubdiv demo src zlib -1.2.7 contrib vstudio vc10 x64 ZlibDllRelease zlibwapi . dll bin Release cp - f c Users opensubdiv demo src ptex x64 lib Ptex . dll bin Debug cp - f c Users opensubdiv demo src ptex x64 lib Ptex . dll bin Release Important Notice that the following scripts start by recursively removing the ..build and ..inst directories. Make sure you modify them to suit your build workflow. Here is a similar script for Nix-based platforms echo Removing build cd .. rm - rf build inst mkdir build cd build echo Running cmake cmake - DPTEX_LOCATION home opensubdiv dev opensource ptex install - DGLFW_LOCATION home opensubdiv dev opensource glfw build - DDOXYGEN_EXECUTABLE home opensubdiv dev opensource doxygen inst bin doxygen - DCMAKE_INSTALL_PREFIX .. inst - DCMAKE_BUILD_TYPE Debug .. Here is a similar script for macOS echo Removing build cd .. rm - rf build inst mkdir build cd build echo Running cmake cmake - DOPENGL_INCLUDE_DIR Applications Xcode . app Contents Developer Platforms MacOSX . platform Developer SDKs MacOSX10 .9 . sdk System Library Frameworks OpenGL . framework Headers - DGLFW_LOCATION Users opensubdiv dev opensource glfw inst - DNO_OMP 1 - DNO_REGRESSION 0 - DCMAKE_INSTALL_PREFIX .. inst - DCMAKE_BUILD_TYPE Debug .. Using Intels C Studio XE OpenSubdiv can be also be built with Intels C compiler icc. The default compiler can be overriden in CMake with the following configuration options - DCMAKE_CXX_COMPILER path to icc executable - DCMAKE_C_COMPILER path to icc executable The installation location of the C Studio XE can be overriden with - DICC_LOCATION path to Intel s C Studio XE Using Clang CMake can also be overriden to use the clang compilers by configuring the following options - DCMAKE_CXX_COMPILER clang - DCMAKE_C_COMPILER clang Step 3 Building CMake provides a cross-platform command-line build cmake -- build . -- target install -- config Release Alternatively, you can use native toolkits to launch the build. The steps differ for each OS Windows launch VC with the solution generated by CMake in your build directory. macOS launch Xcode with the xcodeproj generated by CMake in your build directory Nix run make in your build directory - use the clean target to remove previous build results - use VERBOSE1 for verbose build output Build Targets Makefile-based builds allow the use of named target. Here are some of the more useful target names osd_staticdynamic_CPUGPU The core components of the OpenSubdiv libraries example_name Builds specific code examples by name glViewer, ptexViewer... doc Builds ReST and doxygen documentation doc_html Builds ReST documentation doc_doxy Builds Doxygen documentation Compiling Linking an OpenSubdiv Application Here are example commands for building an OpenSubdiv application on several architectures Linux g -IOPENSUBDIVinclude -c myapp.cpp g myapp.o -LOPENSUBDIVlib -losdGPU -losdCPU -o myapp macOS g -IOPENSUBDIVinclude -c myapp.cpp g myapp.o -LOPENSUBDIVlib -losdGPU -losdCPU -o myapp install_name_tool -add_rpath OPENSUBDIVlib myapp On 64-bit OS-X add -m64 after each g . Windows cl nologo MT TP DWIN32 IOPENSUBDIVinclude -c myapp.cpp link nologo outmyapp.exe LIBPATHOPENSUBDIVlib libosdGPU.lib libosdCPU.lib myapp.obj Note HBR uses the offsetof macro on a templated struct, which appears to spurriously set off a warning in both gcc and Clang. It is recommended to turn the warning off with the -Wno-invalid-offsetof flag. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"cmake_build.html\"\n }, {\n \"title\": \"Code Examples\",\n- \"text\": \"Code Examples Code Examples 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Standalone Viewers OpenSubdiv builds a number of standalone viewers that demonstrate various aspects of the software. OpenGL examples glViewer glFVarViewer glEvalLimit glStencilViewer glPtexViewer glPaintTest glShareTopology DirectX examples dxViewer dxPtexViewer Metal examples mtlViewer mtlPtexViewer Common Command Line Options While the following command line options are common, not all examples support all -- in some cases, certain options are not applicable. -f launches the application in full-screen mode if supported -yup set initial view with Y-axis up Z-up is the default -u apply uniform subdivision to all meshes -a apply adaptive subdivision to create patches for the limit surface -l depth level of uniform or adaptive refinement to apply -c count number of repetitions of the animation loop when supported default of 0 is infinite -anim interpret a given set of Obj files as an animated sequence rather than a set of distinct meshes -catmark apply the Catmark scheme to all meshes in given Obj files default -loop apply the Loop scheme to all meshes in given Obj files -bilinear apply the Bilinear scheme to all meshes in given Obj files objfiles a set of one or more meshes in Obj format requiring a .obj extension that may be distinct meshes or animated versions of a single mesh Common Keyboard Controls Left mouse button drag orbit camera Middle mouse button drag pan camera Right mouse button dolly camera n, p nextprev model 1, 2, 3, ..., 9, 0 specify adaptive isolation or uniform refinement level , - increase decrease tessellation Tab toggle full-screen Esc turn on off the HUD w switch display mode q quit Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"Code Examples Code Examples 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Standalone Viewers OpenSubdiv builds a number of standalone viewers that demonstrate various aspects of the software. OpenGL examples glViewer glFVarViewer glEvalLimit glStencilViewer glPtexViewer glPaintTest glShareTopology DirectX examples dxViewer dxPtexViewer Metal examples mtlViewer mtlPtexViewer Common Command Line Options While the following command line options are common, not all examples support all -- in some cases, certain options are not applicable. -f launches the application in full-screen mode if supported -yup set initial view with Y-axis up Z-up is the default -u apply uniform subdivision to all meshes -a apply adaptive subdivision to create patches for the limit surface -l depth level of uniform or adaptive refinement to apply -c count number of repetitions of the animation loop when supported default of 0 is infinite -anim interpret a given set of Obj files as an animated sequence rather than a set of distinct meshes -catmark apply the Catmark scheme to all meshes in given Obj files default -loop apply the Loop scheme to all meshes in given Obj files -bilinear apply the Bilinear scheme to all meshes in given Obj files objfiles a set of one or more meshes in Obj format requiring a .obj extension that may be distinct meshes or animated versions of a single mesh Common Keyboard Controls Left mouse button drag orbit camera Middle mouse button drag pan camera Right mouse button dolly camera n, p nextprev model 1, 2, 3, ..., 9, 0 specify adaptive isolation or uniform refinement level , - increase decrease tessellation Tab toggle full-screen Esc turn on off the HUD w switch display mode q quit Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"code_examples.html\"\n }, {\n+ \"title\": \"BFR Overview\",\n+ \"text\": \"BFR Overview BFR Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Base Face Representation Bfr Evaluation BfrSurfaceFactory BfrSurface Parameterization BfrParameterization Discontinuous Parameterizations Tessellation BfrTessellation Tessellation Rates Differences from Hardware Tessellation More on BfrSurfaceFactory BfrSurfaceFactoryCache Defining a Thread-Safe SurfaceFactory Internal versus External SurfaceFactoryCache Customizing a BfrSurfaceFactory The BfrSurfaceFactoryMeshAdapter Interface Customizing the Subclass Interface Base Face Representation Bfr Bfr is an alternate API layer that treats a subdivision mesh provided by a client as a piecewise parameteric surface primitive . The name Bfr derives from the fact that the concepts and classes of this interface all relate to the base faces of a mesh. Concepts such as parameterization , evaluation and tessellation all refer to and are embodied by classes that deal with a specific face of the original unrefined mesh. The Bfr interfaces allow the limit surface for a single face to be identified and evaluated independently of all other faces without any global pre-processing. While concepts and utilities from the Far interface are used internally, the details of their usage is hidden. There is no need to coordinate adaptive refinement with tables of patches, stencils, Ptex indices, patch maps, etc. The resulting evaluation interface is much simpler, more flexible and more scalable than those assembled with the Far classes -- providing a preferable alternative for many CPU-based use cases. The main classes in Bfr include SurfaceFactory A light-weight interface to a mesh that constructs pieces of limit surface for specified faces of a mesh in the form of Surfaces. Surface A class encapsulating the limit surface of a face with methods for complete parametric evaluation. Parameterization A simple class defining the available parameterizations of faces and for identifying that of a particular face. Tessellation A simple class providing information about a specified tessellation pattern for a given Parameterization. Bfr is well suited to cases where evaluation of the mesh may be sparse, dynamically determined or iterative Newton, gradient descent, etc. It is not intended to replace the cases for which Far has been designed i.e. repeated evaluation of a fixed set of points but is intended to complement them. While simplicity, flexibility and reasonable performance were the main goals of Bfr , its current implementation often outperforms the table-based solutions of Far for many common use cases -- both in terms of execution time and memory use. An area that Bfr does not address, and where Far remains more suited, is capturing a specific representation of the limit surface for external use. Bfr intentionally keeps internal implementation details private to allow future improvements or extensions. Those representation details may be publicly exposed in future releases, but until then, use of Far is required for such purposes. Evaluation Since subdivision surfaces are piecewise parametric surfaces, the main operation of interest is evaluation. Bfr deals with the limit surface of a mesh as a whole by associating pieces of surface with each face of the mesh. These pieces of surface are referred to in the context of Bfr simply as surfaces and represented by BfrSurface. Each face of the mesh has an implicit local 2D parameterization and individual coordinates of that parameterization are used to evaluate its corresponding Surface. In general, 3- and 4-sided faces use the same parameterizations for quad and triangular patches used elsewhere in OpenSubdiv Parameterizations are defined for other faces more details to follow, so Surfaces for all faces can be evaluated given any 2D parametric coordinate of its face. Given an instance of a mesh, usage first requires the creation of a BfrSurfaceFactory corresponding to that mesh -- from which Surfaces can then be created for evaluation. Construction of the SurfaceFactory involves no pre-processing and Surfaces can be created and discarded as needed. The processes of constructing and evaluating Surfaces are described in more detail below. BfrSurfaceFactory Construction of BfrSurfaces requires an instance of BfrSurfaceFactory. An instance of SurfaceFactory is a light-weight interface to an instance of a mesh that requires little to no construction cost or memory. The SurfaceFactory does no work until a Surface is requested for a particular face -- at which point the factory inspects the mesh topology around that face to assemble the Surface. SurfaceFactory is actually a base class that is inherited to provide a consistent construction interface for Surfaces. Subclasses are derived to support a particular class of connected mesh -- to implement the topology inspection around each face required to construct the Surface. Use of these subclasses is very simple given the public interface of SurfaceFactory, but defining such a subclass is not. That more complex use case of SurfaceFactory will be described in detail later with other more advanced topics. In many cases, it is not necessary to explicitly define a subclass of SurfaceFactory, as the tutorials for Bfr illustrate. If already using OpenSubdiv for other reasons, a FarTopologyRefiner will have been constructed to represent the initial base mesh before refinement. Bfr provides a subclass of SurfaceFactory using FarTopologyRefiner as the base mesh ignoring any levels of refinement for immediate use in such cases. For those cases when no connected mesh representation is available at all i.e. only raw, unconnected mesh data exists construction of a FarTopologyRefiner provides a reasonably efficient connected mesh representation see the Far tutorials for construction details, whose provided subclass for SurfaceFactory is then readily available. Given the different interpolation types for mesh data i.e. vertex, varying and face-varying, the common interface for SurfaceFactory provides methods to construct Surfaces explicitly for all data types. So for positions, the methods for vertex data must be used to obtain the desired Surface, while for texture coordinates the methods for face-varying are usually required, e.g. Surface CreateVertexSurface Index faceIndex const Surface CreateVaryingSurface Index faceIndex const Surface CreateFaceVaryingSurface Index faceIndex const The Surfaces created by these construction methods may all be distinct as the underlying representations of the Surfaces and the indices of the data that define them will often differ. For example, the position data may require a bicubic patch while the face-varying texture data may be linear or a different type of bicubic patch given the different interpolation rules for face-varying and the possibility of seams. While the internal representations of the Surfaces constructed for different data interpolation types may differ, since they are all constructed as Surfaces, the functionality used to evaluate them is identical. BfrSurface The Surface class encapsulates the piece of limit surface associated with a particular face of the mesh. The term surface is used rather than patch to emphasize that the Surface may itself be a piecewise parametric surface composed of more than one patch potentially even a complex set of patches. Surface is also a class template selected by floating point precision, and so typically declared as BfrSurfacefloat. Just as a simpler type name is likely to be declared when used, the simple name Surface will be used to refer to it here. And where code fragments may be warranted, float will be substituted for the template parameter for clarity. Once created, there are two steps required to evaluate a Surface preparation of associated data points from the mesh the actual calls to evaluation methods using these data points The latter is straight-forward, but the former warrants a little more explanation. The shape of a Surface for a base face is influenced by the set of data points associated with both the vertices of the face and a subset of those in its immediate neighborhood. These control points are identified when the Surface is initialized and are publicly available for inspection if desired. The control points are sufficient to define the Surface if the face and its neighborhood are regular, but any irregularity an extra-ordinary vertex, crease, etc. usually requires additional, intermediate points to be computed from those control points in order to evaluate the Surface efficiently. Having previously avoided use of the term patch in favor of surface, the term patch points is now used to refer to these intermediate points. Patch points always include the control points as a subset and may be followed by points needed for any additional patches required to represent a more complex Surface. While the patch points are assembled in a local array for direct use by the Surface, the control points can either be gathered and accessed locally or indexed from buffers associated with the mesh for other purposes e.g. computing a bounding box of the Surface Once the patch points for a Surface are prepared, they can be passed to the main evaluation methods with the desired parametric coordinates. As previously noted, since the Surface class is a template for floating point precision, evaluation is supported in single or double precision by constructing a Surface for the desired precision. Evaluation methods are overloaded to obtain simply position or including all first or second derivatives. So preparation and evaluation can be achieved with the following Preparing patch points void PreparePatchPoints float const meshPoints , PointDescriptor meshPointDescriptor , float patchPoints , PointDescriptor patchPointDescriptor const Evaluating position and 1st derivatives void Evaluate float const uv 2 , float const patchPoints , PointDescriptor patchPointDescriptor , float P , float dPdu , float dPdv const The PointDescriptor class here is a simple struct defining the size and stride of the associated array of points. Any use of mesh points, control points or patch points generally requires an accompanying descriptor. Depending on the complexity of the limit surface, this preparation of patch points can be costly -- especially if only evaluating the Surface once or twice. In such cases, it is worth considering evaluating limit stencils, i.e. sets of coefficients that combine the original control vertices of the mesh without requiring the computation of intermediate values. The cost of evaluating stencils is considerably higher than direct evaluation, but that added overhead is often offset by avoiding the use of patch points. Surfaces should be considered a class for transient use as retaining them for longer term usage can reduce their benefits. The relatively high cost of initialization of irregular Surfaces can be a deterrent and often motivates their retention despite increased memory costs. Retaining all Surfaces of a mesh for random sampling is a situation that should be undertaken with caution and will be discussed in more detail later with other advanced topics. Parameterization Each face of a mesh has an implicit local 2D parameterization whose 2D coordinates are used to evaluate the Surface for that face. Bfr adopts the parameterizations defined elsewhere in OpenSubdiv for quadrilateral and triangular patches, for use quadrilateral and triangular faces But the parameterization of a face is also dependent on the subdivision scheme applied to it. Subdivision schemes that divide faces into quads are ultimately represented by quadrilateral patches. So a face that is a quad can be parameterized as a single quad, but other non-quad faces are parameterized as a set of quad sub-faces, i.e. faces resulting from subdivision A triangle subdivided with a quad-based scheme e.g. Catmull-Clark will therefore not have the parameterization of the triangular patch indicated previously, but another defined by its quad sub-faces illustrated above to be described in more detail below. Subdivision schemes that divide faces into triangles are currently restricted to triangles only, so all faces are parameterized as single triangles. If Loop subdivision is extended to non-triangles in future, a parameterization involving triangular sub-faces will be necessary. Note that triangles are often parameterized elsewhere in terms of barycentric coordinates u,v,w where w 1 - u - v . As is the case elsewhere in OpenSubdiv, Bfr considers parametric coordinates as 2D u,v pairs for all purposes. All faces have an implicit 2D local parameterization and all interfaces requiring parametric coordinates consider only the u,v pair. If interaction with some other tool set requiring barycentric coordinates for triangles is necessary, it is left to users to compute the implicit w accordingly. BfrParameterization BfrParameterization is a simple class that fully defines the parameterization for a particular face. An instance of Parameterization is fully defined on construction given the size of a face and the subdivision scheme applied to it where the face size is its number of verticesedges. Since any parameterization of N -sided faces requires N in some form, the face size is stored as a member and made publicly available. Each Surface has the Parameterization of its face assigned internally as part of its construction, and that is used internally by the Surface in many of its methods. The need to deal directly with the explicit details of the Parameterization class is not generally necessary. Often it is sufficient to retrieve the Parameterization from a Surface for use in some other context e.g. passed to BfrTessellation. The enumerated type ParameterizationType currently defines three kinds of parameterizations -- one of which is assigned to each instance on construction according to the properties of the face QUAD Applied to quadrilateral faces with a quad-based subdivision scheme e.g. Catmark or Bilinear. TRI Applied to triangular faces with a triangle-based subdivision scheme e.g. Loop. QUAD_SUBFACES Applied to non-quad faces with a quad-based subdivision scheme -- dividing the face into quadrilateral sub-faces. Parameterizations that involve subdivision into sub-faces, e.g. QUAD_SUBFACES, may warrant some care as they are not continuous. Depending on how they are defined, the sub-faces may be disjoint e.g. Bfr or overlap in parametric space e.g. Ptex. To help these situations, methods to detect the presence of sub-faces and deal with their local parameterizations are made available. Discontinuous Parameterizations When a face does not have a regular parameterization, the division of the parameterization into sub-faces can create complications -- as noted and addressed elsewhere in OpenSubdiv. BfrParameterization defines a quadrangulated sub-face parameterization differently from the Far and Osd interfaces. For an N -sided face, Far uses a parameterization adopted by Ptex. In this case, all quad sub-faces are parameterized over the unit square and require an additional index of the sub-face to identify them. So Ptex coordinates require three values the index and u,v of the sub-face. To embed sub-face coordinates in a single u,v pair, Bfr tiles the sub-faces in disjoint regions in parameter space. This tiling is similar to the Udim convention for textures, where a UDim on the order of sqrtN is used to preserve accuracy for increasing N Note also that the edges of each sub-face are of parametric length 0.5, which results in a total parametric length of 1.0 for all base edges. This differs again from Ptex, which parameterizes sub-faces with edge lengths of 1.0, and so can lead to inconsistencies in parametric scale typically with derivatives across edges of the mesh if not careful. As previously mentioned, care may be necessary when dealing with the discontinuities that exist in parameterizations with sub-faces. This is particularly true if evaluating data at sampled locations of the face and needing to evaluate at other locations interpolated from these. Interpolation between parametric locations, e.g. A, B and C, should be avoided when discontinuous. In many cases, dealing directly with coordinates of the sub-faces is unavoidable, e.g. interpolating Ptex coordinates for sampling of textures assigned explicitly to the sub-faces. Methods are provided to convert from Bfr s tiled parameterization to and from other representations that use a local parameterization for each sub-face. Tessellation Once a Surface can be evaluated it can be tessellated. Given a 2D parameterization, a tessellation consists of two parts a set of parametric coordinates sampling the Parameterization a set of faces connecting these coordinates that covers the entire Parameterization Once evaluated, the resulting set of sample points and the faces connecting them effectively define a mesh for that parameterization. For the sake of brevity both here and in the programming interface, the parametric coordinates or sample points are referred to simply as coords or Coords -- avoiding the term points, which is already a heavily overloaded term. Similarly the faces connecting the coords are referred to as facets or Facets -- avoiding the term face to avoid confusion with the base face of the mesh being tessellated. Bfr provides a simple class to support a variety of tessellation patterns for the different Parameterization types and methods for retrieving its associated coords and facets. In many cases the patterns they define are similar to those of GPU hardware tessellation -- which may be more familiar to many -- but they do differ in several ways, as noted below. BfrTessellation In Bfr a Tessellation is a simple class defined by a Parameterization and a given set of tessellation rates and a few additional options. These two elements define a specific tessellation pattern for all faces sharing that Parameterization. An instance of Tessellation can then be inspected to identify all or subsets of its coords or facets. The process of tessellation in other contexts usually generates triangular facets, but that is not the case with Bfr . While producing triangular facets is the default, options are available to have Tessellation include quads in patterns for parameterizations associated with quad-based subdivision schemes. For simple uniform patterns, these produce patterns that are similar in topology to those resulting from subdivision Tessellation of 4- and 5-sided faces of a quad-based scheme using quadrilateral facets left and triangular right The name Tessellation was chosen rather than Tessellator as it is a passive class that simply holds information define its pattern. It doesnt do much other than providing information about the pattern when requested. A few general properties about the pattern are determined and retained on construction, after which an instance is immutable. So it does not maintain any additional state between queries. In order to provide flexibility when dealing with tessellations of adjacent faces, the coords arising from an instance of Tessellation are ordered and are retrievable in ways to help identify points along edges that may be shared between two or more faces. The coords of a Tessellation are generated in concentric rings, beginning with the outer ring and starting with the first vertex Ordering of coords around boundary for quad and tri parameterizations. Methods of the Tessellation class allow the coords associated with specific vertices or edges to be identified, as well as providing the coords for the entire ring around the boundary separately from those of the interior if desired. While the ordering of coords in the interior is not defined and so not to be relied upon, the ordering of the boundary coords is specifically fixed to support the correlation of potentially shared coords between faces. The Tessellation class is completely independent of the Surface class. Tessellation simply takes a Parameterization and tessellation rates and provides the coords and facets that define its pattern. So Tessellation can be used in any other evaluation context where the Parameterizations are appropriate. Tessellation Rates For a particular Parameterization, the various tessellation patterns are determined by one or more tessellation rates. The simplest set of patterns uses a single tessellation rate and is said to be uniform, i.e. all edges and the interior of the face are split to a similar degree Uniform tessellation of a quadrilateral and triangle with rates of 5 and 8. More complex non-uniform patterns allow the edges of the face to be split independently from the interior of the face. Given rates for each edge, a suitable uniform rate for the interior can be either inferred or specified explicitly. These are typically referred to as the outer rates and the inner rate. The single rate specified for a simple uniform tessellation is essentially the specification of a single inner rate while the outer rates for all edges are inferred as the same. Non-uniform tessellation of a quadrilateral, triangle and 5-sided face with various outer and inner rates. In the case of Parameterizations for quads, it is common elsewhere to associate two inner rates with the opposing edges. So two separate inner rates are available for quad parameterizations -- to be specified or otherwise inferred Quad tessellations with differing inner rates with matching left and varying outer rates right. Differences from Hardware Tessellation Since the specifications for hardware tessellation often leave some details of the patterns as implementation dependent, no two hardware implementations are necessarily the same. Typically there may be subtle differences in the non-uniform tessellation patterns along boundaries, and that is to be executed here. Bfr does provide some obvious additional functionality not present in hardware tessellation and vice versa, e.g. Bfr provides the following not supported by hardware tessellation patterns for parameterizations other than quads and tris e.g. N-sided preservation of quad facets of quad-based parameterizations while hardware tessellation provides the following not supported by Bfr patterns for so-called fractional tessellation non-integer rates The lack of fractional tessellation in Bfr is something that may be addressed in a future release. Where the functionality of Bfr and hardware tessellation overlap, a few other differences are worth noting indexing of edges and their associated outer tessellation rates uniform tessellation patterns for triangles differ significantly For the indexing of edges and rates, when specifying an outer rate associated with an edge, the array index for rate i is expected to correspond to edge i . Bfr follows the convention established elsewhere in OpenSubdiv of labelingindexing edges 0, 1, etc. between vertex pairs 0,1, 1,2, etc. So outer rate 0 corresponds to the edge between vertices 0,1. In contrast, hardware tessellation associates the rate for the edge between vertices 0,1 as outer rate 1 -- its outer rate 0 is between vertices N-1,0. So an offset of 1 is warranted when comparing the two. Outer edge tessellation rates of 1,3,5,7 applied to a quad with Bfr left and GPU tessellation right. For the uniform tessellation of triangles, its well known that the needs of hardware implementation led designers to factor the patterns for triangles to make use of the same hardware necessary for quads. As a result, many edges are introduced into a simple tessellation of a triangle that are not parallel to one of its three edges. Bfr uses patterns more consistent with those resulting from the subdivision of triangles. Only edges parallel to the edges of the triangle are introduced, which creates more uniform facets both edge lengths and area and reduces their number by one third. This can reduce artifacts that sometimes arise with use of the hardware patterns at lower tessellation rates Uniform tessellation of a triangle with Bfr left and GPU tessellation right. These triangular patterns have been referred to as integer spacing for triangular patches in early work on hardware tessellation. But use of these patterns was generally discarded in favor of techniques that split the triangle into three quads -- allowing the hardware solution for quad tessellation to be reused. More on BfrSurfaceFactory The primary function of BfrSurfaceFactory is to identify and construct a representation of the limit surface for a given face of a mesh. It achieves this by inspecting the topology around the given face and constructing a suitable representation encapsulated in a Surface. The regions around a face can be divided into two categories based on their topology those that are regular and those that are not, i.e. those that are irregular. Recalling the illustration from Irregular versus Irregular Features Patches of regular Surfaces Potential patches of irregular Surfaces The representation of the limit surface for regular regions is trivial -- it is a single parametric patch whose basis is determined by the subdivision scheme e.g. uniform bicubic B-spline for Catmull-Clark. In contrast, the representation of the limit surface for an irregular region cannot be accurately represented so simply. It can be far more complex depending on the features present extra-ordinary vertices, creasing of edges, etc.. It may be as simple as a different kind of parametric patch whose points are derived from those of the mesh, or it may often be a set of patches in a hierarchy resulting from local subdivision. Bfr intentionally hides the details of these representations to allow future improvement. The cost of determining and assembling the representations of irregular Surfaces is therefore often significant. Some of the performance benefits of the SurfaceFactory are achieved by having it cache the complexities of the irregular surfaces that it encounters. In many common use cases, awareness and management of this caching is not necessary as illustrated by the tutorials. But the thread-safe construction of Surfaces is one area where some awareness is required. Other use cases that share the cache between meshes are also worth exploring as they can further reduce potentially significant costs. BfrSurfaceFactoryCache The SurfaceFactoryCache is the class used by SurfaceFactory to cache the topological information that it can reuse for other similarly irregular faces of the mesh. Though it is a publicly accessible class, the SurfaceFactoryCache has little to no public interface other than construction made available to support more advanced cases covered later and in most cases it can be completely ignored. Typically an instance of SurfaceFactory has an internal SurfaceFactoryCache member which is used by that factory for its lifetime. Since that cache member is mutable -- potentially updated when an irregular Surface is created -- it does need to be thread-safe if the SurfaceFactory is to be used in a threaded context. To accommodate this need, SurfaceFactoryCache is defined as a base class with an accompanying class template to allow the trivial declaration of thread-safe subclasses template typename MUTEX_TYPE , typename READ_LOCK_GUARD_TYPE , typename WRITE_LOCK_GUARD_TYPE class SurfaceFactoryCacheThreaded public SurfaceFactoryCache ... For example, a local type for a thread-safe cache using stdshared_mutex from C17 could be simply declared as follows include shared_mutex typedef Bfr SurfaceFactoryCacheThreaded std shared_mutex , std shared_lock std shared_mutex , std unique_lock std shared_mutex ThreadSafeCache Such thread-safe cache types are essential when distributing the work of a single SurfaceFactory across multiple threads. They can be encapsulated in the definitions of subclasses of SurfaceFactory or used to define external cache instances for use with any subclass of SurfaceFactory. Defining a Thread-Safe SurfaceFactory The thread-safety of a SurfaceFactory is purely dependent on the thread-safety of the SurfaceFactoryCache that it uses. With caching disabled, any SurfaceFactory is thread-safe but will be far less efficient in dealing with irregular Surfaces. When a subclass of SurfaceFactory is defined discussed in more detail later, one of its responsibilities is to identify and manage an instance of SurfaceFactoryCache for its internal use. Defining such a subclass is a simple matter of declaring a thread-safe SurfaceFactoryCache type as noted above along with a local member of that type to be used by each instance. Given the widespread use of the FarTopologyRefiner in OpenSubdiv, and the lack of a connected mesh representation in many contexts, a subclass of SurfaceFactory is made available to use a TopologyRefiner as a mesh, i.e. the BfrRefinerSurfaceFactory subclass. Since many OpenSubdiv users may make use of the RefinerSurfaceFactory subclass, and they may have different preferences of threading model, the RefinerSurfaceFactory subclass is similarly defined as a class template to enable threading flexibility. In this case, the template is parameterized by the desired type of SurfaceFactoryCache, which embodies the threading specifications as noted above, i.e. template class CACHE_TYPE SurfaceFactoryCache class RefinerSurfaceFactory public ... ... The default template is the base SurfaceFactoryCache which is not thread-safe, but a simple declaration of a thread-safe cache type is sufficient to declare a similarly thread-safe RefinerSurfaceFactory type include opensubdivbfrsurfaceFactoryCache.h Declare thread-safe cache type see stdshared_mutex example above typedef Bfr SurfaceFactoryCacheThreaded ... ThreadSafeCache Declare thread-safe factory type typedef Bfr RefinerSurfaceFactory ThreadSafeCache ThreadSafeFactory The resulting factory type safely allows the construction of Surfaces and their subsequent evaluation and tessellation to be distributed over multiple threads. Internal versus External SurfaceFactoryCache Typical usage of the SurfaceFactoryCache by the SurfaceFactory is to have the factory create an internal cache member to be used for the lifetime of the factory associated with a mesh. But the data stored in the cache is not in any way dependent on the factory or mesh used to create it. So a cache can potentially be shared by multiple factories. While such sharing is possible -- and the Bfr interfaces intentionally permit it -- any exploration should proceed with caution. Greater public knowledge and control of the cache is ultimately necessary to manage its potentially unbounded memory increase, and support in the public interface is currently limited. A cache stored as a member varialbe and managed exclusively by the factory is said to be internal while one managed exclusively by its client is said to be external. In both cases, the factory deals with retrieving data from or adding data to the cache -- only management of the caches ownership differs, and that ownership is never transferred. A subset of the methods of SurfaceFactoryOptions provide the means of specifying the use of an internal or external cache, or no caching at all Assign an external cache to override the internal Options SetExternalCache SurfaceFactoryCache cache Enable or disable caching default is true Options EnableCaching bool on As noted here, specifying an external cache will override use of a factorys internal cache. Disabling caching takes precedence over both, but is generally not practical and exists mainly to aide debugging. The common use of the internal cache is to create a SurfaceFactory and distribute processing of the Surfaces of its faces over multiple threads, or to construct Surfaces for the mesh for any other purpose while the mesh remains in scope. There is no need to deal explicitly with the SurfaceFactoryCache in these cases. Use cases for an external cache are more varied and explicit, including creating a single external cache to process a sequence of meshes on a single thread cache thread-safety not required creating a separate external cache on each thread to process a set of meshes distributed over multiple threads cache thread-safety not required creating a single external cache for multiple meshes distributed over multiple threads cache thread-safety required, and beware of unbounded memory growth here Future extensions to the public interface of SurfaceFactoryCache may be made to support common use cases as their common needs are made clearer. Customizing a BfrSurfaceFactory One of the goals of Bfr is to provide a lightweight interface for the evaluation of Surfaces from any connected mesh representation. In order to do so, the factory needs to gather topological information from that mesh representation. That information is provide to the factory through inheritance a subclass of SurfaceFactory is defined that fulfills all requirements of the factory. It must be made clear that a subclass can only be created from a connected mesh representation, i.e. a representation that includes connectivity or adjacency relationships between its components vertices, faces and edges. Classes for simple containers of mesh topology used for external formats e.g. USD, Alembic, etc. are generally not connected . Many applications construct a connected mesh representation for internal use when loading such mesh data -- using a variety of techniques including half-edges, winged-edges or table-based relationships. There are many choices here that offer a variety of trade-offs depending on usage e.g. fixed vs dynamic topology and so no best solution. Once constructed and available within an application, Bfr strives to take advantage of that representation. As a minimum requirement for supporting a subclass of SurfaceFactory, a connected mesh representation must be able to efficiently identify the incident faces of any given vertex. As noted earlier, when no such representation is available, users can construct a FarTopologyRefiner for their connected mesh and use BfrRefinerSurfaceFactory. There are three requirements of a subclass of SurfaceFactory fulfill the interface required to adapt the connected mesh to the factory provide an internal cache for the factory of the preferred type extend the existing SurfaceFactory interface for the connected mesh type The first of these is the most significant and is the focus here. The second was mentioned previously with the SurfaceFactoryCache and is trivial. The last should also be trivial and is generally optional at minimum the subclass will need a constructor to create an instance of the factory from a given mesh, but anything more is not strictly essential. It is important to note that anyone attempting to write such a subclass must have an intimate understanding of the topological capabilities and limitations of the mesh representation involved. The SurfaceFactory is topologically robust in that it will support meshes with a wide range of degenerate or non-manifold features, but in order to process topology efficiently, a subclass needs to indicate when and where those degeneracies may occur. A simplified implementation of the BfrRefinerSurfaceFactory is provided in the tutorials for illustration purposes. The BfrSurfaceFactoryMeshAdapter Interface The SurfaceFactoryMeshAdapter class defines the interface used to satisfy the topological requirements of the SurfaceFactory. An implementation for a particular mesh class provides the base factory with everything needed to identify the limit surface of a given face from its surrounding topology. The SurfaceFactory actually inherits the SurfaceFactoryMeshAdapter interface but does not implement it -- deferring that to its subclasses -- since separate subclasses of SurfaceFactoryMeshAdapter serve no other purpose. The limit surface for a face is fully defined by the complete set of incident vertices, faces and edges surrounding the face. But it is difficult to accurately and efficiently assemble and represent all of that required information in a single class or query for all possible cases. So the mesh adapter interface provides a suite of methods to allow the factory to gather only what it needs for the Surface required -- which may differ considerably according to whether the Surface is for vertex or face-varying data, linear or non-linear, etc. The virtual methods required can be organized into small groups devoted to particular aspects of construction. A description of the methods and purposes for each group follows, with more details and exact signatures available in the accompanying Doxygen for the SurfaceFactoryMeshAdapter class. Basic Properties of a Face A small set of simple methods indicate whether the SurfaceFactory needs to create a Surface for a face, and if so, how virtual bool isFaceHole Index faceIndex const 0 virtual int getFaceSize Index faceIndex const 0 These are trivial and self-explanatory. Identifying Indices for an Entire Face If the Surface requested turns out to be linearly interpolated e.g. for varying or linear face-varying data indices for the control point data are all assigned to the face and can be trivially identified virtual int getFaceVertexIndices Index faceIndex , Index vertexIndices const 0 virtual int getFaceFVarValueIndices Index faceIndex , FVarID faceVaryingID , Index faceVaryingIndices const 0 Since multiple sets of face-varying data with different topology may be assigned to the mesh, an identifier needs to be specified both in the public interface when requesting a Surface and here when the factory assembles it. How a face-varying identifier is interpreted is completely determined by the subclass through the implementation of the methods that require it. Specifying the Neighborhood Around a Vertex When the Surface requested is not linear, the entire neighborhood around the face must be determined. This is achieved by specifying the neighborhoods around each of the vertices of the face, which the factory then assembles. For the neighborhood of each face-vertex, the factory obtains a complete specification in a simple VertexDescriptor class. An instance of VertexDescriptor is provided and populated with the following method virtual int populateFaceVertexDescriptor Index faceIndex , int faceVertex , VertexDescriptor vertexDescriptor const 0 Within this method, the given VertexDescriptor instance is initialized using a small suite of VertexDescriptor methods that specify the following information about the vertex and its neighborhood whether the neighborhood is manifold ordered counter-clockwise whether the vertex is on a boundary the sizes of all or each incident face the sharpness of the vertex the sharpness of edges of incident faces These methods are specified between Initialize and Finalize methods, so an interior vertex of valence 4 with three incident quads and one incident triangle might be specified as follows int vertexValence 4 vertexDescriptor . Initialize vertexValence vertexDescriptor . SetManifold true vertexDescriptor . SetBoundary false vertexDescriptor . SetIncidentFaceSize 0 , 4 vertexDescriptor . SetIncidentFaceSize 1 , 4 vertexDescriptor . SetIncidentFaceSize 2 , 3 vertexDescriptor . SetIncidentFaceSize 3 , 4 vertexDescriptor . Finalize Specifying the vertex neighborhood as manifold is critical to allowing the factory to inspect the neighborhood efficiently. A manifold vertex has its incident faces and edges ordered in a counter-clockwise orientation and is free of degeneracies. If it is not clear that a vertex is manifold, it should not be set as such or the factorys inspection of associated data will not be correct. Identifying Indices Around a Vertex When the Surface requested is not linear, the indices of control point data for the entire neighborhood of the face are ultimately required, and that entire set is similarly determined by identifying the indices for each of the neighborhoods of the face-vertices virtual int getFaceVertexIncidentFaceVertexIndices Index faceIndex , int faceVertex , Index vertexIndices const 0 virtual int getFaceVertexIncidentFaceFVarValueIndices Index faceIndex , int faceVertex , FVarID faceVaryingID , Index faceVaryingIndices const 0 As was the case with the methods retrieving indices for the entire face, one exists for identifying indices vertex data while another exists to identify indices for a specified set of face-varying data. Customizing the Subclass Interface Once the topological requirements of a subclass have been satisfied for its mesh representation, minor customizations of the inherited interface of SurfaceFactory may be useful. Consider a class called Mesh and its associated subclass of SurfaceFactory called MeshSurfaceFactory. At minimum, a constructor of MeshSurfaceFactory is necessary to construct an instance for a particular instance of mesh. This is typically achieved as follows MeshSurfaceFactory Mesh const mesh , Options const options In addition to the Mesh instance, such a constructor passes a set of Options i.e. SurfaceFactoryOptions to the base SurfaceFactory. Any additional arguments are possible here, e.g. perhaps only a single face-varying UV set is supported, and that might be specified by identifying it on construction. Given that mesh representations often have their own associated classes that internally contain the actual data, it may be useful to provide a few other conveniences to simplify working with a Mesh. For example, if mesh data is stored in a class called MeshPrimvar, a method to construct a Surface from a given MeshPrimvar may be useful, e.g. bool InitPrimvarSurface int faceIndex , MeshPrimvar const meshPrimvar , Surface float surface which would then determine the nature of the MeshPrimvar data interpolated as vertex, varying or face-varying and act accordingly. It may also be worth simplifying the template complexity here if only one precision is ever required. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"tags\": \"\",\n+ \"loc\": \"bfr_overview.html\"\n+ }, {\n \"title\": \"API Overview\",\n- \"text\": \"API Overview API Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE API Layers Using the Right Tools Use case 1 Simple refinement Use case 2 GL adaptive tessellation drawing of animating mesh Tutorials and Examples API Layers OpenSubdiv is structured as a set of layered libraries. This structure facilitates operation on a variety of computing resources, and allows developers to only opt-in to the layers and feature sets that they require. From a top-down point of view, OpenSubdiv is comprised of several layers, some public, and some private. Layers list Sdc Subdivision Core Sdc Overview The lowest level layer, implements the core subdivision details to facilitate the generation of consistent results. Most cases will only require the use of simple public types and constants from Sdc. Vtr Vectorized Topological Representation Vtr Overview A suite of classes to provide an intermediate representation of topology that supports efficient refinement. Vtr is intended for internal use only. Far Feature Adaptive Representation Far Overview The central interface that processes client-supplied geometry and turns it into a serialized data representation ready for parallel processing in Osd . Far also provides a fully-featured single-threaded implementation of subdivision interpolation algorithms. Bfr Bace Face Representation Bfr Overview A suite of classes to provide parameterization, evaluation and tessellation on the CPU. Bfr is more flexible and more scalable than Osd but potentially less efficient. Osd OpenSubdiv cross platform Osd Overview A suite of classes to provide parallel subdivision kernels and drawing utilities on a variety of platforms such as TBB, CUDA, OpenCL, GLSL and DirectX. Client mesh data enters the API through the Far layer. Typically, results will be collected from the Osd layer. However, it is possible to use functionality from Far without introducing any dependency on Osd. Although there are several entry-points to provide topology and primitive variable data to OpenSubdiv, eventually everything must pass through the private Vtr and Sdc representations for topological analysis. Using the Right Tools OpenSubdivs tiered interface offers a lot flexibility to make your application both fast and robust. Because navigating through the large collection of classes and features can be challenging, here are use cases that should help sketch the broad lines of going about using subdivisions in your application. General client application requirements Surface Limit For some applications, a polygonal approximation of the smooth surface is enough. Others require C 2 continuous differentiable bi-cubic patches ex deformable displacement mapping, smooth normals and semi-sharp creases... Deforming Surface Applications such as off-line image renderers often process a single frame at a time. Others, such as interactive games need to evaluate deforming character surface every frame. Because we can amortize many computations if the topology of the mesh does not change, OpenSubdiv provides stencil tables in order to leverage subdivision refinement into a pre-computation step. Multi-threading OpenSubdiv also provides dedicated interfaces to leverage parallelism on a wide variety of platforms and API standards, including both CPUs and GPUs. GPU Draw If the application requires interactive drawing on screen, OpenSubdiv provides several back-end implementations, including D3D11 and OpenGL. These back-ends provide full support for programmable shading. Use case 1 Simple refinement The following example shows the most simple case to get your mesh refined uniformly. Define a class for the primvar you want to refine. Its required to have Clear and AddWithWeight functions. struct Vertex void Clear x y z 0 void AddWithWeight Vertex const src , float weight x weight src . x y weight src . y z weight src . z float x , y , z 2. Instantiate a FarTopologyRefiner from the FarTopologyDescriptor . Far TopologyDescriptor desc desc . numVertices the number of vertices desc . numFaces the number of faces desc . numVertsPerFace array of the number of verts per face desc . vertIndicesPerFace array of vert indices Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc Call RefineUniform to refine the topology up to maxlevel. refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel 4. Interpolate vertex primvar data at level using FarPrimvarRefiner Far PrimvarRefiner primvarRefiner refiner Vertex const src coarse vertices Vertex dst refined vertices primvarRefiner . Interpolate level , src , dst The topology at the refined level can be obtained from FarTopologyLevel Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int nverts refLastLevel . GetNumVertices int nfaces refLastLevel . GetNumFaces for int face 0 face nfaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face do something with dst and fverts Done See Far tutorial 1.1 for the complete code example. Use case 2 GL adaptive tessellation drawing of animating mesh The next example is showing how to draw adaptive tessellated patches in GL using OpenSubdiv. The osd layer helps you to interact with GL and other device specific APIs. Also for an efficient refinement of animating mesh on a static topology, we create a stencil table to refine the positions changing over time. The following example code uses an OsdGLMesh utility class which composites a stencil table, patch table, vertex buffer and evaluator in osd layer. You can also use those classes independently. 1. Instantiate a FarTopologyRefiner from the FarTopologyDescriptor , same as usecase 1. Setup OsdMesh. In this example we use b-spline endcap. int numVertexElements 3 x, y, z Osd MeshBitset bits bits . set Osd MeshAdaptive , true set adaptive bits . set Osd MeshEndCapBSplineBasis , true use b-spline basis patch for endcap. Osd GLMeshInterface mesh new Osd Mesh Osd CpuGLVertexBuffer , Far StencilTable , Osd CpuEvaluator , Osd GLPatchTable refiner , numVertexElements , 0 , level , bits 3. Update coarse vertices and refine OsdMeshRefine calls OsdCpuEvaluatorEvalStencils mesh - UpdateVertexBuffer vertex 0 , 0 , nverts mesh - Refine Bind index buffer, PatchParamBuffer and vertex buffer index buffer glBindBuffer GL_ELEMENT_ARRAY_BUFFER , mesh - GetPatchTable - GetPatchIndexBuffer vertex buffer glBindBuffer GL_ARRAY_BUFFER , mesh - BindVertexBuffer glEnableVertexAttribArray 0 glVertexAttribPointer 0 , numVertexElements , GL_FLOAT , GL_FALSE , numVertexElements sizeof float , 0 patch param buffer glActiveTexture GL_TEXTURE0 glBindTexture GL_TEXTURE_BUFFER , mesh - GetPatchTable - GetPatchParamTextureBuffer Draw. Since we use b-spline endcaps in this example, there is only one PatchArray in the patch table. You may need to iterate patch arrays as you use other type of endcap. To configure GLSL program for each patch type, see osd shader interface for more details. Osd PatchArray const patch mesh - GetPatchTable - GetPatchArrays 0 Far PatchDescriptor desc patch . GetDescriptor int numVertsPerPatch desc . GetNumControlVertices 16 for B-spline patches glUseProgram BSplinePatchProgram glPatchParameteri GL_PATCH_VERTICES , numVertsPerPatch glDrawElements GL_PATCHES , patch . GetNumPatches numVertsPerPatch , GL_UNSIGNED_INT , 0 As the mesh animates, repeat from step 3 to update positions, refine, and draw. See glViewer and other examples for more complete usage. Tutorials and Examples For more use cases, please see Tutorials and Examples Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"API Overview API Overview 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE API Layers Using the Right Tools Use case 1 Simple refinement Use case 2 GL adaptive tessellation drawing of animating mesh Tutorials and Examples API Layers OpenSubdiv is structured as a set of layered libraries. This structure facilitates operation on a variety of computing resources, and allows developers to only opt-in to the layers and feature sets that they require. From a top-down point of view, OpenSubdiv is comprised of several layers, some public, and some private. Layers list Sdc Subdivision Core Sdc Overview The lowest level layer, implements the core subdivision details to facilitate the generation of consistent results. Most cases will only require the use of simple public types and constants from Sdc. Vtr Vectorized Topological Representation Vtr Overview A suite of classes to provide an intermediate representation of topology that supports efficient refinement. Vtr is intended for internal use only. Far Feature Adaptive Representation Far Overview The central interface that processes client-supplied geometry and turns it into a serialized data representation ready for parallel processing in Osd . Far also provides a fully-featured single-threaded implementation of subdivision interpolation algorithms. Bfr Bace Face Representation Bfr Overview A suite of classes to provide parameterization, evaluation and tessellation on the CPU. Bfr is more flexible and more scalable than Osd but potentially less efficient. Osd OpenSubdiv cross platform Osd Overview A suite of classes to provide parallel subdivision kernels and drawing utilities on a variety of platforms such as TBB, CUDA, OpenCL, GLSL and DirectX. Client mesh data enters the API through the Far layer. Typically, results will be collected from the Osd layer. However, it is possible to use functionality from Far without introducing any dependency on Osd. Although there are several entry-points to provide topology and primitive variable data to OpenSubdiv, eventually everything must pass through the private Vtr and Sdc representations for topological analysis. Using the Right Tools OpenSubdivs tiered interface offers a lot flexibility to make your application both fast and robust. Because navigating through the large collection of classes and features can be challenging, here are use cases that should help sketch the broad lines of going about using subdivisions in your application. General client application requirements Surface Limit For some applications, a polygonal approximation of the smooth surface is enough. Others require C 2 continuous differentiable bi-cubic patches ex deformable displacement mapping, smooth normals and semi-sharp creases... Deforming Surface Applications such as off-line image renderers often process a single frame at a time. Others, such as interactive games need to evaluate deforming character surface every frame. Because we can amortize many computations if the topology of the mesh does not change, OpenSubdiv provides stencil tables in order to leverage subdivision refinement into a pre-computation step. Multi-threading OpenSubdiv also provides dedicated interfaces to leverage parallelism on a wide variety of platforms and API standards, including both CPUs and GPUs. GPU Draw If the application requires interactive drawing on screen, OpenSubdiv provides several back-end implementations, including D3D11 and OpenGL. These back-ends provide full support for programmable shading. Use case 1 Simple refinement The following example shows the most simple case to get your mesh refined uniformly. Define a class for the primvar you want to refine. Its required to have Clear and AddWithWeight functions. struct Vertex void Clear x y z 0 void AddWithWeight Vertex const src , float weight x weight src . x y weight src . y z weight src . z float x , y , z 2. Instantiate a FarTopologyRefiner from the FarTopologyDescriptor . Far TopologyDescriptor desc desc . numVertices the number of vertices desc . numFaces the number of faces desc . numVertsPerFace array of the number of verts per face desc . vertIndicesPerFace array of vert indices Far TopologyRefiner refiner Far TopologyRefinerFactory Descriptor Create desc Call RefineUniform to refine the topology up to maxlevel. refiner - RefineUniform Far TopologyRefiner UniformOptions maxlevel 4. Interpolate vertex primvar data at level using FarPrimvarRefiner Far PrimvarRefiner primvarRefiner refiner Vertex const src coarse vertices Vertex dst refined vertices primvarRefiner . Interpolate level , src , dst The topology at the refined level can be obtained from FarTopologyLevel Far TopologyLevel const refLastLevel refiner - GetLevel maxlevel int nverts refLastLevel . GetNumVertices int nfaces refLastLevel . GetNumFaces for int face 0 face nfaces face Far ConstIndexArray fverts refLastLevel . GetFaceVertices face do something with dst and fverts Done See Far tutorial 1.1 for the complete code example. Use case 2 GL adaptive tessellation drawing of animating mesh The next example is showing how to draw adaptive tessellated patches in GL using OpenSubdiv. The osd layer helps you to interact with GL and other device specific APIs. Also for an efficient refinement of animating mesh on a static topology, we create a stencil table to refine the positions changing over time. The following example code uses an OsdGLMesh utility class which composites a stencil table, patch table, vertex buffer and evaluator in osd layer. You can also use those classes independently. 1. Instantiate a FarTopologyRefiner from the FarTopologyDescriptor , same as usecase 1. Setup OsdMesh. In this example we use b-spline endcap. int numVertexElements 3 x, y, z Osd MeshBitset bits bits . set Osd MeshAdaptive , true set adaptive bits . set Osd MeshEndCapBSplineBasis , true use b-spline basis patch for endcap. Osd GLMeshInterface mesh new Osd Mesh Osd CpuGLVertexBuffer , Far StencilTable , Osd CpuEvaluator , Osd GLPatchTable refiner , numVertexElements , 0 , level , bits 3. Update coarse vertices and refine OsdMeshRefine calls OsdCpuEvaluatorEvalStencils mesh - UpdateVertexBuffer vertex 0 , 0 , nverts mesh - Refine Bind index buffer, PatchParamBuffer and vertex buffer index buffer glBindBuffer GL_ELEMENT_ARRAY_BUFFER , mesh - GetPatchTable - GetPatchIndexBuffer vertex buffer glBindBuffer GL_ARRAY_BUFFER , mesh - BindVertexBuffer glEnableVertexAttribArray 0 glVertexAttribPointer 0 , numVertexElements , GL_FLOAT , GL_FALSE , numVertexElements sizeof float , 0 patch param buffer glActiveTexture GL_TEXTURE0 glBindTexture GL_TEXTURE_BUFFER , mesh - GetPatchTable - GetPatchParamTextureBuffer Draw. Since we use b-spline endcaps in this example, there is only one PatchArray in the patch table. You may need to iterate patch arrays as you use other type of endcap. To configure GLSL program for each patch type, see osd shader interface for more details. Osd PatchArray const patch mesh - GetPatchTable - GetPatchArrays 0 Far PatchDescriptor desc patch . GetDescriptor int numVertsPerPatch desc . GetNumControlVertices 16 for B-spline patches glUseProgram BSplinePatchProgram glPatchParameteri GL_PATCH_VERTICES , numVertsPerPatch glDrawElements GL_PATCHES , patch . GetNumPatches numVertsPerPatch , GL_UNSIGNED_INT , 0 As the mesh animates, repeat from step 3 to update positions, refine, and draw. See glViewer and other examples for more complete usage. Tutorials and Examples For more use cases, please see Tutorials and Examples Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"api_overview.html\"\n }, {\n \"title\": \"Additional Resources\",\n- \"text\": \"Additional Resources Additional Resources 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Links Videos Feature Adaptive GPU Rendering of Catmull-Clark Subdivision Surfaces 2012 Open Subdivision Technology Review 2012 Autodesk User Group At Anaheim 2013 Why Model with Subdivisions 2013 Meet the Experts The OpenSubdiv Project 2013 Links OpenSubdiv Github Repository OpenSubdiv with Mudbox and Maya Videos Feature Adaptive GPU Rendering of Catmull-Clark Subdivision Surfaces 2012 Open Subdivision Technology Review 2012 Autodesk User Group At Anaheim 2013 Why Model with Subdivisions 2013 Meet the Experts The OpenSubdiv Project 2013 Join special guests, Bill Polson, Dirk Van Gelder, Manuel Kraemer, Takahito Tejima, David G. Yu and Dale Ruffolo, from Pixar Animation Studios GPU team, as they show how real time display of subdivision surfaces helps artists be more productive, and how this code is open source and engineered for ease of integration. Generated on 2024-01-12 1326 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n+ \"text\": \"Additional Resources Additional Resources 3.5.0 User Docs API Docs Release Notes Forum Github OSD_ARTICLE_TITLE Links Videos Feature Adaptive GPU Rendering of Catmull-Clark Subdivision Surfaces 2012 Open Subdivision Technology Review 2012 Autodesk User Group At Anaheim 2013 Why Model with Subdivisions 2013 Meet the Experts The OpenSubdiv Project 2013 Links OpenSubdiv Github Repository OpenSubdiv with Mudbox and Maya Videos Feature Adaptive GPU Rendering of Catmull-Clark Subdivision Surfaces 2012 Open Subdivision Technology Review 2012 Autodesk User Group At Anaheim 2013 Why Model with Subdivisions 2013 Meet the Experts The OpenSubdiv Project 2013 Join special guests, Bill Polson, Dirk Van Gelder, Manuel Kraemer, Takahito Tejima, David G. Yu and Dale Ruffolo, from Pixar Animation Studios GPU team, as they show how real time display of subdivision surfaces helps artists be more productive, and how this code is open source and engineered for ease of integration. Generated on 2025-02-13 2004 UTC. User Docs API Docs Release Notes Forum TM 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms of Use\",\n \"tags\": \"\",\n \"loc\": \"additional_resources.html\"\n }, ]\n };\n"}]}, {"source1": "./usr/share/doc/opensubdiv/tutorials.html", "source2": "./usr/share/doc/opensubdiv/tutorials.html", "unified_diff": "@@ -531,15 +531,15 @@\n OBJ format for inspection. [code]\n \"images/hbr_tutorial_2.0.png\"\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -280,14 +280,14 @@\n the same pyramid shape used in hbr_tutorial_0. We then apply the\n Refine() function sequentially to all the faces in the mesh to\n generate several levels of uniform subdivision. The resulting data is\n then dumped to the terminal in Wavefront OBJ format for inspection.\n [code]\n [images/hbr_tutorial_2.0.png]\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/using_osd_hbr.html", "source2": "./usr/share/doc/opensubdiv/using_osd_hbr.html", "unified_diff": "@@ -756,15 +756,15 @@\n vertexBuffer->UpdateData(vtxData+splits[i].second*numVertexElements, splits[i].first, 1);\n }\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -522,14 +522,14 @@\n // Duplicate the vertex data into the split singular vertices\n std::vector > const splits = hbrMesh->GetSplitVertices();\n for (int i=0; i<(int)splits.size(); ++i) {\n vertexBuffer->UpdateData(vtxData+splits[i].second*numVertexElements, splits\n [i].first, 1);\n }\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}, {"source1": "./usr/share/doc/opensubdiv/vtr_overview.html", "source2": "./usr/share/doc/opensubdiv/vtr_overview.html", "unified_diff": "@@ -392,15 +392,15 @@\n Tags were added to the base level that are propagated through the refinement\n and these too are dependent on the scheme, but are applied externally.

\n \n \n \n
\n
\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n
\n \n \n \n \n \n", "details": [{"source1": "html2text {}", "source2": "html2text {}", "unified_diff": "@@ -238,14 +238,14 @@\n savings are considerable.\n Currently there is nothing specific to a subdivision scheme in the refinement\n other than the type of topological splitting to apply. The refinement does\n subdivide sharpness values for creasing, but that too is independent of scheme.\n Tags were added to the base level that are propagated through the refinement\n and these too are dependent on the scheme, but are applied externally.\n ===============================================================================\n-Generated on: 2024-01-12 13:26 UTC.\n+Generated on: 2025-02-13 20:04 UTC.\n \n * User_Docs\n * API_Docs\n * Release_Notes\n * Forum\n TM & \u00a9 1986 - 2017 PIXAR. ALL RIGHTS RESERVED. Terms_of_Use\n"}]}]}]}]}]}